Interpreting Unilateral and Competitive Price Signals: the Moderating Role of Need For Cognition

ABSTRACT - Retailers use unilateral promotions (e.g., x% off or $x off) as well as competitive promotions (price-matching offers, e.g., bring a lower advertised price and we will refund the difference) to attract shoppers. We report an experiment that compares subjects’ preferences for the two types of promotion in the context of a camera-purchase and car-repair decision. The results suggest that the attractiveness of price-matching offers is moderated by the amount of thought subjects bring to the decision (need for cognition; nCog). Low nCog subjects prefer price-matching offers and associate such signals with low market prices as long as competing stores do not run any promotions. When competing stores announce unilateral discounts (e.g., 33% off), low nCog subjects prefer direct discounts to price-matching offers. High nCog subjects, on the other hand, are indifferent between the two types of promotion. Implications for retail strategy and consumer decision-making are discussed.

Citation:

Subimal Chatterjee and Suman Basuroy (1998) ,"Interpreting Unilateral and Competitive Price Signals: the Moderating Role of Need For Cognition", in NA - Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT : Association for Consumer Research, Pages: 151-155.

Advances in Consumer Research Volume 25, 1998      Pages 151-155

INTERPRETING UNILATERAL AND COMPETITIVE PRICE SIGNALS: THE MODERATING ROLE OF NEED FOR COGNITION

Subimal Chatterjee, Binghamton University

Suman Basuroy, Rutgers University

ABSTRACT -

Retailers use unilateral promotions (e.g., x% off or $x off) as well as competitive promotions (price-matching offers, e.g., bring a lower advertised price and we will refund the difference) to attract shoppers. We report an experiment that compares subjects’ preferences for the two types of promotion in the context of a camera-purchase and car-repair decision. The results suggest that the attractiveness of price-matching offers is moderated by the amount of thought subjects bring to the decision (need for cognition; n_Cog). Low n_Cog subjects prefer price-matching offers and associate such signals with low market prices as long as competing stores do not run any promotions. When competing stores announce unilateral discounts (e.g., 33% off), low n_Cog subjects prefer direct discounts to price-matching offers. High n_Cog subjects, on the other hand, are indifferent between the two types of promotion. Implications for retail strategy and consumer decision-making are discussed.

Retailers use a wide variety of promotional offers to attract store traffic, ranging from in-store coupons and bonus buys (Dhar and Hoch, 1996) product displays, feature advertisements, and price discounts (Green, 1995) and free extra products (Diamond, 1992). Such unilateral promotions prominently display the benefits to the consumer (e.g., reduced price, additional merchandise at no extra cost). Of a different nature are competitive promotions such as offers to match price (e.g., "if you see a lower advertised price for the same brand name item, we’ll match that price at the time of purchase") made famous in retail pledges like Nobody Beats the Whiz, or Nobody Beats Midas. Unlike unilateral promotions, competitive promotions do not directly advertise a price cut, but only promise to match a lower price should the shopper find one.

Little attention has been devoted to studying how shoppers infer market prices from price-matching promotions, and how effective such signals are in attracting store traffic (see for exceptions, Chatterjee and Basuroy, 1997a, 1997b). In general, understanding how consumers respond to a signal is important to the success of the signaling strategy (Moore, 1992). Price-matching offers, in particular, are an important avenue for investigation since they can send conflicting messages to shoppers. Shoppers may associate price-matching offers with competition and low market prices, preferring to shop in stores that offer to match the lowest market price compared to stores that do not make such offers. Conversely, shoppers may associate price-matching offers with tacit collusion among sellers and high market prices, preferring not to shop in stores offering to match the lowest market price.

How consumers interpret price-matching offers will depend, among other things, upon how much thought they bring to their decision. For example, shoppers may want to simplify their store choice and use the signal as a decision heuristic (e.g., choose the store that promises to match the lowest market price). Alternatively, the absence of any price information (e.g., current prices, amount of reduction) may make shoppers suspicious and encourage them to shop for the lowest price. If shoppers think about the signal, they may in fact, as shown shortly, be convinced that price-matching offers are a tool for tacit collusion among sellers and high market prices. In this paper we report an experiment that (1) compares the attractiveness of price-matching offers against direct price discounts, (2) determines whether subjects associate high prices or low prices with stores offering to match their competitors’ price, and (3) examines whether the amount of thought that subjects bring to the task moderates their judgment of price-matching offers.

THEORY AND RESEARCH

Price Matching: Signaling Collusion or Competition?

Sellers can use price-matching offers as a strategic tool in various ways (see Levy, 1994; Hess and Gerstner, 1991 for reviews). First, sellers may announce price-matching offers without actually lowering prices in the hope that customers will not expend the effort required to take advantage of such offers (e.g., make price comparisons across stores). Second, and according to the competition interpretation, stores desiring to build market share at the expense of profits may keep their prices low and use price-matching offers to convince customers that they will find the lowest price at the stores. Third, and according to the collusion theory, stores can use price-matching offers to enter into tacit collusion with one another. This happens because price-matching serves as a deterrent to price cuts on two grounds. First, if all sellers offer to match each other’s price, then there is little to be gained by any one seller reducing his/her price since shoppers can get the benefit of the low price at any store. Second, the seller contemplating the price cut knows that other competitors are likely to retaliate by initiating cuts of their own, thereby reduing industry profits (a tit for tat strategy; Cooper, 1986; Belton, 1987).

Taking these arguments one step further, some authors have suggested that price-matching offers can support a collusive price, encouraging sellers to raise prices to levels that maximize joint profits (Salop, 1986; Lin, 1988; Sargent 1993; Baye and Kovenock, 1994; Chen, 1995; Zhang,1995). The rational is explained with the help of an example. Suppose Stores A and B each sell the same product for $10. If Store A increases the price to $12 and announces that it will match Store B’s price, then Store A should not lose any customers to Store B. Shoppers who know about Store B’s price will get the same price ($10) at Store A, whereas the less-knowledgeable shopper will pay the extra $2 (Png and Hirshleifer, 1987). [This does not take into consideration the fact that some buyers, irritated or suspicious by the fact that Stores A's list price is higher than that of Store B, may prefer Store B.] The most profitable counter-strategy for Store B is to do the same as Store A, i.e., raise its price to $12 and announce that it, too, will match Store A’s price (see Salop (1986) for a game-theoretic model).

Empirical data, the little that exist, appear to support the collusion theory. Hess and Gerstner (1991) collected weekly price data for 114 frequently purchased products from five supermarkets in North Carolina between 1984 and 1986. Out of these 114 products, 79 were covered under price-matching guarantees and their prices published in a weekly Price Finder. The other 35 products not included in the Price Finder, due to fluctuations of wholesale prices or legal issues, were not guaranteed to match the lowest market price. The authors found that (1) price-matching offers led to a high degree of price coordination among all supermarkets (those making, as well as those not making, the offers), and (2) over time, there was a significant increase in prices of the products covered by price-matching guarantees relative to the products that were not.

The Moderating Role of Need for Cognition

If shoppers bring little thought to their store-choice decision and are looking for cues to simplify the task, price-matching offers may serve as a convenient decision heuristic (e.g., choose the store that advertises to match the lowest market price). To associate price-matching with collusion however, requires greater deliberation. At the least, the shopper has to mentally simulate different pricing scenarios and analyze how best price-matching offers can be used to the seller’s advantage. Hence, the amount of thought that shoppers bring to the decision should determine the attractiveness of price-matching offers. The dispositional variable, need for cognition (n_Cog) refers to an individual’s proclivity to engage in and enjoy effortful thinking (Cacioppo & Petty, 1982). High n_Cog subjects are likely to think long and hard about any issues, whereas low n_Cog individuals’ tendency is to look for cues or reasons to simplify their decision. We investigate the moderating role of n_Cog on subjects’ reactions to price-matching offers for four reasons.

First, low n_Cog people are typically "cognitive misers" (Cacioppo, Petty, Kao, & Rodriguez, 1986) and are unlikely to devote much thought to any situation. Hence, they are more prone to use price-matching signals as a heuristic to simplify choice. High n_Cog people, on the other hand, are more likely to consider both the competition as well as the collusion aspects of price-matching offers, and base their choice on more substantive cues (e.g., actual price).

Second, n_Cog has been found to moderate reactions to price promotion signals such as % off or limit x with offer (Inman, McAlister, and Hoyer, 1990; Inman, Peter, and Raghubir, 1997). For example, whereas high n_Cog individuals are persuaded by price reductions, low n_Cog subjects are persuaded by discount signals even when the price of the promoted brands is not actually reduced. The current study investigates if similar n_Cog effects extend to price-matching ffers as well.

Third, n_Cog has been found to moderate individuals’ susceptibility to framing effects (Smith and Levin, 1996). In one experiment Smith and Levin (1996) asked subjects to choose between two types of cancer treatments (surgery and radiation therapy) that were described either by number of people surviving (e.g., x out of n people live at the end of the first year), or number of people dying (e.g., n-x people out of n die at the end of the first year). Framing had a significant effect on the choices of low n_Cog subjects, but high n_Cog subjects were unaffected by the frames. If price- matching offers are seen as frames to describe stores in a more competitive light, low, rather than high, n_Cog subjects are more likely to be susceptible to such framing effects.

Finally and fourth, if n_Cog is to be useful in understanding consumer behavior in the actual marketplace, it is important that its hypothesized influences be studied in such settings (Haugtvedt, Petty, & Cacioppo, 1992). Promotional signals provide a particularly relevant marketing context for studying the influence of this variable.

Hypotheses

Suppose that two stores, A and B, in Mall X advertise to match each other’s price, and two other stores, C and D, in Mall Y make no such guarantees. We call this Scenario 1. If you believe in the competition theory, ceteris paribus, you are likely to choose X over Y and associate lower prices with the stores in X compared to the stores in Y. If, on the other hand you believe in the collusion theory, your preferences will be just the reverse. Associating competition and low prices with price-matching offers is relatively simple. However, associating collusion and high prices with price-matching offers require greater deliberation, an effort that low n_Cog individuals are unlikely to expend. Hence, we expect low n_Cog subjects to select X over Y, and associate X with lower prices. High n_Cog subjects, on the other hand, more likely to understand the competitive as well as the collusive aspects of price-matching offers, are expected to be indifferent between X and Y.

Now suppose that the two stores, C and D, in Mall Y, each announce a 33% discount. We call this Scenario 2. Since low n_Cog individuals are likely to make their decisions based upon market signals, a signal that advertises a magnitude (e.g., 33% off current prices) may be more persuasive than a signal that does not. On the other hand, since high n_Cog subjects are persuaded only by the dollar amount of price reductions (Inman, McAlister, and Hoyer, 1991), market signals that do not advertise the actual dollar reduction (price-matching offers, or percentage-off signals without a base price) are not likely to be persuasive. Hence,

H1a: Low n_Cog subjects will choose X more frequently in Scenario 1 than in Scenario 2.

H1b: High n_Cog subjects will be indifferent between X and Y across Scenario 1 and Scenario 2.

H2a: Low n_Cog subjects will associate lower prices with X in Scenario 1 than in Scenario 2.

H2b: High n_Cog subjects will not perceive prices to be different between X and Y across Scenario 1 and Scenario 2.

Finally, high n_Cog individuals, uncertain whether they should associate price-matching offers with more competition or more collusion, are likely to vacillate between X and Y. Low n_Cog individuals, on the other hand, are likely to use the promotion signals (pricematching offers, or discounts) to simplify their decision. Hence,

H3: High n_Cog subjects will find their decisions more difficult than low n_Cog subjects across Scenarios 1 and 2.

EXPERIMENT

Method

Stimuli and measures. Subjects were asked to imagine that they wished to buy a 35 mm. camera, or their car needed a scheduled engine tune-up. The price of the camera (or cost of the service) was estimated around $300. They were told that there were two camera stores (repair shops), A and B, in Mall (Township) X, and there were two other camera stores (repair shops), C and D, in Mall (Township) Y. Stores A and B guaranteed to match each other’s price. They did not guarantee to match Store C or D’s price. For half of the subjects, Stores C and D did not offer any discounts or matching guarantees of their own. For the other half of the subjects, Stores C and D offered a 33% discount on the camera (car-service). Subjects were told that they had time to visit only one Mall (Township), and X and Y were equi-distant from where they lived. Furthermore, to prevent subjects from drawing any quality inferences from the price cues presented, they were told that the quality of the cameras (car-repair) was identical in the two malls (townships).

Following the descriptions, subjects were asked their impression of the price at X and Y on a 9-point scale ranging from 1 (Definitely Lower in X) to 9 (Definitely Lower in Y) with 5 (About the Same in X and Y) as the mid-point. Subjects then made a choice (X, or Y), and responded to a question that ascertained the perceived difficulty of the decision on a 9-point scale ranging from 1 (Very Easy) to 9 (Very Difficult).

TABLE 1

CHOICE OF X: DISCOUNT BY n_Cog INTERACTION

TABLE 2

PRICE AT X: DISCOUNT BY n_Cog INTERACTION

One week later, as part of a different assignment, subjects answered the 18-item n_Cog scale (Cacioppo, Petty, and Kao, 1984). The items included statements like "I would prefer complex to simple problems" and "Thinking is not my idea of fun" (reverse scaled). Subjects marked each item on a 9-point scale ranging from very strong disagreement (-4) to very strong agreement (+4). A composite n_Cog score was created for each subject by averaging the items.

Subjects. Fifty-two undergraduate students enrolled in a business class at a large north-eastern university served as subjects in this experiment and were given course credits for their participation. They were randomly assigned to the four cells of the 2 (Discount offered by Y: no discount, 33% discount) X 2 (Product Class: camera, car-repair) between-subjects design.

Analyses and Results

Choice. Together, H1a (low n_Cog subjects will prefer X more when stores in Y run no discounts compared to when stores in Y run 33% discounts) and H1b (high n_Cog subjects will be indifferent between X and Y independent of whether stores in Y run discounts or not) predict a discount by n_Cog interaction on choice. To test that prediction, subjects’ choices were subjected to a log-linear analysis (SAS CATMOD) with Y’s discount (no discount, 33% discount), product class (camera, car-repair) and n_Cog as predictors. The latter variable was mean-centered and entered into the model as a continuous variable. The discount by n_Cog interaction was significant (c²₍₁₎=4.04, p<0.05; see Table 1). To explicate the interaction, we divided subjects into a low and high group based on a median split of their n_Cog scores (n’s of 26 in each goup). The average score for the low group was -0.57 (sd=0.79) and the average score for the high group was +1.20 (sd=0.62). Consistent with H1a, low n_Cog subjects preferred X more when competing stores offered no discounts compared to when competing stores in Y offered 33% discount (M_{No Discount}=68.75%, M_{33% Discount}=10%; Fisher’s exact test p<0.01). Consistent with H1b, high n_Cog subjects were indifferent between X and Y independent of whether stores in Y offered discounts or not (M_{No Discount}=40%, M_{33% Discount}=50%; Fisher’s exact test p=0.701).

Price. For the purpose of exposition, we rescaled subjects’ price ratings by subtracting 5 (the scale’s mid point) from the subject’s original price-score such that negative numbers implied that perceived prices were lower in X compared to Y, positive numbers implied that perceived prices were higher in X compared to Y, and zero implied that perceived prices were similar across X and Y. Together, H2a (low n_Cog subjects will associate X with lower prices when stores in Y run no discounts compared to when stores in Y run 33% discounts) and H2b (high n_Cog subjects will perceive no price differences between X and Y independent of whether or not stores in Y run discounts) predict a discount by n_Cog interaction on price ratings. To test that prediction, subjects’ transformed price ratings were subjected to an ANOVA (SAS GLM) with discount by Y (no discount, 33% discount), product class (camera, car-repair), and n_Cog as predictors. The discount by n_Cog interaction was significant (F_(1,44)=5.03, p<0.05; see Table 2). As predicted by H2a, low n_Cog subjects associated X with lower prices when stores in Y did not run any promotions compared to when the stores in Y ran 33% discounts (M_{No Discount}=-1.06, M_{33% Discount}=+ 0.90; t₂₄=2.67, p<0.05). As predicted by H2b, whether stores in Y offered discounts or not, did not affect high n_Cog subjects’ price ratings (M_{No Discount}=0.60, M_{33% Discount}=0.13; |t₂₄| <1)

We ran separate t-tests within each discount condition, separately for low and high n_Cog subjects (Table 2). If the transformed price score is significantly less than zero, it would imply that perceived prices are lower in X compared to Y. Similarly, if the transformed price score is significantly greater than zero, it would imply that perceived prices are lower in Y compared to X. As seen in Table 2, low n_Cog subjects associated X with significantly lower prices as long as competing stores in Y did not run any discounts. They associated Y with significantly lower prices when the stores in Y ran a 33% discount. High n_Cog subjects did not think that prices would be different across X and Y independent of whether stores in Y offered discounts or not.

Decision difficulty. H3 predicts that high n_Cog subjects will find their decisions more difficult than low n_Cog subjects. Subjecting subjects’ felt decision difficulty scores to an ANOVA with discount (no discount, 33% discount), product class (camera, car-repair) and n_Cog (high, low) as predictors, revealed a main effect of n_Cog that approached statistical significance (F_(1,44)=2.52, p=0.12). Subjects’ felt decision difficulty ratings were consistent, in direction, with the hypothesis. Low n_Cog subjects found their decisions easier than their high n_Cog counterparts (M_Low=3.69, M_High=4.77; t₅₀=1.94, p=0.06).

DISCUSSION

Price-matching offers can send conflicting signals to shoppers. They may signal collusion among sellers and high market prices, or, cnversely, they may signal competition among sellers and low market prices. Our results suggest that the inferences shoppers draw from price-matching offers depend upon (1) the nature of the competition (e.g., whether competing stores run other types of promotions), and (2) the amount of thought that shoppers bring to the decision. As long as the competing stores did not offer to match prices or offer other types of promotion, low n_Cog subjects associated price-matching signals with low price. When competing stores ran a 33% discount, low n_Cog subjects associated the discount stores with a better deal compared to the price-matching stores. High n_Cog subjects, on the other hand, did not perceive any price difference across X and Y even when the stores in Y ran a 33% discount.

If sufficient thought is devoted to the decision, it becomes apparent that price-matching offers can either foster competition or facilitate tacit collusion among sellers. Our high n_Cog subjects appear to have understood the conflicting possibilities, as reflected in their price ratings (they perceive no difference in prices across X and Y) and their choices (they were never any different from chance level, or 50%). Even when the stores in Y ran a 33% discount promotion, the high n_Cog subjects were not impressed, suggesting that they prefer to base their decisions on more substantive cues (e.g., the amount of reduction). The implication for retailers is clear. Price-matching offers, by themselves, may not attract the vigilant shopper to the store unless the offers signal the amount of price cut. However, when the shopper is hurried or distracted, or otherwise unwilling to devote much thought to the process, such signals may attract them to the stores, as long as the competing stores do not have even more persuasive signals of their own (e.g., deep discounts).

Limitations. Little prior research exists on how consumers interpret price- matching offers. Our experiment takes the first step towards understanding a phenomenon that impinges on a variety of issues ranging from retail strategies to consumer welfare. However, some cautions are in order. First, we focus on one durable product (camera) and one service product (car repair). Since price matching promotions are common across a wide variety of markets, we would need to replicate these results on a wider array of products and services. Second, in our experiments subjects are hurried for time and can select either X or Y. Future research needs to address whether the ability of price matching promotions to attract store traffic diminishes when consumers have sufficient time and resources to make price comparisons. Third, we compare price-matching offers to one particular level of discount (33%). It is possible that low n_Cog subjects may prefer price-matching offers over small discounts (e.g., 5%) and reverse their preference only when the discounts offered are large (e.g., 33%, as in our experiment). Hence, adding extra levels of discount and investigating their moderating effects is a logical next step (see for example, Chatterjee and Basuroy, 1997b). Fourth, it is possible that price-matching offers can influence the perceived quality of the products or service at that store. For example, shoppers may associate price-matching offers with low prices and associate such stores with poor-quality products or services. While we controlled for differences in quality perceptions across X and Y by instructing subjects to assume that the quality of the products were the same across the two malls (townships), it would be interesting to investigate how pricing cues such as matching-offers or discounts affect subjects’ quality perceptions. For example, recent research suggests that high n_Cog individuals associate price-matching offers with lower quality products and services (Chatterjee and Basuroy, 1997b). Fifth, moderators other than the individual’s proclivity for effortful thinking may influence subjects’ interpretations of price-matching offers. For example, collusion between two stores may be more readily established than collusion among many stores. Hence, thenumber of stores offering to match each other’s price may influence how consumers respond to such offers. Finally, our experiment investigates the standard price matching claim (i.e., matching the competitor’s price). Retailers now pledge not only to match, but to beat, the competitor’s price; similarly they offer not only to meet the advertised prices, but also the coupons offered by competing stores. How such different claims impact shoppers’ responses need to be investigated to understand the power of this very popular promotional tool.

REFERENCES

Baye, Michael R. and Dan Kovenock (1994), "How to Sell a Pickup Truck: Beat or Pay Advertisements as Facilitating Devices," International Journal of Industrial Organization, 12 (1), 21-33.

Belton, Terrence M. (1987), "A Model of Duopoly and Meeting or Beating Competition," International Journal of Industrial Organization, 5 (December), 399-417.

Cacioppo, John T., and Richard E. Petty (1982), "The Need for Cognition," Journal of Personality and Social Psychology, 42 (January), 116-131.

Cacioppo, John T., Petty, Richard E., and Chuan-Feng Kao (1984), "The Efficient Assessment of Need for Cognition," Journal of Personality Assessment, 48 (June), 306-307.

Cacioppo, John T., Petty, Richard E., Kao, Chuan-Feng, and Regina Rodriguez (1986), "Central and Peripheral Routes to Persuasion: An Individual Difference Perspective," Journal of Personality and Social Psychology, 51 (November), 1032-1043.

Chatterjee, Subimal and Suman Basuroy (1997a), "Consumer Reactions to Price Matching Signals," Advances in Consumer Research, Vol. 24, 400-404.

Chatterjee, Subimal and Suman Basuroy (1997b), "Competition or Collusion? Consumers’ Interpretations of Price-Matching Signals," working paper, School of Management, Binghamton University, Binghamton, NY 13902-6000.

Chen, Zhiqi (1995), "How Low is a Guaranteed Lowest Price?," Canadian Journal of Economics, 28 (August), 683-701.

Cooper, Thomas E. (1986) "Most Favored Customer Pricing and Tacit Collusion," Rand Journal of Economics, 17 (Autumn), 377-388.

Diamond, William D. (1992), "Just What is a 'Dollar’s Worth?’ Consumer Reactions to Price Discounts Versus Extra Product Information," Journal of Retailing, 68 (Fall), 254-270.

Dhar, Sanjay, and Stephen J. Hoch (1996), "Price-Discrimination Using In-Store Merchandising," Journal of Marketing, 60 (January), 17-30.

Green, Corliss L. (1995), "Differential Responses to Retail Sales Promotions Among African-American and Anglo-American Consumers," Journal of Retailing, 71 (Spring), 83-92.

Haugtvedt, Curtis P., Petty, Richard. E., and John T. Cacioppo (1992), "Need for Cognition and Advertising: Understanding the Role of Personality Variables in Consumer Behavior," Journal of Consumer Psychology, 1 (3), 239-260.

Hess, James D. and Eitan Gerstner (1991), "Price-matching Policies: An Empirical Case," Managerial and Decision Economics, 12 (August), 305-315.

Inman, J. Jeffrey, McAlister, Leigh, and Wayne D. Hoyer (1990), "Promotion Signal: Proxy for a Price Cut?" Journal of Consumer Research, 17 (June), 74-81.

Inman, J. Jeffrey, Peter, Anil C, and Priya Raghubir (1997), "Framing the Deal: The Role of Restrictions in Accentuating Deal Value," Journal of Consumer Research, 24 (June), 68-79.

Levy, David T. (1994), "Guaranteed Pricing in Industril Purchases," Industrial Marketing Management, 23 (October), 307-313.

Lin, Joseph Y. (1988), "Price-Matching in a Model of Equilibrium Price Dispersion," Southern Economic Journal, 55 (July), 57-69.

Moore, Marian C. (1992), "Signals and Choices in a Competitive Interaction: The Role of Moves and Messages," Management Science, 38 (April), 483-500.

Png, I. P. L. and D. Hirshleifer (1987), "Price Discrimination Through Offers to Match Price," Journal of Business, 60 (July), 365-383.

Salop, Steven C. (1986), "Practices that (Credibly) Facilitate Oligopoly Coordination," in Joseph E. Stiglitz and G. Frank Mathewson (eds.), New Developments in the Analysis of Market Structure, Cambridge, MA: MIT Press.

Sargent, Mark T. L. (1993), "Economics Upside-Down: Low Price Guarantees as Mechanisms for Facilitating Tacit Collusion," University of Pennsylvania Law Review, 141 (May), 2055-2118.

Smith, Stephen M. and Irwin P. Levin (1996), "Need for Cognition and Choice Framing Effects," Journal of Behavioral Decision Making, 9 (December), 283-290.

Zhang, John Z. (1995), "Price-matching Policy and the Principle of Minimum Differentiation, " Journal of Industrial Economics, 43 (September), 287-299.

----------------------------------------