The Effect of Power Differences on the Outcome of Consumer Bargaining Situations

ABSTRACT - This study investigated a new power-based model of consumer transaction systems by studying some of the conditions sufficient for their collapse and/or alteration. Sixty bargaining interactions were conducted between automobile salesmen and a well-coached confederate. The outcomes provide preliminary evidence to partially substantiate and encourage further research on the model.


Wesley J. Johnston and Thomas V. Bonoma (1984) ,"The Effect of Power Differences on the Outcome of Consumer Bargaining Situations", in NA - Advances in Consumer Research Volume 11, eds. Thomas C. Kinnear, Provo, UT : Association for Consumer Research, Pages: 170-174.

Advances in Consumer Research Volume 11, 1984      Pages 170-174


Wesley J. Johnston, The Ohio State University

Thomas V. Bonoma, Harvard University


This study investigated a new power-based model of consumer transaction systems by studying some of the conditions sufficient for their collapse and/or alteration. Sixty bargaining interactions were conducted between automobile salesmen and a well-coached confederate. The outcomes provide preliminary evidence to partially substantiate and encourage further research on the model.


Consumer buying behavior, like most of our daily activities, almost always occurs as a public, social, interactive happening. By "social" we do not simply mean that the consumer gets some of his or her information about products from relatives, peers, or other sources who enter the choice equation as another stimulus. Rather, we mean that the act of buying itself ordinarily occurs as part of an "encounter" (Deutsch 1973) or interaction with one or more other people making up the social universe.

Perhaps the best general conceptualization of this social conte>:t for marketing behavior to date has been that of exchange (Bagozzi 1975; Johnston 1978; Krech et al. 1962). With the adoption of the view of marketing as exchange, the smallest meaningful unit of analysis in buying transactions becomes the dyad. "A 'dyadic' or social view of consumer behavior uncovers, emphasizes, and makes explicit the rational quality and character o f interaction between the buyer, consumer or recipient of the goods/services and the seller, marketer, or provider" (Bonoma and Zaltman 1979). The basic and most prominent variables in such interaction systems are not necessarily the significant stimuli to which the individual buyer is exposed, but rather the character of interdependence relationally shared with the other parties in the transaction. Relevant questions from a social perspective include (1) what type of interaction System the buyer and seller have established; (2) to what degree i.. the relationship characterized by different modes of social influence; and (3) what norms, rules, and meaning systems characterize the interaction?

Bonoma (1976 and 1977) has proposed both a general model and a consumer behavior adaption of three specific "power-consumption" systems in which marketing transactions occur. The basic conceptual framework of the model views marketing transactions as one subpart of the social influence process, in which a buyer and seller exert various influence attempts on one another in order to seek a resolution to the consumption transaction. As long as there are elements of competition in the interaction situation, the participants to the consumption transaction will experience varying degrees of social conflict (Chertkoff and Esser 1978). Conflict arises when two or more participants encountering one another (a) jointly desire a resource which is in too-short supply to satisfy all concerned, or (b) when the accomplishment of the behavioral preferences of one party involved in tile interaction would prohibit the accomplishment of another party's behavioral preferences. The exercise of social influence is the result of the conflict existing between the parties involved in the marketing interaction.

The determinant of effective usage of various influence modes during the marketing transaction encounter, as well as the basic type of interaction which occurs, lies in the relative power of the participants involved in the transaction (Cook and Emerson 1978). Thus, because of the nature of the power disparities possible between buyer and seller (a relational concept, which itself is dependent upon other relational constructs like interdependence and conflict), specific and discriminable "systems" of buyer-seller behavior will be isolatable. These include and go beyond the traditional conception of consumer interactions.

When the three logical possibilities of buyer-seller power levels (Bonoma 1976 and 1977) are articulated, i.e., (1) when the buyer is much weaker (or stronger) than the seller, (2) when the buyer and seller approach equivalence in their relative power as regards the consumption transaction understudy, or (3) when power issues are irrelevant to the relationship of buyer and seller; three distinct power-consumption systems emerge. These power-consumption systems can be labeled the "disparity" system, the "functional equivalence" (bargaining) system, and the "group welfare" system of marketing transactions. Though a good deal of the marketing literate has been given over to articulating the interaction rules and results of at least two of these systems (e.g., disparity most of the consumer behavior literature falls in this interaction system, and bargaining--Allen et al. 1975; Anglemar and Stern 1978; Kotler 1972; Mathews et al. 1972; McCarty 1975; Dwyer and Walker 1981), no previous investigation has postulated nor attempted to explore the contextual links between the three power consumption systems.

The research reported in this paper first attempts to describe the three power-consumption systems' characteristics. A field experiment is then reported which examines conditions under which the functional equivalence system's boundaries are tested. Finally, other studies are proposed to further test the power-consumption systems model of consumer behavior.

The Disparity System: The defining characteristic of the disparity power-consumption system is that the seller or producer of goods is ordinarily much stronger than the buyer (or aggregate of buyers) involved in the transaction. Examples include the retail buying situation, as we . as any other high demand market (e.g., gasoline during the OPEC boycott). Consider the following phenomena: (a) the seller requires the buyer to come to the seller's outlet; (b) the seller sets the conditions, constraints, and requirements of purchase (e.g., store hours); (c) the seller sets a price on the goods under consideration, and the buyer may "take it or leave it" (e.g., the administered pricing of department stores).

These exemplary effects of disparity transactions serve to bring home the point that there is a reliable power difference between the consumer and the seller in the latter's favor. The left-most column of Table 1 summarizes our current a priori sense of this system, in which a great majority of marketing analyses of consumer and industrial buying behavior have been positioned.

First, the interaction structure is a stimulus-response one in which the buyer is bombarded by the "4 R's" of product, price, promotion, and place, and makes some consumption decision. The interaction is construed as basically competitive, where the seller is motivated to maximize profit and the buyer to satisfy his/her needs for quality products at minimum prices. It is understood that the seller will try to persuade, cajole, or even bribe (e. g., coupons , in store sales promotions) or threaten ("If you don' t brush your teeth with. . . ") the consumer to make a favorable purchase response. The ratification of such a sales transaction is ordinarily achieved by physical exchange of goods for money; the legal system of threats is employed to ensure both parties abide by the rules. Consequently, the disparity interaction system breeds buyer mistrust, consumerism, and all the other phenomena we are used to observing in the disparity interaction; shoplifting and store vandalism, for instance. No one buyer, and seldom any aggregate of buyers, is strong enough to be "important" to the retail seller in the sense that the seller would change his "production orientation" of profit volume maximization toward a "marketing orientation" of consumer satisfaction (Krech et al . 1962 ) .

The Functional Equivalence System: In the "functional equivalence" transaction system, the buyer and seller approach equality as regards resource control vis a vis the transaction under study. Examples include "big ticket" sales, such as house or auto buying, where each individual buyer represents such a significant portion of revenue to the seller that he or she is considered "irreplaceable" in a manner not present in the disparity system. The structure of buyer-seller transactions in the functional equivalence system is not stimulus-response, but rather a much more dynamic search for mutually satisfying utility realignment points, points at which some commodity held by the seller (e. g., a house) may be traded for some number of units of money held by the buyer. The success criteria of the parties is the favorability of the exchange rate to them; e.g., how much money was exchanged for how valuable a house. The interaction system breeds a "marketing orientation" in the seller, a desire to "satisfy consumer preferences," just because the individual buyer is so important. The system inculcates a bargaining orientation in the buyer as well as desire to treat the seller's initial stance toward the product as tentative "offers," negotiable and malleable .

The same sorts of influence moves occurring in the disparity system are available for use in the functional equivalence system. but are accessible by both buyer and seller, rather than only one party to the transaction. The ratification system in bargaining is not physical exchange, but the much more sophisticated mechanism of contract formation as a particular instance of norm maintenance. In short, because the buyer and seller are equal and important to each other in the functional equivalence system, bargaining is the preferred interaction mote as opposed to a "harder" process of influence imposition as in the disparity system (see Chertkoff and Esser 1976 for a review of experiments in explicitly bargaining).



The Group Welfare System: In the group welfare system, a number of sellers (cartel), buyers (family), or buyers and sellers (co-op) form a group. The maintenance of the group and its welfare is seen as the primary objective, and in an important sense, mileposts in the longer continuum of relationship. Arndt (1979) has recently further discussed these systems, which he calls "domesticated" marketing systems. The orientation in this system, therefore, is a problem-solving one. The parties' goals are group welfare maximization, and a number of unique behaviors appear in this system which are not seen in the disparity or functional equivalence systems. However, the group welfare system was not investigated in this present study, and therefore, we refer the reader to Bonoma and Zaltman (1979) for an in-depth conceptual discussion.

The three power-consumption systems are schematically summarized in Table 1.


To do a field experiment, or survey, and establish that more bargaining type interactions take place in a car dealership, for instance, where buyer and seller are more equivalent in power than in a supermarket, where any one buyer is clearly weaker than the seller, would be a trivial exercise. But, to attempt to break down such normal interaction patterns in disparity or functional equivalence transaction contexts has greater theoretical value, since by studying the boundaries of the interaction structures, we can gain insight into the essential conditions for the particular structure observed.

Hypotheses: It can be hypothesized that a critical differentiating condition between a disparity and functional equivalence interaction, when the fuzzy concept of "relative power" is pinned down, has to do with both the genera] market conditions and the personal characteristics of the buyer. This research examines a manipulation of the personal characteristics of the buyer from a low power/high power perspective, keeping the general market conditions relatively constant

Buyer characteristics contribute to the consumer's possession of power vis a vis the seller. A well dressed, cash-paying (not requiring financing), slightly older, working man or woman may be perceived to have more power (greater resource control) than will, say, an informally dressed (jeans and sweatshirt), cash-poor, slightly younger man or woman. The former will find it easier to bargain in marKeting interactions, while the latter will find themselves in a disparity interaction more often. Both dress and wealth have been used as objective measures of social class and power in other studies (Krech, Crutchfield, and Ballachey 1962). Power theory research by Cook and Emerson (1978) established that an asymmetrical distribution of power amongst transaction participants will produce asymmetrical negotiations, with the more powerful member controlling the bargaining. A bargainer who holds more power commands a more extreme initial bid and yields less profit from the initial position. When a bargainer is less powerful, he will take a less extreme initial position and yield more profit (Thibaut 1950).

Thus, the following hypotheses are suggested for testing: Hypothesis 1: A well-dressed individual will receive better initial offers in car bargaining interactions than will a more casually dressed individual. Hypothesis 2: An individual with sufficient financial resources ("cash available") to purchase an automobile will receive better initial offers in car bargaining interactions than will an individual with scarce financial resources, requiring financing.

More specifically, a higher power consumer will (a) more often find herself in a functional equivalence system, and (D) the outcome for the high power consumer will be better (more favorable exchange ratio) than a low power consumer.

Relevant research which comments on these hypotheses presents a set of results which might be taken to offer preliminary confirmation. Specifically, Wise, Cox, and Floto (1977) report a study whose goal was to "determine the e:tent to which a new car salesman would allow his perception of the sex and the race of a prospect to affect either the price quoted or the services rendered to a prospect (p.108)." Their methodology had a well dressed, middle class, black or white male or female enter twenty new car dealers in the Dayton, Ohio area. The confederate, of whatever race or sex, established contact with the proper salesman, and then asked the salesman what price he or she would have to pay for a car with certain specified equipment. While a number of behavioral observation measures were taken by Wise et al., a critical dimension concerned the price offer made to a new car buyer. Basically, Wise et al. found that black prospects, and especially black females, were asked to pay significantly higher amounts for an identical new car than white prospects. There were no reliable differences between white males and females. These results, if race is an indicator of potential social power, suggest the power-consumption systems theory may have some consumer validity. The following study was designed to test this theory in a less ambiguous manner. [A pilot study was conducted in which twenty car bargaining interactions were conducted. The confederate appeared as either casually dressed and requiring maximum financing, or well dressed and having the ability to pay cash. The result indicated (p<.10) that the high power individual received more and better offers on the cars in question. It was decided to treat tile two power manipulations separately.]

The Study: A well-coached female confederate [The confederate was coached on appropriate dialogue and interaction style with the car salesmen. The confederate was instructed to note whether the salesmen made an initial offer lower than the sticker price, and the difference between the sticker price and the initial seller's offer. At no time were the hypotheses or the underlying theory of this research revealed to the confederate.] of the experimenters' entered sixty car dealerships selected randomly from the Columbus, Ohio SMSA in one of four power manipulation conditions.

Perceptions of high or low power were manipulated using exactly the same person in a manner which has been reported previously in social psychological experiments. Basically, the confederate dressed differently in the high versus lower power dress condition. In the former, she was dressed in a fur coat, with a "dressy" dress, high heel shoes, and carefully coiffured hair. This dress style was referred to as the "working woman" mode.

Contrarily, in the lower power dress manipulation, the exact same female was presented to salesmen in a cloth jacket, dressed in jeans, wearing "earth shoes," and with self-styled and not too carefully done hair. This style of dress was labelled "college student."

The financial ability to purchase an automobile was also systematically varied with the two dress conditions of the confederate. In the high power financial condition, the confederate presented herself as having full ability to pay on the spot for the car, by "letting it slip" that she had a savings account with sufficient funds to purchase the car should she decide to do so, and had no need to finance it.

In the low power financial condition, the confederate made it known that she was looking for a car to purchase with minimum down-payment, and desired financing of the vehicle as a necessity. A summary description of the dress and financial manipulations is given in Table 2.



The car-buying context in America is one that normally conforms to the requirements of a functional equivalence system (bargaining) interaction. For the purposes of this research, only mid-sized (price range $5,200 through $8,300 with mean price of $6,547), domestic automobiles, across the spectrum of manufacturers were included in the experiment. The entire series of sixty interviews was conducted between mid-September and mid-December, 1979.

In each interview, the confederate followed the same exact procedure with the exception of dress and the ability to pay. The confederate entered the dealership and began to look over the on-floor samples searching for a mid-sized automobile, and waiting to be approached by a sales person.

When so approached, the confederate would lead the salesman (all 60 sales representatives were male) to an appropriate automobile, and after glancing at the price list window sticker, would state:

"I really like this car. I see that is costs $XXXX.XX. Can you tell me how much more the total cost would be with tax and title charges? I have cash to pay for the car in a savings account (I will have to finance the car, and can only afford a small down payment)."

This was an implicit offer to pay the full sticker price as if in a disparity situation with a take-it-or-leave-it administered price.

The major dependent variables observed were (a) whether or not the salesman refused to allow the "normal" interaction pattern to be violated (i.e., whether he did not accePt the implicit offer to pay list price) and moved to suggest some lesser sum, and (b) if the salesman did turn to the functional equivalence system of interaction, his initial offer to reduce list Price as a percentage of the list Price.


A 2 (dress condition) by 2 (payment ability) ANOVA was conducted to test for initial differences in the sticker prices of the cars by cells of the experiment. No significant main effects or interactions were noted. The percentage reductions from list price were also analyzed by analysis of variance techniques. A 2 (tress condition) by 2 (payment ability) ANOVA was used to examine the outcome of the sixty car bargaining interactions. The results are presented in Table 3 and Figure 1.





A main effect is apparent for the dress condition manipulation. The "working woman" received greater price reductions than did the "college student." A main effect is also evident for the confederate'; payment ability. When the confederate expressed the ability to pay cash, she received greater price reductions than when she expressed a need for maximum financing with a minimum down payment. Thus, both hypotheses were supported.

It was also predicted that the high power (well-dressed and cash-buying) conditions would generate the fewest refusals to bargain on the part of salespeople as well as the greatest amount of price reduction from the sticker price. This prediction was not significantly substantiated. Figure 2 indicates the number of interactions by treatment condition in which salespeople accepted the offer to pay full sticker price without attempting to make a counter offer. Although the prediction is not significant using a chi-square test, two elements are worth mentioning: (1) in the overwhelming majority of the cases (50 of 60), in all power manipulation conditions, the salesperson rejected the implicit offer to pay full sticker price. This indicates the difficulty of violating the "normal" interaction mode for a particular purchase. And (2) the high dress, cash payment condition had no violations of the functional equivalence system, while the other conditions did display such violations, three on average. This indicates that, while the vast majority of interactions maintained the bargaining nature of the interaction, this was especially true in the highest power condition.




The implications of this research are important to the understanding of consumer behavior because they provide evidence in support of the fact that marketing transactions are not independent of their context. It is essential to understand the necessary and sufficient conditions under which "normal" interaction patterns can be violated in order to understand something about the "normality" of marketing transactions that occur. For instance, with what classes of buyer (a market segmentation question) will a disparity system interaction be dysfunctional? When is bargaining an undesirable interaction system, unnecessary and unsuited to t he resources of the buyer?

The results of this preliminary study provide some information to help examine the boundary between the disparity and functional equivalence systems. Buyer naivety in itself (the offer to pay the list price of an automobile) is not enough to move a consumer from a bargaining interaction system to a take-it-or-leave-it one as indicated by the fact that 50 out of 60 salespeople refused an initial offer to pay full sticker price for an automobile. Naivety coupled with less power and/or a requirement for favorable treatment from other parties (i.e., financing) does begin to disadvantage the consumer to a significant extent as indicated by the differences in percentage of price reductions offered.

An important question raised by this experiment is whether the power manipulations were truly the cause of the differences in the bargaining outcomes. It could be argued that in certain market conditions, offering credit decreases the profitability of a sale. Thus, a seller might be forced to give lower discounts to buyers seeking financing to keep profitability constant. In these conditions, cost, rather than market power conditions, could be driving the price differences between consumers. In informal discussions with a subset of the car salesmen involved in this research, it was learned that their commission was not affected by the financing aspect of the sale. In fact, there was some incentive to encourage the buyer to finance the purchase if the financing could be arranged with the bar.k holding the car dealership's floor plan. With respect to dress, the change from the low power condition to the high power one could also have affected attractiveness. Dion and Stein (1978) have found physical attractiveness to increase interpersonal influence; however, French and Raven (1959) have classified this type of influence as referent power.

For a long time in marketing, and in business and society courses, we have asked the question: "Do the ___ pay more?" The instructor is supposed to fill in the blank with his favorite preference, such as the poor, blacks, the elderly, or whatever. Perhaps it is not just the "___" who pay more, but the power context in general determines what kind of marketing interaction system will be elicited and how well or poorly certain consumers will do in those transactions they encounter. Being poor, or old, or black is just another way of being weak. So is being young, or uninformed, or placed in an emergency situation for which one is unprepared. The "power context" view of marketing we offer here integrates all of these disparate marketing investigation variables and provides a coherent context in which they may be interpreted, theoretically related, and hopefully understood.

This research suggests many avenues for future empirical investigation, including testing the same purchase situation for differences in the sex of the buyer (male/female). Another study currently in process is examining the boundary of the disparity system of interaction.

That study is conceptually similar to this study, but tries to violate the non-bargaining disparity nature of retail sales toward a functional equivalence (bargaining) interaction. A confederate high dress/low dress and cash/credit manipulations will attempt to buy either a floor sample (low demand condition) or "fresh in the box" item (high demand condition) of a major appliance, such as a color TV. The major dependent variable will be the amount of price reduction and frequency of interaction structure violations. It is anticipated that the low demand (floor sample), high buyer resources condition will promote the most willingness on the part of salespersons to attempt violation of the disparity norms surrounding retail sales.


Allen, Bruch E., Kahler, Ruel C., and Tatham, Ronald L. (1975), "A Shopping Simulation Game Approach to the Study of Bargaining Behavior," in Edward M. Mazze, ed., Combined Proceedings. Chicago: American Marketing Association, 439-444.

Angelmar, Reinhard, and Stern, Louis W. (1978), "Development of a Content Analytic System for Analysis of Bargaining Communication in Marketing," Journal of Marketing Research, 15, 93-102.

Arndt, Johan (1979), "Toward a Concept of Domesticated Markets," Journal of Marketing, 43, 69-75.

Bagozzi, Richard P. (1975), "Marketing As Exchange," Journal of Marketing, 39, 32-39.

Bonoma, Thomas V. (1977), "Toward a Social Analysis of Consumption: Buyer-Seller Transactions in Context," in Peter Bennett, Arch Woodside, and Jagdish Sheth (eds.), Foundations of Consumer and Industrial Buying Behavior. New York: Elsevier.

Bonoma, Thomas V. (1976), "Conflict, Cooperation, and Trust in Three Power Systems," Behavioral Science, 21 (February), 495-514.

Bonoma, Thomas V., and Zaltman, Gerald (1979), "Toward a Social Analysis of Consumer Behavior," in Gerald Zaltman and Melanie Wallendorf, Consumer Behavior. New York: Wiley Interscience.

Chertkoff, Jerome M., and Esser, James K. (1978), "A Review of Experiments in Explicit Bargaining," Journal of Experimental Social Psychology, 12 (Fall), 464-486.

Cook, Karen S., and Emerson, Richard M. (1978), "Power, Equity, and Commitment in Exchange Networks," American Sociological Review, 43 (October), 721-739.

Deutsch, Morton (1973), The Resolution of Conflict. New Haven: Yale University Press.

Dion, Karen K., and Stein, Steven (1978), "Physical Attractiveness and Interpersonal Influence," Journal of Experimental Social Psychology, 14, 97-108.

Dwyer, F. Robert and Walker, Orville C., Jr. (1981), "Bargaining in an Asymmetrical Power Structure," Journal of Marketing, 45 (Winter), 104-115.

French, J.R.P., Jr., and Raven, B. (1959), "The Bases of Social Power," in Studies in Social Power, D. Cartwright (ed.). Ann Arbor, MI: Institute for Social Research.

Goffman, E. (1967), "The Nature of Deference and Demeanor," in E. Goffman, Interaction Ritual: Essays on Face-to-Face Behavior. New York: Doubleday/Achor, 47-96.

Johnston, Wesley J. (1978), 'Marketing: The Social Psychology of Conflict, Power, and Influence," American Behavioral Scientist, 21 (March/April), 515-534.

Kotler, Philip (1972), "A Generic Concept of Marketing," Journal of Marketing, 36 (April), 46-54.

Krech, D., Crutchfield, R., and Ballachey, E. (1962), Individual in Society. New York: McGraw-Hill.

Mathews, H. Lee, Wilson, David T., and Monoky, John F. (1972), "Bargaining Behavior in a Buyer-Seller Dyad," Journal of Marketing Research, 9 (February), 103-105.

McCarthy, E. Jerome (1975), Basic Marketing. Homewood, IL: Irwin, 5th Edition.

Thibaut, J.W. (1950), "An Experimental Study of the Cohesiveness of Underprivileged Groups," Human Relations, 3 (April), 251-278.

Walker, Orville C., Jr. (1971), "The Effects of Learning on Bargaining Behavior," in F.C. Allvine, ed ., Combined Proceedings. Chicago: American Marketing Association.

Wise, Gordon L., Cox, Myron K., and Floto, Charles (1977), "Sex and Race Discrimination in the New-Car Showroom: A Fact or a Myth?" Journal of Consumer Affairs, 11 (Winter), 107-113.



Wesley J. Johnston, The Ohio State University
Thomas V. Bonoma, Harvard University


NA - Advances in Consumer Research Volume 11 | 1984

Share Proceeding

Featured papers

See More


Tuition Myopia: Temporal Discounting Induces a Myopic Focus on the Costs of Higher Education

Haewon Yoon, Indiana University, USA
Yang Yang, University of Florida, USA
Carey K. Morewedge, Boston University, USA

Read More


J1. The Effect of Identity Abstractness on Information Processing Styles

Woojin Choi, University of Seoul
Min Jung Kim, Manhattan College
HyukJin Kwon, University of Seoul
Jiyun Kang, Texas State University

Read More


Restrained by Resources: The Joint Effect of Scarcity Cues and Childhood Socioeconomic Status (SES) on Consumer Preference for Feasibility

Lili Wang, Zhejiang University
Yanfen You, New Mexico State University, USA
Chun-Ming Yang, Ming Chuan University, Taiwan

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.