Special Session Summary New Directions in Price Signaling Theory and Research

Joydeep Srivastava, University of California at Berkeley
[ to cite ]:
Joydeep Srivastava (1997) ,"Special Session Summary New Directions in Price Signaling Theory and Research", in NA - Advances in Consumer Research Volume 24, eds. Merrie Brucks and Deborah J. MacInnis, Provo, UT : Association for Consumer Research, Pages: 149-150.

Advances in Consumer Research Volume 24, 1997      Pages 149-150



Joydeep Srivastava, University of California at Berkeley


It has long been recognized that price plays at least two distinctive roles in consumer choice processes: an allocative role and an informative role. While the economic theory of consumer behavior considers the allocative role of price, a considerable body of literature has also treated price as an informational cue. The latter research has however been largely confined to studying the price-quality relationship. In recent years, there has been a growing awareness of the complex role that price and price-related strategies play even as an informational cue (Monroe 1990). Recent behavioral research (Monroe 1990) indicates that there is no simple explanation of how price influences individual purchase decisions and more importantly how consumers derive information from prices. The three papers in this session serve to highlight this complexity associated with buyers subjective perceptions of price. Each paper in the session examines different roles that price and price-related strategies may play in shaping consumers subjective perceptions and price expectations. The first paper by Stiving examines the effect of price endings on consumers quality perceptions. The second paper by Simester identifies conditions under which consumers use either frequency or magnitude cues as indicators of the overall price level of a store. The third paper by Jain and Srivastava shows that price-matching refunds can be used by firms to signal low prices under certain conditions. Despite the differences in the three papers, the common themes underlying the papers are: (1) motivated from an economics of information perspective, (2) game-theoretic framework, and (3) empirical testing of the consumer-side implications of the model. A brief summary of each of the three papers as well as key comments from the discussion are iven below.




Mark Stiving, Ohio State University

The first paper by Stiving addresses the effects of price endings (e.g., 24.95 versus 25.00) on quality perceptions. Many texts provide examples of "psychological pricing" such as odd-evenpricing, i.e., prices end in odd number or even number. Despite the apparent acceptance of these "psychological" phenomenon, research justifying odd-even pricing has been largely of an anecdotal nature. Stiving provides both a theoretical as well as an empirical justification for such pricing practices.

Stiving argues that when determining its pricing policy, a firm makes at least two decisions: the approximate price level and the price ending. He subsequently demonstrates that these two decisions are not independent and the relationship between the two is consistent with the signaling theory. Stiving s theoretical model demonstrates that under most circumstances a firm will want to set a price just below a round number, but when using price to signal quality the same firm would prefer to use a price that is both high and round. He shows that consumers can rationally interpret a combination of high price and 00 price ending as a signal of high quality while a round price in combination with a lower price does not signal high quality. Further, Stiving provides empirical support for his theoretical framework by demonstrating that firms appear to be sending the signal and that consumers are able to receive it.

In contrast to previous research (Dodds and Monroe 1985, Quigley and Notarantonio 1992, Schindler and Kibarian 1995), Stiving is able to demonstrate that consumers do perceive round prices as higher quality. He reported the results of an experiment which was designed around a real product offered by a firm that behaves as though they are signaling quality with price endings. The subjects in the experiment were shown a description (taken directly from the Franklin Quest catalog which sells binders for loose leaf calendars) including the price for each of two binders (vinyl and leather) and were then asked for a quality rating for each binder. This between subjects design varied the price ending between 95 and 00 for both binders, covering the possible combinations of round and just below pricing. The results indicate that consumers are able to receive the signal, perceiving high, round prices as signals of high quality relative to high prices alone. In contrast, subjects do not perceive low, round prices as signals of high quality relative to low prices alone.



Duncan Simester, University of Chicago

Simester s presentation began with a description of a series of experiments reported by Alba, Broniarczyk, Shimp, and Urbany (1994) in which they presented subjects with the prices of selected products at two stores. Although the total price for the products was the same, one store charged a lower price ontwo-thirds of the products while the other store s price was lower on the rest. The authors then examined subjects overall price perceptions of the two stores. Despite various manipulations of subjects prior beliefs, the salience of the selected products and the difficulty of aggregating the price differences, most subjects consistently indicated that they believed that the store with the lower price on most products had lower overall prices. They concluded that when subjects are unable to completely process price data, most rely on the frequency heuristic to help form price judgments.

Simestr pointed out that the findings suggest that subjects were sensitive to the number of products for which each store charged a lower price, but do not show that subjects disregarded the magnitude of the price differences. Providing a rationale for the subjects reliance on the frequency cue, he argues that for a consumer to recognize whether a price of $1.50 for a bottle of ketchup is expensive requires considerable knowledge of relative prices. For this reason, observing a price of $1.50 for ketchup reveals little about a store s attributes. However, when subjects are presented with a price comparison in which one store charges $1.50 and another store charges $1.60 for the same bottle of ketchup, it is reasonable to infer that it is more likely that the store charging the lower price enjoys lower costs on that product. Lack of information also hinders the consumer s ability to evaluate the size of a price difference: does a $0.10 price difference in ketchup represent a greater or smaller cost difference that a $0.20 price difference in cereal? Without an answer to this question, the magnitude of a price difference may reveal no more information about unseen prices than its sign; so reliance on a frequency heuristic can improve price predictions.

He argues that this intuition suggests additional hypotheses. First, if reliance on the frequency heuristic improves the accuracy of price predictions, reliance should continue when data processing is unconstrained. Second, if the information contained in the magnitude of price difference is increased, reliance on the magnitude cue should increase relative to the frequency cue. Thus the information content of the magnitude cue can be raised by reducing the product variance confounding the price differences. In particular, if the price information represents prices for the same product collected in separate months, the size of the price differences should be comparable between months. Subjects presented with monthly price differences for a single product should be expected to place more reliance on the magnitude cue and less reliance on the frequency cue relative to those who observe prices for different products. Simester reported the results of regression analysis of prices collected from various stores which provide support for his contention.



Sanjay Jain, Johannes Gutenberg University of Mainz

Joydeep Srivastava, University of California at Berkeley

Price-matching refund strategies or offers by firms to match the prices of others are very common in both consumer and industrial marketing. The final presentation by Srivastava attempted to answer the question "why do manufacturers and retailers use such price-matching policies?" Srivastava pointed out that while the previous research has justified the use of price-matching refunds either as a means of reducing competition and maintaining high prices or as a device for price discrimination leading to higher prices, casual evidence suggests that price-matching refunds are associated with lower prices. He also noted that consumers have a favorable opinion of price- matching refunds and do not perceive it as a ploy to reduce competition.

In the paper, the inconsistency between extant theory and common wisdom is resolved by developing a model in which consumers are not fully informed of prices in the market. The authors show that price-matching refunds have a different impact on competition and prices in informationally poor environments, than proposed in the prior literature. In particular, they show that in markets where consumers are not perfectly informed about prices at different stores, a price-matching refund can serve as a credible signal of low prices. Thus, such refunds can lead to more intense price competition with lower prices and higher consumer welfare.

The basic model establishes the conditions under which a price-matching refund strategy can be used by firms to signal low prices. Given that irms differ in their cost structure and that consumers are imperfectly informed about prices, price-matching refunds are credible signals of low prices when the difference in the cost structure among the firms is high. Price-matching refunds can also signal low prices when there is a high proportion of high cost firms in the market relative to low cost firms. In contrast to previous research, they find that such a pricing policy can lead to lower prices and increase consumer welfare. Subsequently, the model is extended in two different ways: (1) only a fraction of the consumers actually avail themselves of the price-matching offer (either due to high transaction costs or because they are unaware of lower priced stores); and (2) consumers have prior knowledge about the price levels at different stores. The results indicate that as the proportion of consumers who claim the price-matching offer decreases, high cost stores have more of an incentive to mimic low cost stores by offering to match prices and use it as aprice-discriminatory device. They also find that as the proportion of consumers who have prior knowledge (price expectations) about the general price levels of different firms increases, the incentive for low cost firms to signal lower prices by offering to match prices decreases .

The key implications of the analytical framework are then submitted to empirical testing. In a laboratory experiment, they empirically validate their fundamental assumption by testing whether consumers interpret price-matching refunds as signals of low prices. The results of the experiment also provide strong support for the consumer-side implications of the model.



Russell Winer, University of California at Berkeley

Winer began with a general discussion of the relevance and importance of signaling models. In particular, he pointed out that while all the three papers in the session were concerned with firm-to-consumer signaling, it is important to examine firm-to-firm, consumer-to-firm, and consumer-to-consumer signaling models as well. He noted that the strength of the three papers in the session lies in the fact that not only do these papers develop game-theoretic models but also test these models empirically. He called for more research on topics which are on the interface of economics and psychology. Winer concluded by commenting that inquiries of consumer behavior from an economics perspective serves to enrich and broaden the domain of consumer research in general.


Alba, Joseph, Susan Broniarczyk, Terence Shimp, and Joel Urbany (1994), "The Influence of Prior Beliefs, Frequency Cues, and Magnitude Cues on Consumers Perceptions of Comparative Price Data," Journal of Consumer Research, 21(2), 219-235.

Dodds, William B. and Kent B. Monroe (1985), "The Effect of Brand and Price Information on Subjective Product Evaluations" in Advances in Consumer Research, Vol. 12, 85-90.

Monroe, Kent B. (1990), Pricing: Making Profitable Decisions, Mc-Graw Hill Publishers.

Quigley, Charles J. and Elaine M. Notorantonio (1992), "An Exploratory Investigation of Perceptions of Odd and Even Pricing," working paper, Bryant College.

Schindler, Robert M. and Thomas Kibarian (1995), "Increased Salesof Discounted Items Through the Use of 99-ending Prices," working paper, Rutgers University, Camden.