an exploratory study - CiteSeerX

THE INFLUENCE OF CULTURE ON THE PROCESS OF BUSINESS NEGOTIATIONS: AN EXPLORATORY STUDY John L. Graham* University of Southern California Abstract. The processes of business negotiations in three countries, the United States, Japan, and Brazil, are compared and contrasted. Three dyads from each country were videotaped during a buyerseller negotiation simulation. Both verbal behaviors and nonverbal behaviors were observed and recorded. Observed differences provide the basis for hypothesized differences which might be tested in future work. Despite the increasing importance of cross-national commercial relationships, business negotiations in different countries have received little attention. During the 1970s a few articles appeared in business journals (for example, Van Zandt, 1970; Jastram 1974; Kapoor 1974; and Wells 1977), but they are primarily descriptive and often anecdotal. Recently, more systematic studies of negotiations in foreign countries have been undertaken. Tung (1982) considered business negotiations between American and Chinese (that is, People's Republic of China) executives. Harett and Cummings (1980) compared bargainers' characteristics and behaviors across several cultural groups (1980). Graham (1983) investigated the determinants of business negotiation outcomes in the United States, Japan, and Brazil. These latter studies have proven valuable, but limited in their use of questionnaire items as measures of negotiation processes. In this study simulated business negotiations in three countries-the United States, Japan, and Brazil-are analyzed using observational methods. That is, exploratory methods are employed in which negotiation processes are recorded on videotape and both verbal and nonverbal behaviors during negotiations are compared across the three cultures. The remainder of this paper is organized into 5 sections. First, the theoretical perspective is briefly discussed. Next, methods of data collection are described. Third, the results of the analyses of verbal behaviors are presented. The fourth section consists of the analyses of nonverbal behaviors. Finally, the results are summarized, conclusions are drawn, and hypotheses for future testing are suggested. * John L. Graham is AssistantProfessorof

at the Schoolof Business,UniMarketing of Southern work This was California. versity supportedin partby a grantfrom Motor to Sales the International BusinessEducationand Toyota USA, Incorporated, ResearchProgram(IBEAR)at the School of Business,the Universityof Southern California. Date Received: June 7, 1983; Revised: February 22, 1984; Accepted: April 10, 1984. 81

Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve, and extend access to Journal of International Business Studies ® www.jstor.org

82

JOURNAL OF INTERNATIONAL BUSINESS STUDIES, SPRING 1985

THEORETICAL PERSPECTIVE

Here the theoretical perspective of the research is briefly discussed to provide a context for the presentation of the exploratory methods used and the results reported in later sections. (For a comprehensive review of the relevant literature see Graham 1980.) Most recently, the outcomes of business negotiations have been hypothesized to be the result of several factors which can be classified into 3 categories or kinds of theoretical constructs-individual characteristics, situational constraints and process measures (see Sawyer and Guetzkow 1965, Rubin and Brown 1975). Many empirical measures of both individual characteristics and situational constraints have been tested in previous research. Examples of such individual characteristics might be intelligence, self-esteem, credibility, attractiveness, and power. Examples of situational constraints might include company goals, time limitations, or unequal power relations. Process Measures

A few studies have focused on the process of business negotiations (for example, Pennington 1968; Pruitt and Lewis 1975; Lewis and Fry 1977; Dwyer and Walker 1981). Graham(1983) has defined process measures as "qualitative and quantitative descriptions of the activities involved in a business negotiation-for example, bargaining strategies used" (page 47). Based on an extensive review of the negotiation literature, Rubin and Brown (1975) conclude that the behaviorsof bargainersduring the negotiation process affect negotiation outcomes. The kinds of behaviors listed are opening moves, countermoves, types of appeals, demands, and the like. But little work has been done to investigate relationships among process measures and negotiation outcomes, individual characteristics and situation constraints. Moreover, because researchershave only recently turned their attention to process measures, concepts and operational definitions remain vague and relations not adequately specified. The present study focuses on the development of operational definitions of process measures using observational methods. Further, special attention is given to cultural differences in these process measures. CulturalDifferencesin BusinessNegotiationProcesses

Culture has been a difficult concept to deal with in any consistent way. Anthropologists and sociologists have been arguing over definitions for years. Culture has appeared in the marketing literature primarily as a determinant of buyer behavior (for example, Engel and Blackwell 1982), but operational definitions seem to have varied from study to study. Perhaps the most widely accepted definition is that professed by Linton: "A culture is a configuration of learned behaviors and results of behavior whose component parts are shared and transmitted by members of a particular society" (1945). The important part of the definition for the present research is the idea that behaviors are shared by members of a particular

INFLUENCE OF CULTURE ON BUYER-SELLER NEGOTIATIONS

83

culture. Or as Spiro put it, "members of a given society behave in uniform and predictable ways" (1950). A central goal of the study is to discover what shared behaviors manifest themselves during business negotiations in Brazil, Japan, and the United States. In addition to bargainingbehaviorsbeing consistent within cultures, several authors have suggested that negotiation processes differ across cultures (for example, Sawyer and Guetzkow 1965; Kay 1970; Frake 1970; Van Zandt 1970; Condon 1974). Therefore, a second purpose of this work is to identify how Brazilian, Japanese, and American bargaining processes might differ from one another. METHODS

In this section, the procedures of the study are presented. First, the sample of business people participating in the research is specified. Next, the laboratory setting is described. Third, data collection instruments are presented. Sample

The participants in the experiment are 6 Brazilian businessmen, 6 Japanese businessmen and 6 American businessmen. All were participating in executive or MBA programs in the United States, and all volunteered for this study. All received college education and all have at least 2 years' business experience in their respective countries. The sample was limited to experienced businessmen because Fouraker and Siegel (1963) reported differences in the bargainingbehavior of students and businessmen. A comparison of the experience characteristics of the three cultural groups is included in Table 1. The participants were asked to play the role of either buyer or seller in a two-person buyer-seller negotiation simulation. Laboratory Setting

The negotiation simulation, developed by Kelley (1966) and recently used by Pruitt and Lewis (1975), Lewis and Fry (1977), and Graham (1983), involves bargaining for the prices of 3 products. Final agreements are a combination of 3 prices (that is, letters), 1 for each product. As can be seen in the Appendix the ideal joint solution is AEI, with both buyer and seller making a profit of 52. Differing amounts and types of background information can be included with the basic pay-off matrices, depending on the focus of the research.The simulation is simple enough to be learned quickly, but complex enough to provide usually one half hour of face-to-face interaction. (See procedures.) Nine separate negotiations were conducted-3 Japanese/Japanese, 3 American/American, and 3 Brazilian/Brazilian. Data Collection

The 9 interactions were videotaped using a wide-angleperspective to capture postures, body movements, and interpersonal distances. The camera was mounted in the ceiling of a behavior science laboratory at an American

TABLE1

ProcessMeasuresDevelopedUsingObservationalMethodologiesand Par

Category

Verbal Behaviors

Symbol

BBI to BB12

Variable

Description and Measure

Bargaining behaviors

Angelmar and Stern's (1978) content analysis scheme; the twelve categories are listed a defined in Table 3.

No

Number of "Nos"

Number of times the word " used by bargainer per half h

01

First Offer

The profit level associated w each participant's first offer Buyer offers to buy at price

IC

Initial Concession

The difference in profit asso with the first and second off

SP

Silent Periods

OV

Conversational overlaps

The number of conversationa (10 seconds) per half hour Number of interruptions per

FG

Facial Gazing

Number of minutes players looking at opponent's face, randomly selected ten-minut

T

Touching

Incidents of bargainers touch another per half hour (not i hand shaking)

AAA = profits

Nonverbal Behaviors

Participant's Characteristics

Age Years Work Experience Percentage of "your work involving contact with people outside your firm?"

* N = 5, one Brazilian bargainer never made a first offer or concession.

of 80)


85

university. It was operated by remote control, thereby minimizing obtrusions. Additionally, all participants were asked to evaluate the obtrusiveness of the setting on questionnaires following the negotiation game. All participants reported a minimum of discomfort. ANALYSISOF VERBALBEHAVIORS

As mentioned, a primary purpose in this exploratory work is the identification and clarification of process measures. Consequently, the discussions in the sections to follow are organized as a "list" of process measures. Associated with each item on the "list" are operational definitions, a brief account of the method of measurement, and mention of apparent differences among the 3 cultural groups-Brazilian, American, and Japanese. Table 1 lists the variables developed in this part of the study. They are ordered in the sequence of the discussion to follow. The first step in the measurement and analysis of verbal behaviors during the business negotiations is the transcription of the audio portion of the videotapes. This is a potential source of error in measurement. A complete check of the transcripts did reveal some minor mistakes, and these were corrected. Generally, errors in the transcription process were found to be inconsequential. The second step in the measurement and analysisprocess consisted of translation of the Brazilian and Japanese interactions in which native languages were spoken throughout. Here exists another potential source of measurement error. The translation process involved 2 steps: (1) transcription of the tapes into Portuguese and Japanese, and (2) a meaning translation. Examples of transcripts can be found in the Appendix. All these steps in the process proved to be very time consuming and expensive, thus putting very real limitations on the number of interactions which might be analyzed. The transcripts were further analyzed using 3 schemes. First, a content analysis scheme was employed to identify the bargainingbehaviors of each participant. Second, each participant's concession strategies were summarized. Third, the number of times bargainersused the word "no" was tallied. Each of these steps is discussed in detail below. Content Analysis

Measurement. Recently, Angelmar and Stem (1978) have described a content analysis scheme developed specifically for the analysis of bargaining communications in business settings. Utterances by participants are classified into 12 categories. The categories and definitions are listed in Table 3. Examples of coding are included in the Appendix. Angelmar and Stern report positive results from a reliability and validity assessment of the system applied to written communications. The present study includes the first attempt to apply the scheme to transcripts of conversations. Coding transcribed conversations is a more difficult undertaking; spoken words are the only channel of communication. Transcriptsdo not include information communicated through other channels such as proxemics, prosody,

86


kinesics, or facial expression. Theory indicates that these channels also may be important for accurate interpretation and measurement of conversational contributions.

Two coders were employed in classifying segments of the conservation into 12 bargaining categories. This investigator coded all 9 interactions and a research assistant (ignorant of the theory and hypotheses involved in the study) coded 3 interactions to provide a reliability check. The author is cognizant of the possible bias involved in using coders informed about the theory applied in the research. However, resource constraints necessitated this less-than-ideal state. Significantly, analysis of discrepancies in coding between the 2 coders revealed this source of bias to be minimal. Despite the limitations described above, the coding system proved adequately reliable in this work. As can be seen in Table 2, both unitizing and coding reliabilities compare favorably with those reported by Angelmar and Stem (1978, p. 99). TABLE2 Content AnalysisScheme Reliabilityand Content Validity Type of Test

Unitizing Reliability

Coding Reliability

Content Validity

Type of Analysis Performed Angelmar & Stern's Japanese American (1978) Results

Brazilian

Differences in the number of units between coders as a percentage of sum of units

3.0%

9.5%

3.5%

3.4%

Agreements concerning unit boundaries as percentage of sum of agreements and disagreements

69%

77%

86%

85%

66% Agreements as percentage of total number of codings (Number of units coded

69%

59%

63%

190

556

482)

Percent units coded in "other" category

10.9%

6.7%

17.1%

0.7%

1.8%

6.5%

542

326

2-3%

Percent Units coded in "restatements and clarifications" category (Number of units coded

1719)

The content validity of the scheme applied to conversational data is a more difficult issue. A substantial number of units were classified into the "other"

category (17.1%) and to the "restatements and clarifications" category (6.5%) for the Brazilians. In the latter case, a statement which duplicated the statement previous to it was classified separately. The statements which did not fit into any of the other 12 categories were assigned to the "other" category. The large discrepancy between Angelmar and Ster's results (2.3%) and the findings for the study (17.1%) is due to the aforementioned


87

differences between written and spoken communication. Many conversational bits were incomplete (that is, in some cases not even complete words) and consequently unclassifiable. This latter circumstance was strikingly true for the Brazilian interactions. As will be discussed in more detail in a later section, the Brazilian style of conversation apparently includes many more interruptions than either the American or Japanese styles. This, in addition to longer negotiations (Americans averaged 24 minutes, Japanese 33 minutes, and Brazilians 51 minutes), accounts for the larger number of units coded for the Brazilian interactions. Findings. Based on the classification of the various units, percentages of each bargaining category were calculated for each participant for each negotiation. These findings are summarized in Table 3. As the data indicate, the percentages of each category are surprisinglyconsistent across 3 groups. The only substantial discrepancies are that the Brazilians appear to make fewer promises and commitments and more commands than either the American or Japanese business people. Examination of the units classified into the "other" category suggests inclusion of a 13th category to be included in future work. A substantial number of the 17.1% "other" units could be classified as "statements of fact," regardingthe other bargainers' previous statements or actions. Below are listed examples of such units classified into the "other" category. "Youknowthatis not true." "Butyou arethe onewho confessedto me, Curte,thattelevision,you havea selling problem..." " .. at the endyou cameherecrying-atthe lastminuteyou wantsomeone...." FindingsRelated to the Patternof BargainingBehaviors

Theory suggests that the process of marketing negotiations differs across cultures. More specifically, this author (1980) reported that the "pattern" of interaction in Japanese sales negotiations differs from that of American sales negotiations. Pattern here is defined as a time-ordered sequence of events. One way to view the "pattern" of interaction is through the perspective of content analysis-the various bargaining behaviors being the "specific events" referred to above. The summary data presented in Table 3 provides some information about pattern, but it provides no information regardingtime-ordering. In this section the dimension of time is added to the analysis. Each interaction was divided into 4 equal periods. Thus, a 40-minute interaction was divided into 10-minute quarters, a 20-minute interaction into 5-minute quarters, and so on. Percentagesof each bargainingbehavior were calculated for each participant for each quarter, and the percentages were averaged across participants. The results of this analysis are presented in Table 4. Three propositions (suggested by Van Zandt 1970) regardingdifferences in the pattern of interaction can be investigated through the data displayed in Table 4. The first, that Japanese ask more questions, is not supported

88


TABLE3 SummaryContent Analysis Findings Units in individual categories as percentage of total units Symbol

Bargaining Behaviors and Definition

BB1

Promise. A statement in which the source indicated his intention to provide the target with a reinforcing consequence which source anticipates target will evaluate as pleasant, positive, or rewarding. Threat. Same as promise, except that the reinforcing consequences are thought to be noxious, unpleasant, or punishing. Recommendation. A statement in which the source predicts that a pleasant environmental consequence will occur to the target. Its occurrence is not under the source's control. Warning.Same as recommendation, except that the consequences are thought to be unpleasant. Reward. A statement by the source that is thought to create pleasant consequences for the target. Punishment. Same as reward, except that the consequences are thought to be unpleasant. Positive normative appeal. A statement in which the source indicates that the target's past, present, or future behavior was or will be in conformity with social norms. Negative normative appeal. Same as positive normative appeal, except that the target's behavior is in violation of social norms. Commitment. A statement by the source to the effect that its future bids will not go below or above a certain level. Self-Disclosure. A statement in which the source reveals information about itself. Question. A statement in which the source asks the target to reveal information about itself. Command. A statement in which the source suggests that the target perform a certain behavior. Other. Unclassifiable statements. Restatements and Clarifications. A statement which duplicates the speaker's previous statement.

BB2

BB3

BB4 BBS

BB6 BB7

BB8

BB9

BB10 BB11

BB12 BB13 BB14

Japanese American

Brazilian

7

8

3

4

4

2

7

4

5

2

1

1

1

2

2

1

3

3

1

1

0

3

1

1

15

13

8

34

36

39

20

20

22

8

6

14

not included in the analysis not included in the analysis

by the data. The percentages of questions for the 3 groups is surprisingly similar and consistent across quarters. The second proposition is that Americans use more aggressive persuasive tactics than Japanese. Looking at totals for commands reveals little difference between the 2 cultural groups. However, viewing these categories across quarters reveals an interesting pattern. Americans appear to being using aggressive persuasive tactics


89

earlier than the Japanese. Notice the difference in quarter 2 in commands, and the apparent change in pattern in quarters 3 and 4. The conclusion would be that Americans do not necessarily use "more" aggressivetactics than Japanese bargainers, but they do begin using them "sooner" in the interaction. The third proposition is that Japanese negotiators tend to make concessions toward the end of the negotiations. An examination of the data in Table 7 does not support this final proposition. Indeed, the Americans appear to make more commitments toward the end, the percentage growing through quarters 1 to 4 (3%, 9%, 13%,28%). The pattern of commitments for Japanese bargainersis more consistent (4%, 19%, 12%, 19%). This last proposition will be discussed further when concession strategies are examined from a somewhat different perspective. TABLE4 A Comparisonof Patternsof BarginingBehaviors (percentbargainingfor Japanese,N = 6 / Americans,N = 6 / BraziliansN = 6) Behaviors

promise threat recommendation warning reward punishment positive normative appeal negative normative appeal commitment self-disclosure question command * 4/6/1

1st Quarter

2nd Quarter

3rd Quarter

4th Quarter

4/6/1* 3/1/1 10/6/5 0/2/1

6/3/3 1/7/3

9/13/3 4/3/1

8/9/3 7/3/3

2/3/3 3/1/1 0/2/1 0/7/3 1/0/0 3/0/1

5/5/4 3/1/0

7/1/6 1/0/0

2/2/1

0/1/2

1/2/4 0/1/1

19/9/12 33/33/34 29/22/26

12/13/3 29/34/45 16/20/18 16/9/16

11/04 0/2/0 2/4/0 19/28/11

1/1/0 3/1/1 15/13/8

31/30/32 15/20/22 9/2/18

34/36/39 20/20/22 8/6/14

3/4/4 2/1/1 2/0/1 2/1/1 4/3/6 46/48/46 21/20/29 5/8/4

5/14/15

3/0/1

Total 7/8/3 4/4/2 7/4/5 2/1/1 1/2/2 1/3/3

represents promises = 4% for Japanese, 6% for Americans, and 1%for Brazilians.

The only noticeable difference in the Brazilian bargainingprocess appears to be in the use of commands. Like the Americans, the Brazilians appear to use commands more toward the later stages of the negotiation (4%, 15%, 16%, 13%). FindingsRelated to the Use of the Word"No"

During the process of coding the Brazilian negotiations, it became obvious that the Brazilian bargainers tended to use the word "no" much more frequently than either American or Japanese bargainers. Therefore, all transcripts were searched for the use of the word "no," and each was tallied. The difference between the Brazilian transactions and the others is striking (see Table 1). Most often, the use of "no" in the Brazilian negotiations was not as an answer to a question, but instead, it preceded a statement of disagreement. Four separate examples are listed below.


90

"Ah! No way. There'sno way, no, typewriterswe have settled already." "No, you gaveme 60 for televisions." "No, no, no."

"No. No, I didn't agree.We did some arrangement." BiddingBehaviors

For this part of the analysis, profit levels associated with participants' bids on the 3 products in the negotiation simulation were summarized. Profits associated with first offers (01) and initial concessions (IC = first offers less second offers) were compared across the 3 groups. As can be seen in Table 1, the Brazilian bargainersasked for much more in the way of profits initially; similarly initial concessions also appear to be higher for the Brazilian bargainers. As noted in Table 1, one of the Brazilian bargainerssucceeded in avoiding making any commitments or counter offers until the final handshake. This approach is unique in all the bargainingevents observed by this author, in all countries and in both the field and the laboratory. ANALYSISOF NONVERBALBEHAVIORS

In this section of explorations into the process of buyer-seller negotiations, nonverbal aspects of the videotaped interactions were considered. First the rhythm of the conversations will be discussed, specifically examining silent periods and conversational overlaps. Next, gaze direction of the participants will be considered. Lastly, findings related to touching duringnegotiations are presented. All measurements in this section have been derived irrespective of the verbal content of the interactions to avoid potential bias. ConversationalCoordination

Communication theory suggests that when 2 people are effectively sharing ideas, their communication behaviors-both verbal and nonverbal-will be rhythmically coordinated (Condon 1968; Gumperz 1979; Erickson 1976). Here 2 measures of conversational coordination, "silent periods" and "conversational overlaps," are operationally defined, and findings are reported below. Silent Periods. Silent periods are operationally defined as gaps in conversations 10 seconds or more in duration. The time period of 10 seconds was selected somewhat arbitrarily, but it is a long enough period of silence to appear unnatural to most American observers. The tapes were searched for gaps in conversations of 10 seconds or more, and these gaps were noted on the transcripts and tallied (see Table 1). Silent periods occurred more frequently in the Japanese interactions (5.5/ 30 min.) than either the American (3.5/30 min.) or Brazilian (none were discovered. Conversational Overlaps. The concept of "interactional synchrony," the unconscious coordination of verbal and nonverbal behaviors of 2 or more participants in a conversation, is discussed at length by this author (1980).


91

One possible measure of this construct is the number of conversational overlaps or interruptions during a conversation. Conversational overlaps are defined here as periods when both speakers are talking simultaneously, or when the conversational contribution of one speaker overlaps that of the other speaker. Identification of such overlaps is independent of the verbal content of the interactions. In the present work, the videotapes were searched for overlaps, and such interruptions in the flow of conversation were noted on the transcripts. The number of overlaps (interruptions) by each participant was totaled and divided by the time of negotiation to arriveat values (OV) which may be compared across interactions. Table 1 shows that the Brazilian bargainersinterrupted more than twice as often as either the Japanese or American bargainers.Moreover, a review of all the videotapes revealed another difference. More than just interrupting one another, the Brazilianbargainerstended to talk simultaneously for extended periods of time, seemingly "fighting for the floor." Alternatively, when the American and Japanese bargainers interrupted, one speaker or the other stopped talking. Facial Gazing

The third nonverbal variable to be considered in this part of the study is facial gazing. Other researchers have found significant relationships between facial gazing and outcomes of negotiation games (Lewis and Fry 1977). Moreover, several authors have suggested differences in facial gazing behavior across cultures (Argyle and Cook 1976). In this study, facial gazing is defined as the percentage of time a bargainer gazes at the face of his opponent. Randomly selected, 10-minute videotape excerpts of each of the 9 interactions served as data here. Using a stopwatch, 2 observers recorded the time each participant spent gazing at his opponent's face during the 10 minutes, giving the measure FG (see Table 1). The method used was very similar to that reported by Lewis and Fry (1977), except that here videotapes were reviewed rather than realtime interactions. Using videotapes is a more reliable technique, allowing reviews and reliability checks. Intercoder reliability (calculated using Guetzkow's (1960) formula, the difference in the number of units between coders as a percentage of sum of units) was found to be 9%, an adequate level. Apparent differences were discovered across the three cultural groups. The Brazilians spent considerably more time gazing at negotiation opponents than either the American or Japanese bargainers(see Table 1). Touching

In none of the 6 American and Japanese interactions did a bargainertouch another bargainer,except for beginning and ending hand shakes. However, as indicated in Table 1, Brazilian bargainerstouched one another relatively frequently and in all three sessions.

92

JOURNALOF INTERNATIONALBUSINESSSTUDIES, SPRING 1985

CONCLUSIONS

The apparent differences between Japanese and American processes of buyer-seller negotiation can be summarizedalong 6 dimensions of observed behavior. Contrary to the hypothesis suggested by Van Zandt (1970), Japanese, like Americans, use aggressive persuasive tactics. However, in Japanese negotiations they appear to be used only by buyers and in the later stages of the negotiations (when all else fails). This is consistent with the differences in the roles and behaviors of Japanese buyers and sellers described by Nakane (1970) and Graham (1983). Silence is a strong persuasive strategy often used in Japanese negotiations (Ueda 1974 and Van Zandt 1970). The Japanese use the word "no" less frequently as Ueda (1974) predicts. Facial gazing is much less frequent duringJapanesenegotiations. Finally, 2 other process variables seem to be culture-specific-first offers and initial concessions. In the 3 interactions taped, Japanese asked for consistently higher profit solutions when making the initial offer in a negotiation. Americans were more apt to offer a "fair" price, one that was closer to the eventual solution. Americans also tended to make larger initial concessions. Other than these 6 differences, American and Japanese bargainingprocesses appear to be similar. Alternatively, observations of Brazilian executives participating in the negotiation simulation suggest a number of striking differences. The Brazilian businessmen made fewer commitments and more commands. First offers were more "greedy" than either the Japanese or the American offers. The Brazilian bargainers disagreed with bargaining partners (as measured by the number of "nos") with strikingly greater frequency than either the Japanese or American businessman. Finally, Brazilian bargainers interrupted very frequently, spent more time gazing at partners' faces, and touched during the bargaining process. That is, the Brazilian bargaining process observed appears to be different from both the Japanese and American style in almost every respect. These results, of course, are not definitive. The small sample size does not allow for tests of statistical significance. How representative the participants are is problematic. They are all experienced businessmen, which is an improvement over most other business negotiation research where students are used as surrogates for bargainers. But how well 6 businessmen represent a "cultural style" cannot be determined. Finally, external validity of the experimental setting is questionable. However, this work represents an improvement over most other business negotiation researchby the use of face-to-face communication instead of written or electronic means. The value and strength of this study is due to the observational methods used to measure the negotiation process. Videotaping allows for multiple observers and multiple observations concurrent with the bargaining process. Thus, the reliability and validity of the process measures developed does not depend on a priori experimental manipulations or post hoc participant self-reports. The methods developed in this study are time consuming and expensive (translations) but, as these findings suggest, potentially fruitful.

INFLUENCE OIF CULTURE ON BUYER-SELLER NEGOTIATIONS

93

Cross-culturalinteractions, or bargaining between buyers and sellers from different countries, were not considered here. This author (1980) does report findings from such studies. Generally, the findings indicate that such cultural differences in bargaining process as described above are potential sources for friction and misunderstandings between bargainers and increased transaction costs for international commercial relationships. Indeed, the findings of this study suggest that Americans and Japanese will have difficulty bargainingwith Brazilian business people. To the American eye, the Brazilian bargaining style might be described as very aggressive, pushy, or even rude. One of the important areas of future research in this area will be the examination of American/Brazilian bargaininginteractions. Who makes adjustments, and what things change? The findings of this study suggest that substantial differences in bargaining style exist across cultures. Reliable and valid measures of negotiation processes have been developed (see Table 1). This exploratory work deserves follow-up research with larger sample sizes. Increased statistical power associated with larger samples would allow for investigations of not only the culture - process relationship but also more complex relations such as culture - process -> negotiation

The findings of such studies

outcomes.

will hold important implications for training business executives and students to manage more efficiently the international transactions of the future, increasingly more global market place. APPENDIX A Payoff Matricesfor Kelley's (1966) NegotiationGame Prices1 Televisions A B C D E F G H 1

Seller Profits

Buyer Profits2

40 35 30 25 20 15 10 5 0

Typewriters 24 21 18 15 12 9 6 3 0

Air Conditioners 16 14 12 10 8 6 4 2 0

Televisions Typewriters 0 2 4 6 8 10 12 14 16

0 3 6 9 12 15 18 21 24

Air Conditioners 0 5 10 15 20 25 30 35 40

1 If bargainers were to agree on price "E" for all 3 products, then Buyers would make a total profit of 40 (20 + 12 + 8), and Sellers would make a total profit of 40(8 + 12 + 20). 2. Profits are adjusted (multiples of those listed above) to reflect realistic levels, given the products, markets, and currencies involved.


94

APPENDIX A Details of Laboratory Procedures 1. Each participant was allowed 15 minutes to read the written instructions (i.e., either a buyer or seller position sheet and appropriate payoff matrix) and plan negotiation strategies. Questions of clarification were answered during this time. 2. At the end of the 15 minute preparation period, the participants were seated across from one another at a table, given final verbal instructions, and left alone. 3. The final instructions consisted, in part, of the following statements: "The game usually takes about thirty minutes to complete." "There is a one-hour time limit."

APPENDIX B Transcript of Excerpt of Brazilian Negotiation (Translated from Portuguese)

•

3

B1 Humm.

S Let's give a final review. /we would stay with option . . for you, "H" for typewriters and could be "H" for air conditioning/but the delivery will not be parcelled-there's no way . . . we must compensate it in some way, gosh, I cannot work (j./ I want-televisions we would go back to option "C." I think I'm proposing a ^when ( good deal to you. B Yes, but we must start. . . Curte . . from this point: what is settled is not expen9:55:50 sive ... I'D S Then let me .. let me... ({ B . . . we have agreed for television letter "B" with 60 days term . . . S Weare... ( 9:56:01 B . . .then we won't discuss this anymore. Let's discuss the other two packages, ~ otherwise we'll go back all over again . . . ()3L~ S No, no, now let's, well. . . B. . . and we keep going and coming. .. S Let's make a little concession . . . > B . . .the day after tomorrow we'll still not have a good deal. @ S For me the air conditioner is an appliance, I mean-what we are offering is our lowest cost already, "I" is... B But you are the one who confessed to me, Curte, that televisions you have a selling (i problem . . . S No, there's a request. .. i B There will be no more of this model. . . ( S No televisions... ah, ok! Televisions yes, ok, all right. > B It's ok, but then you . . . I ok but. .. S It's B . . .you already gave me "B," 60-day terms. I would like not to talk about this anymore, because my promotion is with typewriters and I can see a relatively good market for them . . . S Humm. ( 9:55:212

3

1 B = buyer S = seller 2 time in hours:minutes:seconds 3 coding of Bargaining Behaviors (Angelmar and Stern 1978)


95

REFERENCES Angelmar, R., and Stern, L. W. "Development of a Content Analytic System for Analysis of Bargaining Communication in Marketing." Journal of Marketing Research, February 1978, 93-102. Argyle, M., and Cook, M. Gaze and Mutual Gaze. Cambridge, England: Cambridge University Press, 1976. Bagozzi, R. P. "Marketing as Exchange: A Theory of Transactions in the Market Place." American Behavioral Scientist, March-April, 1978. Blumer, H. Symbolic Interactionism Perspective and Method. Englewood Cliffs, NJ: Prentice-Hall, 1969. Bonoma, T. V. "Buyer-Seller Transactions in Context." Paper presented for the Association of Consumer Research Annual Conference, Atlanta, Georgia, October 28-31, 1976. Condon, J. C. "Perspective for the Conference." In J. C. Condon and M. Saito, eds.,Intercultural Encounters with Japan. Tokyo: Simul Press, 1974. Condon, W. S. "Linguistic-Kinesic Research and Dance Therapy." A.D. T,A. Convention Proceedings, 1968. Dwyer, R. F. and Walker, 0. C. "Bargaining in an Asymmetrical Power Structure." Journal of Marketing, Winter 1981, 104-115. Engel, J. F. and Blackwell, R. D. Consumer Behavior. Chicago: The Dryden Press, 1982. Erickson, F. "Talking Down and Giving Reasons: Hyper-explanation and Listening Behavior in Inter-racial Interviews." Paper delivered at the International Conference on Non-verbal Behavior, Ontario Institute for Studies in Education, Toronto, Canada, May 11, 1976. Fouraker, L. E. and Siegel, S. "Bargaining Behavior." New York: McGraw-Hill, 1963. Frake, C. 0. "Struck by Speech: The Yakan Concept of Litigation." In J. C. Gumperz and D. Hymes, eds., Directions in Sociolinguistics. New York: Holt, Rinehart and Winston, 1972. Graham, J. L. "Cross-Cultural Sales Negotiations: A Multilevel Analysis." Unpublished Ph.D. dissertation, School of Business, University of California, Berkeley, 1980. Graham, J. L. "Brazilian, Japanese, and American Business Negotiations." Journal of International Business Studies, Spring/Summer 1983, 47-61. Green, P. E.; Gross, I.; and Robinson, P. J. "Behavioral Experiment in Two-Person Bargaining." Journal of Marketing Research, November 1967, 374-379. Guetzkow, H. "Unitizing and Categorizing Problems in Coding Qualitative Data." Journal of Clinical Psychology 6, (1960): 47-53. Gumperz, J. J. "Sociocultural Knowledge in Conversational Inference." 28th Annual Roundtable Monograph Series on Languages and Linguistics, Georgetown University, 1979. Harnett, 0. L. and Cummings, L. L. Bargaining Behavior. An International Study. Houston, TX: Dane Publications, Inc., 1980. Jastram, R. W. "The Nakado Negotiators." California Management Review Vol. XVII, No. 2, Winter 1974, 88-90. Kapoor, A. "MNC Negotiations: Characteristics and Planning Implications." Columbia Journal of WorldBusiness, Winter 1974, 121-130. Kay, P. "Some Theoretical Implications of Ethnographic Semantics." Current Directions in Anthropology, Bulletins of the American Anthropological Association, Vol. 3:2, 1970. Kelley, H. H. "A Classroom Study of the Dilemma in Interpersonal Negotiations." In K. Archibald, ed., Strategic Interaction and Conflict. Berkeley Institute of International Studies, University of California, 1966. Lewis, S. A. and Fry, W. R. "Effects of Visual Access and Orientation on the Discovery of Integrative Bargaining Alternatives." OrganizationalBehavior and Human Performance 20, 1977, 75-92. Linton, R. The Cultural Background of Personality. New York: Appleton-Century-Crofts, 1945. Mathews, H. L.; Wilson, D. T.; and Monoky, J. F. "Bargaining Behavior in a Buyer-Seller Dyad." Journal of Marketing Research, February 1972, 103-105. Nakane, C. Japanese Society. Berkeley, CA: University of California Press, 1970. Pennington, A. L. "Customer-Salesman Bargaining Behavior in Retail Transactions." Journal of Marketing Research, August 1968, 255-262. Pruitt, D. G. and Lewis, S. A. "Development of Integrative Solutions in Bilateral Negotiation." Journal of Personality and Social Psychology, Vol. 31:4, 1975, 621-633.

96


Rubin, J. Z. and Brown, B. R. The Social Psychology of Bargaining and Negotiation. New York: Academic Press, 1975. Sawyer, J. and Guetzkow, H. "Bargaining and Negotiation in International Relations." In H. C. Kelman, ed., International Behavior, A Social-Psychological Analysis, New York: Holt, Rinehart and Winston, 1965, 464-520. Spiro, M. "Culture and Personality: The Natural History of a False Dichotomy." Psychiatry, 14:1, 1950, pp. 19-46. Thiabaut, J. W. and Kelley, H. H. The Social Psychology of Groups. New York: John Wiley and Sons, Inc., 1959. Tung, R. L. "U.S.-China Trade Negotiations." Journal of International Business Studies, Fall 1982, 25-38. Ueda, K. "Sixteen Ways to Avoid Saying 'No' in Japan." In J. C. Condon and M. Saito, eds.,Intercultural Encounters in Japan. Tokyo: Simul Press, 1974. Van Zandt, H. F. "How to Negotiate in Japan." Harvard Business Review, November-December 1970, 45-56. Wells, L. T. "Negotiating with Third World Governments." Harvard Business Review, JanuaryFebruary 1972, 72-80.