رویکرد رفتار مصرف کننده به مدل سازی رقابت انحصاری

In this paper, we attempt to integrate research on consumer information processing and the consumer choice process with the goal of proposing a general framework for modeling consumer behavior in monopolistically competitive industries. Following a pattern of inductive reasoning, we posit a set of consumer behavior propositions that is consistent with observed results from context effects experiments and the phased decision-making literature. We propose that, faced with many competing brands in a monopolistically competitive environment, consumers can be said to construct consideration sets on the basis of non-compensatory rules and subsequently to choose from among competing brands within a consideration set on the basis of compensatory rules. We identify five product–market characteristics that consumers use as heuristics to maximize the probability of making the optimal brand choice while minimizing the cost of acquiring and processing information about competing brands. We propose that consumers use memory and stimuli based information to evolve their unique perceptions of these product–market characteristics. As a follow-up to our inductive approach, we show that the empirically documented context effects are consistent with our behavioral propositions. Finally, we use the propositions to explain several classic cases of consumer behavior observed in the beer, ice cream, and automobile industries.

Within the psychology and consumer behavior literatures, there exists a wealth of experimental data focusing on consumer choice in the presence of “brands” (i.e., heterogeneous variations on a product). Much of the experimental findings, including the attraction effect, the substitution effect, the compromise effect, the lone-alternative effect, and the polarization effect, are collectively known as context effects. To date, the literature on context effects and the consumer choice process remains largely empirical in nature. Lacking is a theoretical foundation that attempts to explain, in terms of consumer behavior, why context effects exist. In concluding their seminal work on the attraction and substitution effects, Huber and Puto (1983, p. 41) remark that context effects “could be specified as a part of a general framework”. Simonson (1989, p. 159) similarly challenged researchers. While the marketing literature has devoted much effort to detecting context effects, the mainstream economic literature has remained relatively silent on the topic. Our hope is to create a bridge between these two disciplines by offering a theoretical framework that attempts to explain the empirical observations. We believe that evidence from the context effects and consumer choice experiments suggests a compelling framework within which economists can create rich models of consumer behavior in monopolistic competition and marketing researchers can refine their experiments in context effects. In this paper we develop a general framework that is consistent with the phased decision-making literature and empirical evidence from published context effects experiment. In developing the general framework, we suggest a possible integration among three disparate streams of research: consumer information processing, context effects, and the consumer choice process. Our general framework rests on four important premises. 1. In a monopolistically competitive environment, consumers may be uncertain about the attributes of some or all of the brands, and may be unaware (and cognizant of their unawareness) of the existence of some brands. In an attempt to maximize the likelihood of making the best purchase decision while minimizing the cost of acquiring and processing information necessary for that decision, consumers can rely on heuristics (cf., Bettman, 1979, Bettman et al., 1991, Biehal and Chakravarti, 1986 and Nedungadi, 1990). 2. Consumer choice involves sequential stages that result from consumers’ attempts to resolve uncertainty. This notion of phased decision-making is well documented (cf., Bettman, 1979, Biehal and Chakravarti, 1986 and Kardes et al., 1993). 3. Consumers rely on available information (which maybe be imperfect, incomplete, and/or obsolete) to construct their perceptions of product–markets (the set of all competing brands known to the consumer juxtaposed according to the perceived similarity of salient attributes and the anticipated impact of those attributes on the consumer’s utility; for a review of this research stream see Alba et al., 1991 and Kardes, 1994). 4. Consumers’ perceptions of product–markets influence their subsequent consideration and choice (cf., Day et al., 1979 and Ratneshwar and Shocker, 1991). We identify five salient characteristics of product–markets that consumers can use as choice heuristics: cluster size, cluster variance, cluster frontier, granularity, and brand variance. These product–market characteristics provide information for a non-compensatory short-listing of brands and subsequent compensatory comparison of short-listed brands.1 We organize the remainder of the paper as follows. In the next section, we explain the need for the framework, review pertinent literature, and present the core propositions. Next, we show how our framework is consistent with observed context effects and demonstrate strategic applications of the framework. Finally, we offer concluding remarks on the implications of our framework for future research.

In this paper, we draw on more than twenty years of theoretical and empirical literature on consumer behavior and context effects to suggest a theoretical framework for modeling strategic behavior in monopolistically competitive industries. We define five characteristics of product–markets that can serve as heuristics in the choice process, and show that these heuristics impact the consumer’s probabilities of consideration and choice-given-consideration for brands in the product–market. Finally, we postulate that consumers will use experiences garnered from their consumption decisions to attenuate external and internal uncertainties, thereby causing their perceived product–markets to evolve iteratively. The framework opens the door to a broader application of context effects. For example, there is much economic literature on voter preferences that would benefit from application of context effects in which political candidates are seen as competing brands with different attributes (cf., Caprara & Zimbardo, 1997). A possible counter-intuitive result from this research (along the lines of the attraction effect) would be that some candidates could benefit more by donating to a competitor’s campaign than by increasing their own campaign efforts. For example, when a candidate is well-known, contributing a fixed amount of dollars to the campaign of a lesser known, though similar (i.e., within the same cluster) competitor, could increase the probability of consideration for the candidate by more than would spending those same dollars on the candidate’s own campaign. The argument here is that the elasticity of the probability of consideration (i.e., the relative change in the probability of consideration given a change in campaign spending) for the candidate is greater than the elasticity of the probability of choice-given-consideration. Another possibly counter-intuitive result from this research (along the lines of tradeoff contrast) is that some candidates could benefit from being specific about their positions on topics while others might benefit from being deliberately unclear because one candidate’s lack of clarity might push the cluster frontier farther away from all candidates such that, while the probability of choice-given-consideration falls for all, the probability might increase for some relative to others. Another intriguing application is in the realm of stock price analysis where stocks can be viewed as brands with salient attributes of price, risk, and various financial measures. Interestingly, the dot-com market bubble can be modeled as an increase in the number of brands in the technology cluster. Our model predicts that an increase in the number of tech stocks would increase the probability of consideration for all stocks in the tech cluster. The framework provides a background for further research into the implications of risk-preference on consumer choice. Consumers who are risk preferential may exhibit behavior that violates some of the framework’s propositions. For example, a risk preferential consumer may exhibit a greater probability of consideration for smaller rather than larger clusters. Also, a risk averse consumer may exhibit a greater probability of choice-given-consideration for brands with low variance even if they are positioned further away from the cluster frontier than other brands with higher variances. Lastly, the framework and referenced empirical work suggest that some utility models would benefit from explicitly accounting for uncertainties with regard to product attributes (i.e., brand variance) and uncertainties with regard to mental clustering of brands (i.e., cluster variance). In the presence of non-extreme consumer involvement and many competing brands, treating utility as determinate may lead to incorrect conclusions as uncertainty and the consumer’s response to uncertainty, vis-à-vis the consideration-choice process, play significant roles. Further, the framework suggests that utility maximization might be considered as an on-going, stochastic, adaptive process in which consumers use evidence gleaned from past consumption to mitigate uncertainty and refine heuristics for use in future consumption decisions.