Culture and Negotiation Strategy

The increasing number of business transactions occurring across national borders is reflected in the general acceptance of culture as a contextual variable in management research. Underlying cultural dimensions (such as values for what is important or norms for what is appropriate) have frequently been explored to explain cultural differences in managerial behavior and decision making (Lytle et al. 1995). In the area of conflict management, individualism/collectivism, egalitarianism/hierarchy, and saving face are just some of the values and norms that researchers have relied on to explain differences in negotiator behavior and negotiation outcome (see Leung 1998 for a review).

In the quest to understand culture's influence on negotiation, Brett and colleagues (1998) measured cultural values, norms, and joint gains in a transactional negotiation involving people from six national cultures: France, Russia, Japan, Hong Kong, Brazil, and the United States. We failed to find a direct relationship between the cultural values or norms measured and joint gains in negotiation. In other words, negotiators from cultures where hierarchical status (Schwartz 1994) is valued can negotiate joint gains to the same degree as negotiators from cultures where egalitarian status is valued (Brett et al. 1998). Negotiators from both types of cultures can also negotiate poor to moderate joint gains (Brett et al. 1998).

For example, as reported by Brett et al. (1998), U.S. and Japanese negotiators (who achieved higher joint gains than negotiators from Brazil, France, Russia, or Hong Kong) strongly agreed about the appropriateness of information sharing in negotiation. However, they had different cultural values for hierarchy (the Japanese were more hierarchical and the U.S. more egalitarian) and different norms regarding how appropriate it is to use distributive tactics in negotiation. Japanese and Russian negotiators, who had extremely different levels of joint gains (Japanese achieved the highest joint gains and Russians achieved the lowest joint gains), had similar hierarchical values and norms for the use of distributive tactics, but diverged with respect to the appropriateness of information sharing. These examples show that there was no single combination of cultural values and norms that led to the negotiation of joint gains.

The research reported in this article attempts to inform the findings of Brett et al. (1998) by examining the relationship between culture, negotiation strategy, and joint gains in the same six culture groups. First, we explore how culture may explain differences in communication and power strategies. Then, we build upon our value and norm data to develop hypotheses about how negotiators in different cultures use information sharing and power strategies. Hypotheses are evaluated using discourse analysis on content-coded transcripts of the negotiations previously studied.

Our research focuses on the relationship between culture and negotiation strategy to explain the outcome data reported by Brett et al. (1998). Because we found differences in cultural values and norms for information sharing and distributive tactics, we expect the six cultures also to differ in how they share information and use distributive tactics. To explore these differences, we developed theories on how communication and power strategies may differ across cultures.

The literature on cross-cultural communication suggests that different cultures may have different information-sharing strategies in negotiation (Hall 1976). Brett et al. (1998) found that U.S. and Japanese negotiators had similar norms for information sharing in negotiation, but subsequent research found that U.S. negotiators embraced direct information-sharing strategies while Japanese negotiators used indirect information-sharing strategies (Adair, Okumura, and Brett 2001). Other accounts of U.S. and Japanese negotiation styles also report different communication styles; for example, Japanese negotiators are less likely to say “no” and more likely to remain silent than U.S. negotiators (Graham and Sano 1989; March 1990). The various norms and strategies for information sharing seen in the U.S. and Japan suggest that culture may account for different communication strategies in the six cultures under investigation.

Low/high context is a cultural dimension that refers to the amount of information contained in an explicit message versus implicit contextual cues (Hall 1976). In low-context cultures, like the U.S., messages are transmitted explicitly and directly, and communication is action-oriented and solution-minded (Ting-Toomey 1985). In high-context cultures, like Japan, messages are transmitted indirectly and implicitly, and communication is allusive (Ting-Toomey 1985).

The low/high context dimension has direct implications for perceptions of how much the other party should rely on contextual cues and how much explicit information is enough (Hall and Hall 1990; Cohen 1991). Consequently, negotiators from low-context cultures should rely primarily on direct communication strategies, and negotiators from high-context cultures should rely primarily on indirect communication strategies.

We expect that both direct and indirect communication strategies can lead to high joint gains through different communication behaviors. To generate joint gains or enlarge the pie, negotiators must share enough information to understand each other's priorities and identify trade-offs and compatible issues. It is possible for negotiators to exchange sufficient information using either direct or indirect information-sharing strategies. For example, if a person says, “I prefer a higher salary to a higher signing bonus,” he/she is directly conveying his/her priorities to a potential employer. The same information could be conveyed indirectly through a series of offers by saying, “I would consider $60,000 and a $5,000 signing bonus,” and later offering “$63,000 and no signing bonus.”

Direct information exchange conveys explicit information on priorities. We expect that low-context cultures will use direct information-sharing strategies as measured by Adair, Okumura, and Brett (2001), such as exchanging information about preferences and priorities, making explicit comparisons and contrasts between the parties, and reacting explicitly to offers and proposals.

Indirect communication requires that negotiators infer the other party's priorities from contextual cues such as offers generated or entertained by the other party. We expect that high-context cultures will use indirect information-sharing strategies as measured by Adair, Okumura, and Brett (2001), such as offers and counteroffers. Offers provide information indirectly because one party must infer the other party's priorities from a series of offers made over time. Indirect communication through offers is similar to what Pruitt (1981) describes as information exchange through heuristic trial-and-error processes.

Power in negotiation is the basis of one party's ability to gain advantage over the other party (Lewicki, Saunders, and Minton 1997) through positional, distributive, or influence tactics (French and Raven 1959; Pruitt 1981). Egalitarianism versus hierarchy is a cultural value with implications for the use of power strategies in negotiation. Hierarchical cultural values stress the importance of status and power whereas egalitarian cultural values stress sameness, equal opportunity, and achievement (Hofstede 1980; Schwartz 1994).

Brett et al. (1998) found that hierarchical cultures in comparison to egalitarian cultures were more likely to espouse norms for distributive tactics. Distributive tactics (i.e. making threats or using arguments) are power strategies that are focused on individual, not joint, gains (Pruitt 1981 and 1983). Distributive tactics are normative in hierarchical cultures because negotiators use positional and persuasive arguments to make status and power differences clear.

For example, in hierarchical Japan, parties are likely to use power strategies to establish their position of power and claim their portion of the pie (Graham and Sano 1989; March 1990). Likewise in hierarchical Russia, negotiators positioning for power may seem quite competitive and pushy to an outsider (Roemer et al. 1997). Because distributive tactics are normative in these cultures, using them to establish power differentials may even serve to motivate negotiators to push harder for a better deal (March 1990). Therefore, we expect that power strategies can lead to high joint gains in hierarchical cultures that have norms for distributive tactics.

We think it likely that power strategies could jeopardize joint gains in egalitarian cultures where distributive tactics are not normative. In these cultures, negotiators may prefer to leave power differentials murky so that negotiation outcome is purely a function of what BATNA (Best Alternative to a Negotiated Agreement [Fisher, Ury, and Patton 1991]) each negotiator brings to the table (Adair, Okumura, and Brett 2001). If power strategies are not normative or expected, negotiators may interpret their positional flavor as hard bargaining or even contentious behavior. They may respond with hard bargaining or contentious behaviors of their own, leading to a conflict spiral, low joint gains, or impasse (Brett, Shapiro, and Lytle 1998; Pruitt 1981; Pruitt and Lewis 1975).

We expect that, in general, negotiators from hierarchical cultures will use power strategies more than negotiators from egalitarian cultures. We consider three types of power strategies as in the study by Adair, Okumura, and Brett (2001). Legitimacy makes reference to one's company, status, or competitors to persuade. In hierarchical cultures, company status and negotiator role are sources of legitimate power that have implications for negotiator behavior (Brett et al. 1998; Leung 1997). For example, in Japan, it is normative for a low-status company to defer to a high-status company and a seller to defer to a buyer (Graham, Johnston, and Kamins 1997). Informational persuasion, such as positive and negative substantiation, is a fact-oriented form of argument. Informational persuasion with sanctions adds the element of a threat, for example, a reference to BATNA or reservation price that conveys a party's alternatives or power to walk away from the table.

Different communication and power strategies in the six cultures studied previously (Brett et al. 1998) may help account for between-culture differences in joint gains. In this section, we reference empirical and descriptive data to profile expected communication and power strategies in the six cultures. We expect the cultures to vary in their relative use of direct versus indirect information strategies based on whether the culture is characterized as low- or high-context. We also expect the cultures to vary in their relative use of power strategies based on underlying power assumptions as measured by values for egalitarianism/hierarchy, norms for the appropriateness of distributive tactics, and definitions of power in negotiation.

As noted by Brett et al. (1998), France is an enigma. Descriptive accounts suggest it is both a high- and low-context culture. French diplomacy, which dominated eighteenth and nineteenth century European politics, certainly was high-context — nuanced, secretive, and indirect. Evidence of this subtle communication style (Hall and Hall 1990) along with an inductive problem-solving approach and reliance on context (Schneider and Barsoux 1997) suggest that French negotiators will use relatively more indirect communication. Yet, today, French institutions (e.g. education) emphasize an analytical approach to problem solving and attention to facts (Morrison, Conaway, and Borden 1994), suggesting that negotiators may also use direct communication.

The French culture is both egalitarian and hierarchical. France has long had a powerful socialist, working class left. But at the same time, the right wing, with its origins in the monarchy, remains a powerful force in France. Brett et al. (1998) found that French negotiators were unlikely to endorse hierarchy as a cultural value or distributive tactics as normative in negotiation. In addition, French negotiators expected information and BATNA, rather than company status or role, to determine power in negotiation. This empirical evidence suggests that French negotiators will avoid power strategies in negotiation. However, these results contradict accounts that France is a hierarchical culture (Hall and Hall 1990) where status, rank, and power play an important role (Morrison, Conaway, and Borden 1994; Schneider and Barsoux 1997). Such descriptive accounts suggest that French negotiators will use more power strategies in negotiation.

Most descriptive accounts place Russia as a high-context culture (see Morrison, Conaway, and Borden [1994] for an exception). Russian communication is indirect and holistic (Berdiaev 1990), even secretive (Rajan and Graham 1991). Russians process information subjectively and associatively, characteristics that are also typical of high-context communication norms (Morrison, Conaway, and Borden 1994). These accounts suggest that Russian negotiators will use indirect communication strategies.

In descriptive accounts, power and hierarchy frame Russian social norms (Lefebvre and Lefebvre 1986; Smith 1991), and norms for distributive tactics govern interactions with an out-group (Roemer et al. 1997). Brett et al. (1998) found that Russian negotiators identified hierarchy as a guiding cultural value, distributive tactics as normative in negotiation, and negotiator role as a source of power in negotiation. The consistent descriptive and empirical evidence suggests that Russian negotiators are likely to use more power strategies in negotiation.

Japan has been characterized as a high-context culture from analyses of its communication style, legal system, and literature (Hall 1976). Hall (1976) does describe a low-context side of the Japanese culture that may be evident in some ceremonial business negotiations, but in general the conversational style at home and at work is subtle, indirect, and implicit. Therefore, we expect that Japanese negotiators are likely to use indirect communication.

Prior research suggests that Japan is a hierarchical culture (Hofstede 1980; Brett and Okumura 1998; Schwartz 1994; Tinsley 1998) where role and status are normative sources of power (Brett et al. 1998; Leung 1997; Brett and Okumura 1998). In Japan, distributive tactics are normative in negotiation (Brett et al. 1998; Graham and Sano 1989; Brett and Okumura 1998). This consistent evidence leads us to expect that Japanese negotiators will use power strategies in negotiation.

The Hong Kong Chinese culture is described as high-context (Ting-Toomey 1985), where associative information processing is the norm (Morrison, Conaway, and Borden 1994). As an example of indirect communication styles, Morrison, Conaway, and Borden (1994) describe that “yes” and “no” may have subtle meanings and are not direct responses in negotiations with Hong Kong Chinese. These accounts suggest that Hong Kong Chinese negotiators will use indirect communication strategies.

Descriptive accounts emphasize the role of power in Hong Kong Chinese society; for example, legitimate power determines behavioral norms (Morrison, Conaway, and Borden 1994) and strong social controls (Schneider and Barsoux 1997). Likewise, empirical data indicate that Hong Kong Chinese negotiators embrace hierarchy as a cultural value and are concerned with authority and power in negotiation (Brett et al. 1998; Tinsley and Brett 2001). These accounts suggest that Hong Kong Chinese negotiators will use power strategies in negotiation.

Descriptive accounts suggest that Brazilians approach problems indirectly, which is indicative of a high-context culture (Morrison, Conaway, and Borden 1994). Based on the use of nonverbal cues in communication, including touching, tone of voice, and relationships, Oliveira (2001) also categorizes Brazil as a high-context culture using Hall's (1976) criteria. These accounts suggest that Brazilian negotiators will use indirect communication strategies.

According to descriptive accounts, hierarchy, class, and status play important roles in Brazilian society (Morrison, Conaway, and Borden 1994; Schneider and Barsoux 1997). Brett et al. (1998) report empirical evidence of both low- and high-power norms in Brazil. The Brazilian participants endorsed norms for distributive tactics in negotiation, suggesting that Brazilian negotiators will use power strategies in negotiation. However, they also were more egalitarian than hierarchical and viewed both information and alternatives (BATNA) as power in negotiation, suggesting that Brazilian negotiators will not use power strategies in negotiation.

The U.S. has been classified as a low-context culture where communication is direct (Hall 1976) and negotiations are analytical and fact-oriented (Morrison, Conaway, and Borden 1994). The deductive problem-solving approach common in the U.S. means that negotiators are more comfortable with searching and experimentation than relying on precedents such as history and context (Schneider and Barsoux 1997). Based on this evidence, we expect that U.S. negotiators are likely to use direct communication strategies.

The U.S. has been classified as an egalitarian culture (Brett et al. 1998; Hofstede 1980; Morrison, Conaway, and Borden 1994; Schwartz 1994) where negotiators do not consider distributive tactics appropriate in negotiation (Brett et al. 1998; Brett and Okumura 1998). Brett et al. (1998) found that U.S. negotiators viewed BATNA as power in negotiation. Relative to other cultures, the U.S. approach has been characterized as interests-based and problem solving (Graham, Mintu, and Rogers 1994; Tinsley 1998). Therefore, we expect that U.S. negotiators will be less likely to use power strategies in negotiation.

This research tests how cultural differences in communication and power norms affect negotiator strategies. The U.S. has the most evidence for being a low-context culture characterized by direct communication. Russia, Japan, Brazil, and Hong Kong are generally classified as high-context, and France, as we noted earlier, has been classified as both high- and low-context, depending on the situation. Because the U.S. culture is likely to be the most low-context of those sampled, we hypothesize:

Because Russia and France are characterized by an attention to factual information that may affect the exchange of information on preferences and priorities, we hypothesize:

Because Russia, Japan, and Hong Kong have the most evidence of a high-context culture characterized by indirect communication, we hypothesize:

Russia and Japan have the most evidence of a power-focused culture, as they were high on three different indicators of power in the Brett et al. (1998) study: hierarchical values, espousing distributive tactics, and seeing role information as power. Brazil espoused distributive tactics and saw role information as power, but held more egalitarian values. Hong Kong, while holding hierarchical values and seeing role information as power, nonetheless eschewed the use of distributive tactics. Thus, while both Brazil and Hong Kong have some power focus, Russia and Japan have a stronger power focus.

Because Brazil and Hong Kong have more evidence of a power-focused culture than France or the U.S., we hypothesize:

Negotiators from the six different cultures completed a common negotiation task called “Cartoon,” involving the sale of syndicated rights to a children's television cartoon. The simulation, adapted from Tenbrunsel and Bazerman (1995), stipulates a fixed five-year, one-hundred-episode contract for the cartoon series Ultra Rangers. The negotiable issues include one distributive issue — price of each episode, and two issues that negotiators can logroll — financing and runs (the number of times each episode may be shown in the five-year period).

In addition, both parties are aware of the future availability of a second cartoon, Strums, a compatible issue that can create value for both parties. Finally, there is an opportunity for a contingent contract if the parties realize the potential to bet on their different rating expectations. Table One shows each party's positions on the issues as well as the value of their alternative deals. Negotiators’ confidential information contained the information for their role from the appropriate column in Table One.

The Japanese, Russian, and Brazilian materials were translated from the U.S. version with only minor changes in context (e.g. a Chicago independent TV station versus a Tokyo independent TV station). Bilingual speakers translated and back-translated all role instructions and simulation materials for non-English materials.

In the Japanese version, all numbers in Table One were converted into yen using the current exchange rate at the time of data collection. Because it is common for Russians to conduct business transactions with rubles converted to the more stable U.S. dollar rate, the Russian participants negotiated the sale in U.S. dollars. The Brazilian participants were attending a negotiation-training course in the U.S., so they also negotiated the case in U.S. dollars. Because the business classes in France and Hong Kong were taught in English, and students normally received their materials in English, these roles were not translated. Participants from all six cultures conducted negotiations in their native language.

Participants in the Brazilian and Japanese data sets were executives participating in a negotiation-training program in the United States. Participants in the French, Russian, and U.S. data sets were MBA students who negotiated Cartoon as the first exercise in a course or workshop on negotiation. Participants in the Hong Kong Chinese data set were undergraduate students who negotiated Cartoon as the first negotiation exercise in a course.

Participants in the Brazilian and one of the classes in the U.S. data set had an average age of 38 years; participants in the Japanese, Russian, French and one class of the U.S. data set averaged 27 years of age; participants in the Hong Kong Chinese data set were younger, averaging 23 years. There were no significant differences in cultural values or norms between the two U.S. data sets, so these samples were combined for all data analyses.

The study draws on transcripts for 139 same-culture dyads: 9 French, 45 Russian, 22 Japanese, 16 Hong Kong, 7 Brazilian, and 40 U.S.

After a standard introduction to the negotiation exercise, participants completed a cultural value survey and then were given role assignments and confidential information for their roles. Participants were paired into same-role dyads for a one-hour preparation session during which researchers circulated and were available for questions.

Then, participants were assigned to buyer–seller pairs and given one and one-half hours to negotiate. Negotiation dyads were assigned so that no two buyers who prepared together were paired with any two sellers who prepared together. The negotiation sessions were audiotaped.

After the negotiation, participants completed a form describing their agreement as well as questionnaires about their preparation and negotiation sessions. Then, results were shown to the group and participants were debriefed.

The independent variable was culture: U.S., French, Russian, Japanese, Hong Kong Chinese, and Brazilian. Brett et al. (1998) report the between-culture differences on cultural values, norms for negotiation, and power expectations in negotiation that were used to develop hypotheses.

The dependent variables were constructed by coding the transcripts of the negotiations. By coding, we mean categorizing direct information, indirect information, and power strategies represented in negotiators’ actual recorded conversations, according to the following criteria:

Direct Information-Sharing Strategies

• Preferences and priorities about an issue or offer

• Comparisons and contrasts, marked by a reference to similarities or differences between the parties

• Reactions including direct positive, neutral, or negative reactions to preferences, offers, or proposals

Indirect Information-Sharing Strategies

• Offers including single-issue, multi-issue, and multi-issue with trade-offs.

Power Strategies

• Informational persuasion, consisting of convincing with information and argument

• Informational persuasion with sanctions, consisting of persuasion with references to BATNA and/or reservation price

• Legitimacy, including information about one's own company or competitors

For example, consider the following exchange:

In this exchange, the seller's speaking turn was coded as informational persuasion (“the program has been extremely successful” and “it's got great value”) and a single-issue offer (“7 million”). The buyer's speaking turn was coded as a negative reaction (“that's not cheap”), informational persuasion (“it's risky for us”), and a question about a single-issue (“how about Strums?”). A summary of the codes with some additional examples appears in Appendix  One.

Hypotheses One through Five were tested by analyzing the relative frequency of use for each dependent variable in each culture. In other words, controlling for the fact that some dyads negotiated longer than others (different transcript length) and some people talked more than others (different numbers and lengths of speaking turns), we tested whether each culture group used direct information sharing, indirect information sharing, and power strategies more or less than the other culture groups.

The results across the six cultures are presented visually in Figures One through Nine. Note that in these figures, the zero-axis indicates perfect proportionality, that is, no cultural differences. Bars above the zero-axis indicate a strategy occurring with a frequency greater than we would expect if there were no cultural differences. Bars below the zero-axis indicate a strategy occurring with a frequency less than we would expect if there were no cultural differences. Empirical results of hypotheses tests are presented in the “Notes” section.¹

Hypothesis Two predicted that Russian and French negotiators would discuss preferences and priorities more than Hong Kong Chinese, Japanese, and Brazilian negotiators. This hypothesis was not supported, and in fact high levels of preferences and priorities in the Brazilian sample led to a significant effect in the opposite direction. Another look at Figure One shows that Russian, French, and Japanese negotiators were the least likely to discuss preferences and priorities.

Hypothesis Three predicted that Russian, Japanese, and Hong Kong Chinese negotiators were more likely to use indirect communication strategies than French, Brazilian, and U.S. negotiators. A loglinear analysis comparing Russia, Japan, and Hong Kong to Brazil, France, and the U.S. revealed support for the hypothesis: relative to the other three cultures, Russian, Japanese, and Hong Kong Chinese negotiators were more likely to share information indirectly through offers. As indicated in Figure Four, Russian and Japanese negotiators were extremely likely and U.S. negotiators were extremely unlikely to exchange information indirectly through offers, while Hong Kong Chinese did not use indirect communication as often as expected.

Hypothesis Four predicted that Russian and Japanese negotiators would use power strategies more frequently than the Hong Kong Chinese and Brazilians. This hypothesis was supported with respect to informational persuasion (Figure Five) and informational persuasion with sanctions (Figure Six), but not for legitimacy, which showed a trend in the opposite direction (Figure Seven).

Russian and Japanese negotiators were more likely than negotiators from the other four cultures to use power strategies in terms of reference to BATNA or reservation price (informational persuasion with sanctions) and persuasive argument (informational persuasion). Contrary to our prediction, Russian negotiators were the least likely to discuss company, competitors, and relationships (legitimacy) as a power tactic.

Hypothesis Five predicted that Hong Kong Chinese and Brazilian negotiators were more likely to use power strategies than French and U.S. negotiators. A loglinear analysis comparing Brazil and Hong Kong to France and the U.S. revealed support for the hypothesis with respect to informational persuasion: Brazilian and Hong Kong Chinese negotiators were more likely to use informational persuasion than French and U.S. negotiators (Figure Five). Brazilian and Hong Kong Chinese negotiators also referred to BATNA and reservation price as well as company, competitors, and relationships more frequently than U.S. negotiators but less frequently than French negotiators, so the hypothesis was not supported with respect to informational persuasion with sanctions or legitimacy.

Exploratory analyses revealed that U.S. negotiators were much more likely and Russian negotiators were much less likely than the other culture groups to ask questions (Figure Eight) and exchange general information about the product (Figure Nine) during the negotiation.

While we did not hypothesize about differences across cultures on questions or general information exchange because these are basic negotiation behaviors enacted to establish a common understanding of what is being negotiated, the fact that U.S. negotiators asked more questions and talked about general information relatively more than the other cultures could be further evidence of direct communication strategies supporting Hypothesis One. These negotiators may have been attempting to make sure that there was an explicit, common understanding of the situation before tackling the problem.

Brett et al. (1998) reported differential joint gains for the six cultures in this sample. By looking at behavioral frequencies of negotiation strategies, this study offers insight into why some cultures maximized joint gains and others did not.

Brett et al. (1998) report that U.S. and Japanese negotiators maximized joint gains. Our analysis of negotiation behaviors reveals that these groups took two different paths to joint gains. U.S. negotiators used a lot of direct information exchange and asked numerous questions. Presumably, this open exchange of information allowed negotiators to identify trade-offs and expand the pie. Japanese negotiators frequently used indirect information exchange and persuasion with intermediate amounts of questions and general information exchange. Japanese negotiators inferred the priority information necessary for trade-offs from offers, justifying their offers with persuasive arguments and occasional direct communication.

Brett et al. (1998) report that Brazilian and French negotiators achieved intermediate joint gains. Our analysis of negotiation behaviors helps to understand why French negotiators did not maximize joint gains but leaves some questions regarding the Brazilian negotiators. Brazilian negotiators frequently shared information on preferences and priorities, but they did not exchange other forms of direct information and made extremely few offers. It is unclear why information on preferences and priorities was not enough to identify trade-offs and maximize joint gains for Brazilian negotiators.

French negotiators spent their time establishing context. They exchanged general information on the products and they referenced their own company and competitors. They may have spent too much time clarifying the situation and not enough time identifying issues and priorities. French negotiators displayed extremely low frequencies of offers and persuasion.

Brett et al. (1998) report that Russian and Hong Kong Chinese negotiators generated the lowest joint gains of the six cultures studied. Our analysis of negotiation behaviors suggests that joint gains suffered for these groups because Hong Kong negotiators did not exchange enough information and Russian negotiators were too focused on power.

Similar to the Japanese, Russian negotiators exchanged many offers and frequently used informational persuasion and references to BATNA. The Russians differed from the Japanese because frequencies of other kinds of communication were extremely low. In contrast, the Japanese negotiators displayed intermediate frequencies of direct information exchange: asking questions and discussing general information. It seems that Russian negotiators focused on power and positioning to the exclusion of information necessary to expand the pie. This kind of negotiating may have been frustrating and confrontational, explaining a tendency to close prematurely.

Hong Kong Chinese negotiators used average levels of most behaviors measured. One exception was their frequent use of persuasion, suggesting a positional focus that may have prevented the identification of trade-offs. They also infrequently responded with direct reactions, suggesting that offers may have been proposed without parties making a commitment or moving forward. Because Hong Kong Chinese negotiators did not share much direct or indirect information, there may not have been enough information revealed to identify trade-offs. As with Russia, the Hong Kong Chinese data suggest a rather frustrating, ambiguous atmosphere that may have contributed to premature closure.

Pruitt (1981) proposes that maximizing joint gains depends on getting enough information out on the table, through either direct or indirect methods. Our data suggest that negotiators need to exchange information directly (e.g., U.S. negotiators) or they need to combine indirect information with some other form of direct communication (e.g., Japanese negotiators).

The French and Hong Kong Chinese negotiators simply did not exchange enough information. The Russian negotiators exchanged offers but did not balance the offers and a strong power focus with other forms of information exchange. The Brazilian negotiators suggest that direct information on preferences and priorities alone does not provide enough information to maximize joint gains. Perhaps negotiators also need to use direct reactions, like in the U.S., to move negotiations forward.

We can infer low/high context communication styles and power preferences from our behavioral data on direct and indirect information exchange and power-based influence. These inferences help to classify negotiation norms for the six culture groups.

The U.S. negotiators were clearly low-context and low-power as we originally hypothesized. They focused entirely on direct information exchange to the exclusion of indirect information and power-based influence.

Brazilian negotiators were extremely high in one form of direct information exchange and extremely low in indirect information exchange, suggesting that they relied primarily on low-context communication. They also did not use extreme amounts of any power behaviors, suggesting they were not particularly focused on power. The data clarifies prior descriptions of Brazil as both low- and high-power, but does contradict prior classifications of Brazil as high-context. The data may portray a Brazilian business culture, where low-context and low-power North American norms are becoming more prevalent.

Our behavioral data confirm prior accounts of Japan and Russia as high-context and high-power. Why these two cultures had such different negotiation outcomes is a provocative question for further inquiry. We mentioned that Russian negotiators did not exchange direct or general information. They may also have exchanged primarily single-issue offers, which do not convey as much information as multi-issue offers. In combination with the low norms for information exchange reported by Brett et al. (1998), this may have contributed to a much lower level of information and understanding than for the Japanese negotiators.

Our data does not confirm prior empirical and descriptive accounts of Hong Kong as a high-context, high-power culture. The Hong Kong Chinese negotiators did not display particularly large frequencies of indirect information exchange or power-based influence. One possible explanation is that the Hong Kong negotiators were undergraduates lacking the work and negotiation experience of the other participants.

Consistent with other accounts of France, we found no clear indicators of whether the French negotiators were low- or high-power. Intermediate frequencies of direct information exchange and low frequencies of indirect information exchange suggest that French negotiators were more low- than high-context. But, clear conclusions cannot be drawn from our data. It seems that, as with their negotiation outcome, French negotiators’ reliance on context and power is intermediate.

Our data show that, for the most part, negotiation strategies support the traditional classification of cultures based on values and norms (Hofstede 1980; Schwartz 1994; Hall 1976). Negotiators from Brazil and the U.S. used primarily direct information exchange and avoided power-based influence. Negotiators from Japan and Russia used primarily indirect information exchange and power-based influence. Negotiators from France used a mixture of behaviors that confirm descriptions of France as an enigma. Only negotiators from Hong Kong did not use behaviors that were consistent with cultural descriptions. The mixed negotiation outcomes across culture types are consistent with the findings of Brett et al. (1998) that culture type is not directly related to joint gains.

Our findings indicate a direct relationship between information exchange and joint gains in the six cultures. In a negotiation with integrative potential, getting information out on the table is critical to identify priorities and trade-offs. Our data suggest that relying on context alone to convey information is not enough. Even high-context negotiators must use some direct information exchange to generate joint gains.

The data also suggests that exchanging information on preferences and priorities alone is not enough. Other types of essential direct information exchange are comparisons and contrasts to identify trade-offs and direct reactions to keep the negotiation moving forward.

We found that power-based strategies were not directly related to joint gains in the six cultures. Japanese negotiators were able to use power-based influence and generate joint gains, but Russian and Hong Kong Chinese negotiators were unable to do so. It seems that power-based influence does not hinder joint gains in a culture with values and norms for power when there is sufficient information exchange. In this case, negotiators successfully enlarge the pie and use power-based influence to position and claim more. If power-based influence is used without sufficient information exchange, negotiators get stuck in positional bargaining and are unable to generate joint gains.

Our findings suggest that understanding the other parties’ negotiation strategies can help negotiators direct the negotiation process to facilitate the generation of joint gains. One way to understand the other party's strategies is to learn how the culture is classified in terms of values and norms and infer the strategies and behaviors likely to be used.

By understanding how the other party is likely to share information, a negotiator can strategically plan to exchange information to identify priorities and trade-offs. By understanding whether the other party is likely to use power-based influence, a negotiator can strategically plan how to integrate or avoid power tactics that may or may not be normative for the other party.

The next step in this research is to compare how differential use of these negotiation strategies relates to negotiation outcomes in intercultural negotiation. As opposed to looking at culture as having a main effect on negotiation strategies and outcome as we have done here, we also would like to look at how culture moderates the outcome effects associated with different strategies in different cultures. For example, Tinsley and Pillutla (1998) show that culture moderates the outcome effects of power scripts in U.S. and Hong Kong negotiations. Because we have not found values, norms, or strategies other than certain forms of information exchange directly related to joint gains in the six cultures, looking at culture as a moderator may help to explain differential joint gains across cultures.