Draft November 12, 2006
PROCOMPETITIVE JUSTIFICATIONS FOR EXCLUSIVE DEALING:
PREVENTING FREE-RIDING AND CREATING
UNDIVIDED DEALER LOYALTY
Andres V. Lerner*
Recent exclusive dealing antitrust case law has magnified the importance of procompetitive justifications. While the minimum market share for antitrust liability under Section 1 has increased substantially over time, making it increasingly difficult for plaintiffs to successfully challenge exclusive dealing contracts on Section 1 grounds,(1) there has been a simultaneous recent movement making it easier for plaintiffs to successfully challenge exclusive dealing contracts on Section 2 monopolization grounds when a procompetitive rationale cannot be provided for exclusivity. In these circumstances Section 2 antitrust liability may be found even when plaintiffs have not established that distribution has been effectively blocked to rivals by the exclusive dealing arrangements.
These two disparate trends in Section 1 and Section 2 exclusive dealing law are perhaps most recognizable in Microsoft.(2) The Justice Department lost on its exclusive dealing Section 1 claims at the district court, with the court holding that Microsoft's exclusive distribution contracts with Internet access providers and personal computer manufacturers did not foreclose Netscape from distributing its browser.(3) However, Justice prevailed on its Section 2 exclusive dealing monopolization claims, which the Appeals Court affirmed. Microsoft was condemned not because its exclusive browser distribution contracts effectively foreclosed the market to its rivals, but because the contracts, which controlled the "most efficient" means of browser distribution, were held to be pretextual.(4) Microsoft's exclusive contracts, therefore, unnecessarily placed rivals at a competitive disadvantage.(5) Because there was no reasonable procompetitive rationale for Microsoft's exclusivity restrictions, the court concluded that the contracts did not involve "competition on the merits."(6)
More recently, a similar result occurred in Dentsply.(7) Dentsply, a manufacturer of artificial teeth with a 75 to 80 percent market share,(8) entered exclusive dealing contracts with its dealers, who sold Dentsply teeth along with other supplies to dental labs.(9) The Department of Justice challenged the contracts on Section 1 and Section 2 grounds, maintaining that exclusivity had the effect of foreclosing rival artificial teeth manufacturers from the primary channel of distribution with no procompetitive rationale. The district court rejected Dentsply's attempt to provide procompetitive justifications for its exclusive dealing contracts as pretextual, but did not condemn the arrangements because it held that sufficient alternative distribution channels were available for rival manufacturers. Specifically, "because direct distribution is viable, non-Dentsply dealers are available, and Dentsply dealers may be converted at any time," the court concluded that Dentsply's exclusive contracts did not have an anticompetitive effect.(10)
Once again, the Department of Justice did not challenge its Section 1 loss, and only appealed the court's Section 2 ruling. The Appeals Court reversed the district court's rejection of Section 2 antitrust liability because it concluded that Dentsply effectively foreclosed the "preferred distribution channels-- in effect the 'gateways' -- to the artificial teeth market" without a valid procompetitive rationale.(11) Dentsply's exclusive contracts, therefore, placed its rivals at a competitive disadvantage without any procompetitive benefit to balance against this anticompetitive effect. While the court in Dentsply does not use the Microsoft terminology, that Dentsply's use of exclusive dealing did not involve "competition on the merits", the reasoning suggests that a firm with market power must have a non-pretextual procompetitive rationale for using an exclusive dealing contract that imposes an extra burden on competitors. Without a valid procompetitive justification for exclusivity, the balancing of procompetitive justifications and anticompetitive effects is easily tipped towards antitrust liability.
In addition to reinforcing the increasing legal importance placed on procompetitive justifications for exclusive dealing contracts, Dentsply illustrates the extremely narrow economic foundation upon which procompetitive justifications currently rest. The two procompetitive justifications for exclusive dealing offered by Dentsply, that exclusive dealing prevented dealer free-riding on manufacturer-supplied investments and created dedicated dealers that more actively promoted Dentsply products, were firmly rejected by the district court as making no economic sense, and this conclusion was fully accepted by the Appeals Court. The court emphasized that the prevention of dealer free-riding on Dentsply's investments, a widely accepted rationale for exclusive dealing presented by Howard Marvel in his now classic article and reiterated in his testimony as the economic expert retained by Dentsply, did not fit the facts of the case.(12) In particular, the court held that (a)Dentsply did not provide investments to its dealers that could be used to sell rival manufacturer's products, (b)there was no evidence of dealers switching customers (dental labs) to rival brands, and (c)contrary to the economic theory of free-riding, where exclusive dealing has the beneficial effect of increasing the manufacturer's incentive to make promotional investments, Dentsply executives testified that absent exclusive dealing Dentsply would have increased its promotional investments.(13)
The only other procompetitive justification offered by Dentsply was the claim that exclusive dealing created dedicated distributors that devoted their efforts to promoting Dentsply products.(14) While "undivided dealer loyalty" has been accepted by a number of courts as a procompetitive motivation for exclusive dealing,(15) the Dentsply court maintained that this rationale lacks an economic basis because inter-dealer competition generally provides dealers with the incentive to supply dealer services.(16) In fact, the court noted that this rationale for exclusive dealing was explicitly rejected by Marvel in his article as making no economic sense and the court fully accepted Marvel's economic analysis in rejecting Dentsply's undivided dealer loyalty rationale.(17)
The economic analysis underlying the court's rejection of Dentsply's procompetitive rationales for exclusive dealing is an example of the common error of trying to determine if the facts of a case fit a preconceived economic model instead of developing an appropriate economic model that best explains the facts. In particular, Dentsply's procompetitive rationales were rejected because they failed to correspond with a particular, fairly narrow economic theory of how exclusive dealing prevents dealer free-riding and the economic assumption that competition between dealers necessarily leads to the desired level of dealer promotional effort.
The major purpose of this paper is to expand the economic framework under which we conduct our analysis of exclusive dealing contracts. We provide an economic basis for the fundamental business reality that manufacturers often want their dealers to supply more promotion than the dealers would independently decide to supply. This leads manufacturers to contract with (and compensate) their dealers for providing increased promotion. Using examples taken from important exclusive dealing cases we show that dealers have an incentive to violate these contractual arrangements in three distinct ways that can usefully be described as free-riding and demonstrate how exclusive dealing may be used to mitigate all three forms of free-riding. Within this more realistic economic framework exclusive dealing is shown to prevent free-riding in cases where a manufacturer does not provide any promotional assets to its dealers (but merely compensates dealers for their increased promotional efforts), and where dealers do not switch their sales efforts to the promotion of rival brands (but merely fail to supply the promotion the manufacturer has contracted and paid for). Consequently, the use of exclusive dealing to avoid free-riding and to create dedicated dealers are justifications that have much broader applicability than previously recognized.(18)
The standard avoidance of dealer free-riding theory of exclusive dealing refers to cases where a manufacturer makes investments in promotional assets that it provides to its dealers free of charge.(19) These investments often include, for example, dealer display fixtures or salesperson training. The manufacturer then expects its dealers to use these assets to promote its products, and not the products of rival manufacturers. In Beltone, for example, these manufacturer-supplied promotional investments consisted of sales leads.(20) Beltone, a hearing aid manufacturer, advertised its products in magazines where prospective customers filled out cards requesting additional information. Beltone, which sold its products through exclusive dealers that were granted exclusive sales territories,(21) supplied these sales leads to the dealer located nearest the prospective customer.(22)
Since Beltone directly provided its dealers with significant promotional assets in the form of sales leads, it is obvious that Beltone would want its dealers to use these investments to sell its products. However, when a manufacturer directly supplies a dealer with investments that potentially can be used to sell other manufacturers' products, a potential dealer free-riding problem is created. Specifically, dealers will have an incentive to use the manufacturer's investments to sell a rival manufacturer's brand if they can earn a higher profit margin on the rival brand.
The dealer's profit margin is very likely to be higher on an alternative, low brand name product because the dealer generally can purchase such alternative products at lower wholesale prices. Low brand name products have lower wholesale prices because the manufacturers of these products do not bear the costs of supplying promotional investments to dealers. In addition, dealers demanding a low brand name product usually will have a choice of several highly substitutable alternative suppliers, each of whom faces a highly elastic demand by dealers. The wholesale price of the alternative, low brand name product, therefore, will be much closer to marginal manufacturer production cost than the branded product.
It may appear that the alternative, low brand name manufacturer is free-riding on the brand name manufacturer that has made the promotional investment, in effect using the brand name manufacturer's investment as its own. However, it is the dealer that is ultimately engaging in free-riding by violating its implicit contract with the manufacturer when using the manufacturer-supplied assets to sell a rival manufacturer's product. In addition, it is primarily the dealer that is benefiting from the switching because competition between low brand name rival manufacturers in supplying the substitute product will mean that rival manufacturers are unlikely to significantly profit from the free-riding.
A similar exclusive dealing case, where the manufacturer directly supplied significant promotional investments valued by its dealers, is Ryko.(23) Ryko, a manufacturer of automatic car-wash equipment, used exclusive dealing/exclusive territory contracts with its dealers, who were responsible for promoting the sales of Ryko equipment to car washes and gasoline stations in their designated areas.(24) As part of the marketing process Ryko made sales presentations to national gasoline companies, who then would decide whether to recommend the product to their gasoline station dealers.(25) Once Ryko convinced the national gasoline company of the potential value of the product, this information was supplied to the operators of the gasoline company's stations in their areas.(26)
The Ryko dealer was responsible for completing the sale by convincing individual gas stations to purchase the Ryko system.(27) However, without Ryko first devoting resources to obtain approval of its system from the national gasoline company, it is much less likely that local Ryko dealers would have been able to make the ultimate sale to the gasoline stations.(28) Litigation arose because Eden Services, a local Ryko distributor, promoted its own water reclaim system at the expense of the Ryko system when making its presentation to potential gasoline station buyers in violation of Ryko's exclusive dealing contract.(29) Ryko then terminated Eden and Eden sued, challenging the exclusive on antitrust grounds.(30)
In general, dealers will have to make an extra effort to switch consumers to rival brands when consumers visit a dealer, such as a Beltone hearing aid dealer, with an expectation of purchasing a particular manufacturer's product. However, dealers will have an incentive to devote resources to switching sales from the manufacturer's product to an alternative, low brand name product that the dealer can sell at a lower price because the dealer earns a greater profit margin on the alternative product. For consumers that do not have a strong preference for the manufacturer's product, switching may be accomplished merely by the dealer asserting that the lower-priced, lower brand name alternative product is "just as good" as the manufacturer's product the consumers may have initially asked for.
In contrast, dealers may not always disclose to customers that a substitution is being made, and therefore can "pass off an inferior product as the supplier's own"(31) In cases where buyers are unaware that the dealer has switched them to an alternative, possibly inferior product, the cost to the dealer of switching buyers may be very low or nonexistent and the investments provided by a manufacturer to its dealers can be costlessly used by dealers to sell alternative products.(32)
It is obvious why a manufacturer that wishes to prevent the type of potential free-riding illustrated in Beltone and Ryko may use exclusive dealing. An exclusive, by prohibiting dealers from selling any competing hearing aid brand or water reclaim system, prevents Beltone and Ryko dealers from engaging in free-riding on the manufacturer's investments by preventing dealer switching of consumers to other brands. And by permitting the manufacturer to capture the return on its investments, the exclusive encourages manufacturers to make valuable investments in generating sales.(33)
This procompetitive rationale for exclusive dealing was accepted in both Beltone and Ryko. In Beltone the FTC held that, by protecting Beltone's promotional investments, the exclusive was justified because it encouraged Beltone to make promotional investments.(34) Similarly, in Ryko the court noted that Eden's behavior in promoting its own water reclaim system at the expense of the Ryko system in violation of the exclusive dealing contract amounted to free-riding on Ryko's marketing efforts. The court concluded that the exclusive contract was procompetitive because it made it more likely that manufacturers would undertake valuable marketing activities in the first place because they need not fear that the increased sales created by such activities would be partially lost to other firms.(35)
This standard analysis of exclusive dealing as a way to prevent free-riding on manufacturer promotional investments was rejected by the court in Dentsply because the court found that Dentsply's promotional investments were "purely brand-specific" and, therefore, dealers could not free-ride by using such investments to sell rivals' products.(36) It is unclear exactly what the court understood "purely brand-specific" to mean. Any manufacturer investment that gets the customer into the dealership or otherwise increases a customer's demand for the dealer's services, including brand-specific advertising, creates the potential for dealer free-riding by switching customers to rival brands. Beltone and Ryko clearly illustrate that a dealer may free-ride upon manufacturer investments that are brand-specific. Marvel similarly argued in Dentsply that without exclusive dealing it would not have been economic for Dentsply to undertake the required branded promotion to introduce new products because of the potential dealer free-riding that would exist.(37)
Whether dealers could in principle free-ride on Dentsply's investments or not is irrelevant because in rejecting Marvel's analysis the court noted there was an absence of any evidence of dealers switching dental labs to rival brands.(38) Furthermore, rather than exclusive dealing encouraging increased manufacturer investments, the court cited testimony by Dentsply executives that absent exclusive dealing Dentsply actually would have increased these brand-specific investments.(39) However, dealer free-riding can occur in circumstances where manufacturers do not make any investments whatsoever (section III) and where dealers do not switch customers to rival brands (section IV) and, therefore, where the absence of exclusive dealing may increase manufacturer promotion as an inefficient substitute for lost dealer promotion.
The court in Dentsply concluded that inter-dealer competition provides dealers with the correct incentive to promote a manufacturer's products and that, therefore, the supply of dealer services can be left up to competition among dealers without any contractual interference by the manufacturer with regards to requiring exclusivity.(40) However, an important economic fact about real world business relationships is that in most markets manufacturers desire more dealer promotion of their products than dealers would independently provide.(41) Contrary to the assumption made by the court in Dentsply, uncontrolled dealers will, in general, supply less than the desired (joint manufacturer and dealer profit-maximizing) amount of promotional services.
Marvel acknowledges that dealers will sometimes supply less than the amount of promotional services desired by the manufacturer, but only when dealers engage in inter-dealer free-riding, where dealers do not supply customer service, such as product demonstrations, but instead encourage their customers to first go to a full-service dealer to receive these services before purchasing the product from them at a lower price.(42) Because dealers can free-ride on the services provided by competing dealers in this way, each dealer will have an incentive to supply less than the optimal quantity of dealer services. However, Marvel maintains that, even in these circumstances, "exclusive dealing is not an efficient means by which to promote increases in dealers services."(43) If such inter-dealer free-riding existed, exclusive dealing would have no effect in mitigating the free-riding and would not encourage dealers to increase the supply of services since an identical inter-dealer free-riding incentive continues to exist under exclusivity.(44) Consequently, Marvel and the Dentsply court conclude that "enhancing dealer services cannot be the justification for exclusive dealing."(45)
However, even in the absence of such inter-dealer free-riding, manufacturers cannot leave it entirely up to their dealers to decide how much promotion to supply in marketing their products because dealers will not take account of the profit earned by the manufacturer on the incremental sales produced by the dealer's promotional efforts. The incremental profit earned by manufacturers on additional sales is often significant because most manufacturers face a negatively sloped demand and, hence, set dealer wholesale prices above marginal cost.(46) Dealers will not take account of this profit earned by manufacturers on incremental sales in determining their level of promotion.
In general, a dealer's failure to take account of incremental manufacturer profitability on additional sales will not lead to any distortion with regard to insufficient dealer price competition or supply of services that have significant inter-dealer demand effects. Although dealers do not take account of the manufacturer's profit when deciding to, for example, lower price, a lower dealer price produces a much larger increase in an individual dealer's demand than in the manufacturer's demand since consumers purchasing the manufacturer's product at other dealers switch their purchases to the dealer that has lowered its price. Because the individual dealer response to price changes will be much larger than the manufacturer response due to these inter-dealer competitive effects, the dealer's incentive to lower price will be the same as the manufacturer's incentive to lower price even if the manufacturer's margin on incremental sales is substantially greater than the dealer's margin. Inter-dealer competition largely eliminates any distortion with regards to individual dealer incentives to engage in price competition.(47)
Similarly, when dealers supply a service that produces large inter-dealer effects, such as the supply of free, convenient parking that shifts sales between dealers, competition between dealers leads dealers to provide the desired quantity of these services. An individual dealer that supplies the service experiences a shift out in its demand resulting in an increase in the price it can charge for its products and/or the sale of increased quantity. Once again, although the dealer will not consider the extra manufacturer profit from incremental sales produced by the provision of free parking, inter-dealer competition will lead dealers to provide free parking because it attracts customers from other dealers. Therefore, when dealers make decisions about supplying such services, the dealer's higher elasticity of demand compared to the manufacturer's elasticity of demand with respect to provision of the services will offset the dealer's lower profit margin on incremental sales compared to the manufacturer since the services can be thought of as similar to a lower price.
With regard to a dealer's promotion of a manufacturer's product, however, dealers will not undertake the promotional efforts required to generate the incremental sales that would be profitable to the manufacturer because there are unlikely to be any significant inter-dealer demand effects from a dealer's promotion of a manufacturer's product, which we refer to as inter-brand promotion.(48) In contrast to the provision of non-price services such as free parking, inter-brand promotion generally will not lead consumers to switch dealers in response to its provision.(49) However, the extra sale made by an individual dealer as a result of its promotion represents an extra sale to the manufacturer.
Neither can a dealer profit from providing inter-brand promotion by charging customers for the promotion. This is because a dealer's inter-brand promotion is aimed at "marginal" consumers who, absent the promotion, would not otherwise purchase the product. Retailer promotion increases the reservation values of some of these consumers so that they become equal to (or greater than) the retail price, and the consumers decide to purchase the product. The promotion can be thought of as a way to provide a targeted effective price discount to these particular "marginal" consumers to induce incremental sales of the manufacturer's product. To operate, therefore, such promotional services must be provided to marginal consumers free of charge. The dealer cannot charge for the promotion or it would defeat its very purpose.(50)
Consequently, the dealer's failure to take account of incremental manufacturer profit and inability to charge customers for its promotional efforts will lead to too small an amount of dealer inter-brand promotion. Although manufacturers would be willing to pay the dealer's increased costs of providing additional promotion that would be more than covered by the manufacturer's additional profit on the incremental sales generated by the dealer's promotion, dealers will not independently provide sufficient inter-brand promotion because they will not cover their increased cost of providing the promotion.
Because manufacturers will want greater dealer promotion provided for their products than dealers would choose to supply on their own, they must find a way to get the desired (joint manufacturer and dealer profit-maximizing) quantity of promotion supplied by dealers. This is what creates the economic motivation for the manufacturer to supply promotional assets to its dealers free of charge that underlies the standard case of dealer free-riding. More generally, because much promotion is efficiently provided by dealers at the point of sale, manufacturers will contract with their dealers to supply an increased level of promotion, with manufacturer supply of free promotional assets only one possible element in the overall manufacturer-dealer contractual arrangement.
Manufacturer contracts with dealers for added promotion may be either explicitly written documents or, more generally, implicit contracts. An example of an explicit contract was involved in the landmark Supreme Court exclusive dealing case, Standard Fashion.(51) Standard Fashion, a manufacturer of dress patterns, employed exclusive dealing, requiring the retailers who sold its patterns not to handle any competing patterns.(52) Standard also explicitly contracted with its retailers to actively promote its patterns.(53) In particular, Standard required its retailers to provide a pattern department at "a prominent position on the ground floor in the store,"(54) a designated "lady attendant" to give "proper attention to the sale" of patterns,(55) and a minimum inventory level.(56)
The fact that Standard Fashion contracted with its retailers for promotional inputs is evidence that retailers did not have the correct incentive to independently supply promotional services. As with most products that primarily involve intellectual property, Standard Fashion had a low marginal cost of producing additional copies of its existing patterns and, because it faced a negatively sloped demand for its somewhat unique patterns, a profit-maximizing wholesale price that was substantially above its marginal cost. Consequently, any incremental sales made by a retailer were highly profitable for Standard Fashion. Standard Fashion found it necessary to contractually specify particular retailer promotional inputs because, as described above, its retailers would not take account of this significant incremental profit earned by Standard on each additional sale produced by retailer promotional efforts.
Even when an explicit contract for specific dealer promotion, as in Standard Fashion, does not exist, it is likely that an implicit contractual arrangement is present between the manufacturer and its dealers. For example, in both Beltone and Ryko the manufacturer did not merely provide its dealers with promotional assets and then, as long as dealers did not engage in switching the manufacturer's investments to the sale of rival products, leave it completely up to the dealers to decide how much promotional effort to supply. In addition to providing significant promotional investments to its dealers (in the form of individual customer sales leads in Beltone and a national sales effort to gasoline companies in Ryko), Beltone and Ryko both desired their dealers to supply additional complementary promotional services and compensated their dealers for supplying such services.(57)
Whether the manufacturer's contract with its dealers is explicitly written or only implicitly understood, the manufacturer generally will self-enforce rather than court-enforce the contractual arrangement. That is, whether or not the manufacturer explicitly writes what it expects its dealers to do with regard to the supply of promotion, when a dealer does not perform as expected, the manufacturer generally will not take the dealer to court to demand performance but will merely terminate the relationship.(58)
In order for the threat of termination to assure dealer performance with regard to the supply of adequate promotion, dealers must earn more by supplying the explicitly or implicitly contracted for level of promotion than they could earn by violating the contract and supplying the level of promotion that they independently find in their interests to supply. Therefore, manufacturers must establish a contractual arrangement where dealers expect to more than cover their increased costs of supplying the contracted for promotion. If dealers are to be incentivized to supply the higher, desired level of promotional services, they must earn a profit premium over and above their increased costs of providing the contracted for promotion.(59)
A dealer profit premium is required for the self-enforcement mechanism to operate because each dealer, in deciding whether to perform according to the contractual arrangement and supply the desired level of promotional services or not, will compare the net present value of the profit from performing as contracted, , with the net present value of the profit from not performing and violating the contract, . Since dealers can earn extra profit for a period (before they are terminated by the manufacturer for non-performance) by deciding not to supply the promotion they have contracted and been compensated for and only supply the quantity of promotion it is in its narrow interests to provide, is positive. The dealer saves the extra net cost of supplying the higher, contracted for promotion (and may earn an increased margin on the products it switches consumers to) until the manufacturer detects non-performance and terminates the dealer.
The dealer will perform as contracted with the manufacturer if and only if these gains from not performing are less than the gains from performing.
Therefore, to assure that the dealer performs as contracted and supplies the larger, contracted for level of promotional services, the dealer must earn a profit stream, the present discount value of which, , is greater than the short-term gain from not performing, . Consequently, the manufacturer must more than merely compensate dealers for their higher costs of providing increased promotion. If the manufacturer merely compensated dealers for their higher costs, would equal zero and dealers would not perform. Manufacturers must create a distribution arrangement where each dealer earns a profit premium above the higher costs of supplying the contracted for level of promotional services the present discounted value of which, , or what the dealer will lose upon termination, is greater than the short-run gain that can be earned by a dealer by not performing as implicitly contracted, .
Making sure that is greater than is a primary economic role of granting dealers, such as the Beltone and Ryko dealers, exclusive territories.(60) The dealer earns a rent because it is the sole supplier within an area and is compensated for supplying the desired level of promotional services. , the present discounted value of the dealer's profit stream earned from the grant of the exclusive territory, is the capital value of the dealer's distributorship and the potential sanction that the dealer will bear if it is terminated for nonperformance. The manufacturer then monitors dealer performance (for example, by comparing a dealer's sales to other dealers' sales or by sending monitors to secretly observe dealer behavior) and terminates those dealers that do not perform as implicitly contracted and compensated for by the manufacturer.
When it is efficient to have a large number of dealers selling the manufacturer's product within an area, a manufacturer will not use an exclusive territory to generate the desired level of compensation to assure dealer performance, , as was used in Beltone and Ryko. Instead, the manufacturer may lower the wholesale price and use defacto minimum resale price maintenance to insure that inter-dealer price competition does not compete away the dealer's compensation.(61) This is, in fact, how Standard Fashion compensated its retailers for the desired, contractually specified level of promotional efforts.(62) In particular, Standard Fashion set its wholesale prices at 50percent of retail label prices and contractually required its retailers "not to sell Standard Patterns except at label prices."(63) Minimum resale price maintenance created a per unit sales profit stream for Standard Fashion retailers that would cover the desired retailer-supplied promotion and prevent the profit stream from being competed away in lower retail prices by inter-retailer price competition. Standard Fashion's retailers knew that if they followed the contractual arrangement and supplied the specified promotional efforts, they could expect to earn enough to cover their higher costs of supplying the desired promotion plus an added profit premium to assure performance.
If the costs of providing extra promotional services are largely per unit time (for example, the rent of a ground floor location, the salary of attendants and extra inventory), without minimum resale price maintenance each retailer would have the incentive to lower its price to capture increased sales from infra-marginal consumers who knew they wished to purchase the particular product and were price sensitive. If uncontrolled, this price competition between dealers would eliminate Standard Fashion's per unit sale compensation to retailers for supplying the desired extra promotional services.
Whether a manufacturer provides its dealers with significant promotional investments (the standard free-riding case) and also pays its dealers for supplying additional complementary promotional services, as in Beltone and Ryko, or uses a contractual arrangement where it does not provide any investments to dealers but solely compensates dealers for making the additional promotional investments, as in Standard Fashion, the essential nature of the contractual arrangement between a manufacturer and its dealers is the same. The manufacturer is contracting with its dealers to supply additional promotion with certain understandings. In Beltone and Ryko the understanding is that each dealer will use the promotional leads provided by the manufacturer in combination with its own promotional efforts to sell Beltone hearing aids or Ryko car washing equipment. Beltone and Ryko are paying their dealers to supply the desired amount of complementary dealer promotion with a valuable exclusive territory. In Standard Fashion the manufacturer is paying for dealer-supplied promotional inputs with resale price maintenance with the understanding that adequate dealer-provided promotional inputs, some of which were contractually specified, will be used to promote and sell Standard Fashion patterns.
In the cases we have described, Beltone, Ryko and Standard Fashion, the potential exists for dealers to free-ride by violating the implicit contract with the manufacturer regarding the supply of promotion. One form of such potential dealer free-riding involves dealers using the promotion paid for by the manufacturer to sell rival products. Similar to the standard type of free-riding where dealers use promotional investments supplied by the manufacturer to sell rival products, a short-term profit incentive exists for dealers to use the extra promotional services purchased by a manufacturer to sell rival products with higher dealer profit margins.
As in the standard free-riding case described in Section II, the dealer generally can earn more by convincing consumers to purchase lower-priced rival manufacturer products because the rival products have lower wholesale prices. Rival manufacturers do not bear the costs of purchasing the dealer promotion and face highly elastic dealer demand. For example, retailers selling Standard Fashion patterns will have a profit incentive to use the sales staff, floor space, and other inputs paid for by Standard Fashion to switch consumers to a competing pattern that the retailer makes more profit on and can describe to consumers as "just as good." Although the dealer free-riding motivation is identical to the standard case, it does not appear to fit the standard free-riding framework because there need not be any manufacturer-supplied investments that dealers use to sell rival products. Instead, all the investments are provided by the dealers, but are paid for by the manufacturer. (64)
Since manufacturer compensation of dealers for promotion is contingent on the dealer's sales of the manufacturer's products, a dealer that switches consumers to a rival's products may not appear to be free-riding since manufacturer compensation is reduced. As described above, manufacturer compensation of dealers for promotion often takes the form of an extra profit margin on the dealer's sales of the manufacturer's products via the grant of an exclusive territory or the enforcement of minimum resale price maintenance.(65) Therefore, a dealer that switches sales to rivals will be paid less by the manufacturer.
Dealer switching of sales to rival products, however, is free-riding because dealers operating under a per unit sales compensation arrangement generally are paid for their promotional efforts on the basis of a profit premium they earn on their total sales of the manufacturer's products, not on the difficult to measure incremental sales produced by their extra promotional efforts. Therefore, the dealer will have an incentive to use the promotional inputs paid for by the manufacturer to make incremental sales of more profitable rival products. The dealer will continue to receive most of the manufacturer's compensation on infra-marginal sales, yet use its promotional assets to sell more profitable rival products.(66)
Our analysis implies that exclusive dealing may be used to prevent free-riding in cases where the manufacturer has not made significant investments. Free-riding occurs because dealers use the promotional resources paid for by the manufacturer to sell rival products; exclusive dealing prevents this in the same way as exclusive dealing prevents free-riding in the standard case where dealers use the promotional assets supplied by the manufacturer to sell rival products, namely by preventing dealer switching. Whether dealers are prevented from using the promotional assets supplied by the manufacturer or paid for by the manufacturer to switch consumers to rival products, the procompetitive effect of the exclusive dealing is the same. Therefore, contrary to the court's analysis in Dentsply, there need not be manufacturer investments that dealers may free-ride upon to justify exclusive dealing in terms of the prevention of free-riding.
Exactly how exclusive dealing operates to prevent dealer switching of manufacturer-supplied or manufacturer-paid for promotion to the sale of rival products is not as obvious as it may seem. We cannot assume that because a manufacturer contractually specifies exclusivity that this, by itself, prevents dealer switching and dealer free-riding. In spite of exclusive dealing, some dealers may attempt to switch consumers to rival products if they can get away with it for a sufficiently long period before the manufacturer detects the contract violation and terminates. In most of these cases the manufacturer will not attempt to legally enforce its exclusive dealing contract, that is, take the free-riding dealer to court to demand exclusivity. Instead, dealers that violate exclusive dealing provisions will be terminated. This is what occurred, for example, in Ryko, where Eden, the dealer violating the exclusive, was terminated by the manufacturer, Ryko.(67) SBeltone and Dentsply, which did not involve private litigation but government challenges to exclusive arrangements, there is evidence in the record that the manufacturer in both these cases also merely terminated non-performing dealers.(68)
The role of exclusive dealing in these manufacturer/dealer contractual arrangements is in facilitating manufacturer self-enforcement of the dealer performance contract. Because an exclusive is included in the contract, representatives from Beltone that policed dealers could much more easily detect a non-performing dealer. If a Beltone representative observed a product from a rival hearing aid company at a Beltone dealer, it could infer that the dealer was violating the implicit "best efforts" promotion contract. The manufacturer, therefore, could terminate the dealer without determining if the dealer actually was switching customers produced by Beltone-supplied leads to competing brands. The exclusive serves the economic purpose of defining dealer performance in a way that decreases the manufacturer's cost of monitoring and detecting dealer non-performance.(69)
In terms of the self-enforcement model summarized by equation(1), if exclusivity makes it easier for the manufacturer to detect dealer non-performance, this reduces the potential dealer short-run gain from free-riding, , and, hence, reduces the required premium stream the manufacturer must promise the dealer in order to assure dealer performance. Therefore, exclusivity reduces the costs to Beltone of self-enforcing the contractual arrangement by decreasing the manufacturer's costs of sharing more of the joint profit with dealers, for example, by granting larger or otherwise more profitable dealerships.(70)
The fact that dealers are compensated by manufacturers for supplying promotion on the basis of all their sales implies that dealers also have an incentive to engage in a type of free-riding that does not involve switching sales to a rival manufacturer. In particular, dealers have the incentive to violate the manufacturer's implicit contract by not supplying the contracted for level of promotion. Dealers that reduce their promotional efforts save the cost of supplying the extra, contracted for promotion but continue to receive the manufacturer's compensation on the infra-marginal sales that continue to be made without the dealer's promotion. Dealers lose profit on the incremental sales that would have been induced by the promotion they fail to supply, but as described in III.A. it is not in the dealers' independent interests to bear the costs of undertaking this promotion required to induce these incremental sales.
This potential free-riding problem, where dealers do not have the incentive to supply the full amount of promotional effort contracted and paid for by the manufacturer, exists in all the cases we have discussed where the manufacturer is paying dealers for added promotion with a restricted distribution arrangement. Because restricted distribution arrangements compensate dealers on the basis of all their sales, including the infra-marginal sales that would be made without the contracted-for promotional efforts, an incentive exists for the dealer not to supply the contracted-for promotion. Therefore, if the manufacturer uses an exclusive territory (Ryko, Beltone) or resale price maintenance (Standard Fashion) to pay dealers for supplying extra promotion, the very nature of the promotion compensation arrangement creates an incentive for dealers to supply less promotion than contracted for, even when there is no possibility of a dealer switching its promotion to rival brands.
To illustrate these economic forces consider another important exclusive dealing case, Roland Machinery.(71) The case dealt with the sale of construction equipment by Roland Machinery Company, a distributor of International Harvester's line of construction equipment. Since construction equipment has relatively high fixed costs and a wholesale price that is greater than marginal cost, it is likely that Dresser Industries (which acquired International Harvester's construction equipment business) desired greater point-of-sale dealer promotion to induce profitable incremental sales than dealers would independently supply on their own. Dresser compensated its dealers for their extra promotional efforts by limiting the number of dealerships and granting each dealer a profitable, relatively large exclusive marketing territory. For example, Roland Machinery was the sole International Harvester distributor serving the 45 county area of central Illinois.(72)
In contrast to the exclusive territories granted in Beltone and Ryko, Dresser did not use an explicit exclusive dealing contract with its distributors. Instead, Dresser unilaterally enforced its distributor contracts as if exclusivity was expected. In particular, only eight months after signing its distribution agreement with Dresser, Roland Machinery applied for a dealership from Komatsu, a Japanese manufacturer of competitive construction equipment.(73) Immediately after Roland Machinery signed its Komatsu distribution agreement, Dresser notified Roland that it planned to terminate its International Harvester distributorship under the terms of its distribution agreement where termination could occur without cause on 90days' notice. (74)
Although Dresser was compensating Roland for extra dealer promotion, it is not obvious that there was an economic incentive for Roland to use the promotional services paid for by Dresser to promote Komatsu machines, that is, to engage in switching. In contrast to the standard free-riding analysis, Komatsu is not a low cost, no brand name product that Dresser could make more money switching buyers to. Moreover, there was no evidence presented that Roland had engaged in switching demanders of International Harvester equipment to Komatsu. Instead of preventing free-riding, Dresser maintained that it preferred exclusive dealing because it wished to distribute through dealers that had "undivided loyalty" and, therefore, an increased incentive to promote its products.(75) The court accepted Dresser's "undivided loyalty" procompetitive rationale for exclusive dealing, concluding that exclusive dealing "leads dealers to promote each manufacturer's brand more vigorously than would be the case under nonexclusive dealing."(76)
The "undivided loyalty" rationale for exclusive dealing accepted by the court in Roland Machinery has been accepted by a number of courts. For example, a similar rationale for exclusive dealing arrangements was accepted in Joyce Beverages v. Royal Crown Cola.(77) The court concluded that Royal Crown Cola's requirement that its bottler distributors not carry any other cola brand in order that distributors use their "best efforts to achieve maximum distribution and sale"(78) had the effect of increasing rather than inhibiting competition because the exclusive "insures that the bottler devotes undivided loyalty to its particular brand and that it competes vigorously against all competing brands."(79) Identical reasoning also was used to justify the exclusive in Hendricks Music Co. v. Steinway, where the court held that "[i]t is perfectly legitimate and, in fact, procompetitive, for manufacturers to insist that their dealers devote undivided loyalty to their products and not to those of their competitors."(80)
While this reasoning that exclusive dealing encourages dealer promotion by creating dealers who have undivided loyalty is intuitively appealing and is consistent with the academic marketing literature, which recognizes that manufacturers may use exclusive dealing where they "hope to obtain more dedicated and knowledgeable selling,"(81) the reasoning does not have a rigorous economic basis. In fact, Howard Marvel makes a cogent argument in his paper that the "undivided loyalty" rationale for exclusive dealing makes no economic sense. Marvel assumes that the only reason dealers may not adequately provide services to consumers is if they are taking advantage of the efforts of other dealers by engaging in the inter-dealer free-riding that was the focus of Sylvania, that is, if dealers are letting other dealers provide product demonstrations and other valuable dealer services to consumers before the consumers purchase the product from dealers who do not provide the services.(82) Marvel correctly argues that exclusive dealing would not solve this type of inter-dealer free-riding problem. Dealers who handle only one brand still have the incentive not to supply services to consumers and to free-ride on the promotion provided by other, full-service dealers.(83) Marvel further argues that if there is not an inter-dealer free-riding problem, dealers will have adequate incentives to provide services to consumers, leading him to conclude that "exclusive dealing is not an efficient means by which to promote increases in dealer services."(84)
Because a rigorous economic basis for an "undivided loyalty" rationale for exclusive dealing did not exist, the Dentsply court did not accept Dentsply's attempt to justify its exclusive contracts with dealers because of "the need for dealers to focus their efforts in order to effectively promote the company's teeth and service laboratory customers."(85) The Dentsply court, very well aware of Howard Marvel's explicit rejection of the possibility that exclusive dealing could legitimately be used to encourage increased dealer promotion of a manufacturer's product, noted that "Prof. Marvel has not endorsed this particular rationale for exclusive dealing To the contrary, he stated in his 1982 paper that enhancing dealer services cannot be the justification for exclusive dealing."(86) The court fully accepted Marvel's economic reasoning in rejecting an undivided dealer rationale for exclusive dealing. The court concludes that "[t]he 'focus dealer services' rationale is not a valid justification for using exclusive dealing in the tooth industry because dealers have every incentive on their own to make sure that their level of service for any given tooth brand does not suffer If a customer is dissatisfied with the service it receives from one Dentsply dealer, it will simply buy [Dentsply] teeth from another dealer."(87)
Judge Posner in Roland Machinery makes an attempt to provide an economic explanation for the undivided loyalty rationale for exclusive dealing in concluding that Dresser had "a plausible argument that an exclusive dealer would promote its line more effectively than a nonexclusive dealer, and by doing so would increase competition in the market for construction equipment."(88) Posner argues that by granting Roland Machinery an exclusive territory in central Illinois, Dresser had put "all of its eggs in the Roland basket", and therefore if Roland did not promote Dresser products vigorously, Dresser would not have another dealer to "fall back on."(89) Because Dresser would suffer a significant decrease in sales if Roland did not promote Dresser products vigorously, this gives Dresser an economic reason to "want that dealer to devote his efforts entirely to selling Dresser's brand."(90)
However, Judge Posner does not tell us why Roland would not have the correct incentive to promote Dresser's products absent exclusive dealing. Nor does Judge Posner Judge explain why a manufacturer would want its dealers to exclusively distribute its products even in the absence of exclusive territories. Moreover, Judge Posner does not tell us how exclusive dealing works in this context to incentivize dealers such as Roland to increase their supply of promotional services, merely asserting that Roland's acceptance of an exclusive dealing contract "indicates [Roland's] commitment to pushing that brand; he doesn't have divided loyalties If the dealer carries several brands, his stake in the success of each is reduced."(91) But this does not explain why an exclusive dealer will find it in its economic interests to promote a brand more intensively.
While it may seem intuitively appealing that a manufacturer may use exclusive dealing to create a dedicated distribution arrangement so that dealers have "undivided loyalties", a rigorous economic explanation of why manufacturers want dealers to provide more promotional services than they would otherwise provide and exactly how exclusive dealing induces dealers to do so is not in the economic literature. In fact, the idea that exclusive dealing will increase desired dealer promotion is contrary to generally accepted current economic analysis. Dentsply was unfortunate enough to have a judge that was highly cognizant of the state of the economic knowledge on this issue.
To understand how exclusive dealing creates an incentive for dealers to more actively promote the manufacturer's product in a case such as Roland Machinery, consider the analogous example of automobile manufacturers who wish to have their dealers provide more promotion than the dealers would independently decide to supply in order for the manufacturer to take advantage of the significant profit it can earn on incremental sales. For example, a manufacturer may desire its dealers to remain open longer hours, hire a larger number of knowledgeable and persuasive salespeople, and have the dealer's salespeople spend a greater amount of time demonstrating its products and supplying test drives to "marginal consumers" than the dealer would otherwise supply. Automobile manufacturers, therefore, implicitly contract with their dealers for an increased supply of dealer promotional activity.
The usual contractual arrangement is for the automobile manufacturer to compensate its dealers for providing this extra promotion by limiting the number of dealerships and, thereby, granting each dealer a valuable franchise. Because each dealership serves a particular geographic area, dealers are, in effect, granted a profitable sales and aftermarket service business as a way to compensate for the provision of promotional services.(92) If the manufacturer does not believe the dealer is adequately promoting its products, the dealer's allocation of automobiles will be reduced and, in extreme cases, the dealer could be terminated.
An automobile dealer that handles multiple brands has an increased incentive to violate this implicit contract and free-ride on the manufacturer's compensation arrangement. In contrast to the examples of free-riding described above, this generally does not occur by dealer switching, that is, by a dealer using the promotional resources that have been supplied or paid for by the manufacturer to switch consumers without strong brand preferences to lower-priced, non-brand name rival products. More relevantly, dealers can violate the implicit contract by not undertaking as much promotional effort in selling the manufacturer's product as the manufacturer has implicitly contracted and paid for.
Consider, for example, a case where a customer who is leaning towards the purchase of a Honda comes into a Toyota dealership to check out the Toyota. Toyota desires the salespeople at its dealers to take the time and make their "best efforts" to extol the advantages of Toyota compared to Honda. However, if the dealer's salespeople have the ability to sell Hondas in addition to Toyotas, it will not be in their interests to undertake this extra promotional effort. It will be more economic for the salespeople to save the additional costs associated with the extra time and effort that would have to be expended to sufficiently increase the possibility of a Toyota sale and, instead, merely sell the Honda by telling the customer he is right, "Yes, Honda is better." From the salesperson's and the dealership's point of view, everything else equal, there is a higher profit margin on the Honda sale because there are lower selling costs associated with the Honda sale. Although selling the Honda increases the dealer's profit, failing to actively promote Toyota products in this circumstance violates the implicit contract between Toyota and the dealer. Toyota has provided the dealer with a valuable dealership in return for the dealer's implicit promise to actively promote Toyota products, which the dealer has failed to do.
Although the salesperson has not engaged in switching, this violation of the implicit manufacturer contractual arrangement is analytically similar to the type of free-riding we discussed in sections II and III where the dealer switches a consumer to an alternative, rival brand. In the switching type of free-riding the dealer actively promotes an alternative brand because the profit margin on the alternative brand is greater. In the type of free-riding we are discussing here, the dealer decides not to promote the manufacturer's brand because it costs the dealer less to sell the alternative brand. In switching free-riding the dealer is pushing consumers to higher margin products that have lower costs because the manufacturers of the alternative products have not borne the costs of promotion; in this type of free-riding the dealer is similarly selling a product that has a higher margin because the dealer itself can save the costs of promoting the manufacturer's product.
When the dealership is exclusive, the incentive of the dealer to violate the implicit contract by not supplying the contracted and paid for promotion is substantially reduced because the Toyota dealer does not have the ability to sell the Honda initially favored by the customer. Therefore, if the dealer is to make the sale, it must promote the Toyota product.(93) The exclusive dealing contract, therefore, serves the economic purpose of more closely aligning the incentives of the dealer with the incentives of the manufacturer by creating undivided dealer loyalty.
This analysis of undivided dealer loyalty can be presented more rigorously. Without exclusive dealing, when a customer leaning towards a Honda comes into the dealership to check out the Toyota, the dealer will choose the level of Toyota promotion which maximizes its expected net profit from making the Toyota sale. This can be represented by the following dealer profit function:
That is, dealers will choose to supply a level of Toyota promotional services, S, so as to maximize their expected net return, where is the dealer's profit margin from selling a Toyota, is the dealer's profit margin from selling a Honda (which we can assume can be sold with little or no promotion), p is the dealer's probability of making the Toyota sale, which is assumed to be positively related to the amount of Toyota promotional services supplied by the dealer, is the cost of promoting the Toyota, which is also a positive function of .
The profit maximizing level of Toyota promotion that a non-exclusive dealer chooses is given by:
Dealers will invest in making Toyota promotional expenditures up to point where the increased cost associated with providing increased promotion, C'(S), equals the increase in the expected profitability of selling a Toyota, or the increased probability of making a Toyota sale as promotional expenditures are increased, p'(S), multiplied by the dealer profit difference between a Toyota and Honda sale, .
Equation(3) implies that if , that is, the dealer's profitability of selling the Toyota and Honda are the same, it will never pay the dealer to promote the Toyota when a customer is leaning towards a Honda. There is no incremental profit associated with the supply of Toyota promotion. A dealer, and specifically a dealer's sales staff, would not waste its time and resources if it can make just as much money with less effort by selling a Honda.
On the other hand, if the dealer is an exclusive Toyota dealer, even when a customer is leaning towards the purchase of a Honda, the dealer will spend resources promoting the Toyota up to the point where:
The dealer will increase its promotional expenditures until the increased cost of the promotion is equal to the expected profitability of the promotion, that is, the increased probability of selling a Toyota multiplied by the profit margin on the incremental Toyota sale.
Therefore, even if the profit margin on selling a Toyota is not greater than on selling a Honda, the dealer will promote Toyota under an exclusive because if the dealer does not sell the Toyota, the dealer does not make any sale. In Roland Machinery, for example, the dealer's motivation to promote is altered under exclusive dealing because if the dealer does not make an International Harvester sale, it does not sell anything. Consequently, when a customer comes into an exclusive dealership, the manufacturer can be assured that the salesperson will "push" the manufacturer's products more than if the dealership were not exclusive.
The dealer promotion supplied under an exclusive, given by (4), is still not the manufacturer-dealer joint-profit maximizing level of promotion, which occurs at
where is the Toyota manufacturer's profit in selling an additional car. However, exclusive dealing clearly does move dealer promotion closer to the desired level by better aligning incentives than a contractual arrangement without an exclusive.
Dealer incentives to promote under alternative contractual arrangements represented in equations (3) - (5) can be compared by transposing the equations so that the left hand side of each equation is equal to , the marginal cost of additional Toyota promotion divided by the increased probability of making a Toyota sale from the additional Toyota promotion. Dealers will choose a level of Toyota promotion where this ratio will equal the additional profitability of making a Toyota sale under the alternative conditions. These economic forces are illustrated in Figure1, which plots as a function of the quantity of S.(94)
Figure 1 indicates that without an exclusive dealing contract, the dealer will provide inter-brand promotional services equal to SNE. If , then the dealer will not provide any additional inter-brand promotional services demanded by the manufacturer (SNE=0). More generally, the inter-brand promotional services independently provided by the dealer will be significantly lower than the level the manufacturer desires. Exclusive dealing moves the dealer incentive to supply the inter-brand promotional services desired by the manufacturer from SNE to SED because the dealer cannot save promotion costs by engaging in free-riding by making an easy sale of a competing brand. However, the exclusive, by itself, does not induce dealers to supply the amount of promotion implicitly contracted for by the manufacturer that maximizes the joint profit of the manufacturer and dealer, SMAX. The manufacturer must still monitor dealer performance and self-enforce the implicit contract to insure that dealers go the remainder of the way and provide the joint profit-maximizing level of promotional services.
Monitoring and enforcing dealer performance may occur, for example, by comparing a dealer's sales volume with the sales of other dealers, by using performance measures such as customer surveys to determine sales and service satisfaction, and by providing dealers with a profit stream that can be lost by termination or reduced by a decrease in the dealer's supply of "hot" models. The fact that exclusive dealing moves dealers a significant portion of the way towards the desired level makes it easier for the manufacturer to use these self-enforcement mechanisms. Specifically, because dealers operating under exclusive dealing find it in their own independent self-interests to supply increased promotion, the short-run dealer profit incentive to supply less than the desired level of promotional services, , is reduced. Therefore, the amount of costly manufacturer policing (which controls ) and the required self-enforcing profit premium the manufacturer must pay dealers, , is also reduced by exclusive dealing.
The common assumption made by economists that manufacturers can rely completely on inter-dealer competition to supply the desired level of promotional services is not generally valid. Manufacturers, in fact, often contract with and compensate their dealers to supply more promotion than the dealers would independently supply. Because dealers are supplying more promotion than they otherwise would find profitable to supply absent the manufacturer contract, a dealer short-run profit incentive exists to violate the contract. We have seen that contractual violations and dealer non-performance may occur in three distinct ways: (1)dealer switching of manufacturer-supplied promotional assets to the sale of rival products, the recognized case of dealer free-riding, as well as (2)dealer switching of manufacturer paid-for promotion to the sale of rival products and (3)dealer undersupply of manufacturer paid-for promotion. In all three cases dealers can be said to be free-riding on the manufacturer's arrangement for compensating in receiving manufacturer assets or compensation and then not meeting their end of the bargain by adequately promoting the manufacturer's products.
Many manufacturer-dealer contractual arrangements are likely to contain the potential for all three forms of dealer free-riding. For example, in Beltone and Ryko, where manufacturer promotional investments were supplied to dealers, manufacturers also made payments to dealers for added promotion. Therefore, in addition to the first type of dealer free-riding discussed in the Beltone and Ryko decisions, the second and third types of dealer free-riding was also clearly present. Exclusive dealing, in addition to preventing dealers from using the promotional assets supplied by the manufacturer to sell rival products also prevented dealers from using their own promotional efforts to sell rival products (free-riding type two) and had the effect of increasing independent dealer incentives to promote (reducing free-riding type three). More generally, in all the legal cases we have examined exclusive dealing can be thought of as assuring manufacturers they will receive the increased dealer promotion they are paying for by facilitating self-enforcement of the manufacturer's contractual arrangement. By constraining the ability of the dealer to switch promotional efforts to rival products and by aligning dealer and manufacturer incentives, exclusive dealing has the effect of decreasing the dealer's short-run profit potential from not performing as contracted, .
This economic framework expands what one should look for in determining whether there is a legitimate procompetitive justification for an exclusive dealing contract. In particular, the presence of free-rideable manufacturer investments and dealer switching, the conditions focused upon by the court in Dentsply, are not necessary conditions for determining whether a prevention of free-riding justification for exclusive dealing makes economic sense. All that is required for exclusive dealing to be used to prevent dealer free-riding is that dealers have a significant economic role in the promotion of the manufacturer's product, that manufacturers are compensating dealers for the supply of additional promotion and that exclusive dealing encourages such extra dealer promotion by facilitating manufacturer enforcement of its implicit contract for dealer promotion.
An undivided dealer loyalty rationale for exclusive dealing, therefore, may have been consistent with the evidence in Dentsply. The Department of Justice's proposed Findings of Fact unambiguously indicates that Dentsply relied significantly on dealers to promote the sale of its products.(95) The Department of Justice offered these facts regarding the crucial role of Dentsply's dealers in promoting Dentsply's products in order to demonstrate that the distribution of artificial teeth through dealers was economically essential in order for a manufacturer of artificial teeth to compete effectively. But in demonstrating the anticompetitive effect of Dentsply's exclusive dealing contracts in controlling a key channel of distribution, the government also established the necessary economic conditions for a procompetitive undivided dealer loyalty justification for Dentsply's exclusive dealing contracts.
Moreover, the use of exclusive dealing to create dedicated dealers that would more actively promote Dentsply products is fully consistent with the testimony by Dentsply executives that absent exclusive dealer Dentsply would have increased its own promotional investments.(96) Because the role of exclusive dealing is to encourage dealer promotion and not to protect manufacturer investments against dealer switching of customers to rival brands, we would not expect Dentsply to decrease its investments absent exclusive dealing. Since there is little or no dealer switching, Dentsply need not fear that its investments would be appropriated. Instead, Dentsply would find it economic to increase its promotional investments as a less efficient substitute for the preferred promotion that dealers would supply if they operated under exclusive dealing contracts.
Because exclusive dealing is very likely to increase dealer promotional efforts by creating dedicated dealers, antitrust analysis of exclusive dealing contracts becomes more difficult. The relatively easy cases where nothing is placed on the procompetitive justification side of the scale will be much rarer. However, recognizing that an exclusive dealing contract serves a procompetitive purpose does not mean that the contract cannot be anticompetitive. For example, while the exclusive dealing contracts used by Dentsply likely had a legitimate procompetitive purpose in encouraging dealer promotion, the Appeals Court found the contracts to have anticompetitive effects. But the more likely presence of procompetitive efficiency reasons for adopting exclusive dealing does mean that one cannot use the "no economic sense" test as a necessary condition for finding antitrust liability in exclusive dealing cases.(97)
However, before balancing the procompetitive and anticompetitive effects of exclusive dealing contracts, it is important to remember that balancing is the final step of the analysis. The first step is for plaintiffs to establish a significant anticompetitive effect of the exclusive contract. Recent developments in antitrust case law with regard to exclusive dealing as illustrated by Microsoft and Dentsply, where there was a finding of Section 2 but no Section 1 liability, does not mean that this first step of demonstrating a significant anticompetitive effect may be skipped.(98) Since an exclusive dealing contract may appear to inherently exclude rivals, or at the very least place rivals at a disadvantage, there is a danger that the absence of a procompetitive justification will always tip the competitive balance in favor of liability if we do not require demonstrating a significant anticompetitive effect. The "no economic sense" test would then become a defacto sufficient condition for antitrust liability.
What the analysis in this paper should teach us is that exclusive dealing contracts are more likely to be used as an element of the competitive process than previously believed. The fact that the expanded role of exclusive dealing in facilitating the enforcement of efficient distribution contracts we present in this paper was unrecognized suggests that there may very well be as yet undiscovered other legitimate economic purposes served by exclusive dealing contracts.(99) Therefore, before concluding that an exclusive contract should be condemned because it does not involve "competition not on the merits", we must unambiguously establish that the contract disturbs the competitive process by anticompetitively foreclosing a significant share of distribution.
* Klein: Professor Emeritus of Economics, UCLA and Director, LECG, LLC; Lerner: Principal, LECG, LLC.
1. "The recent decisions uniformly favor defendants where foreclosure levels are 40percent or less." Jonathan M. Jacobson, Exclusive Dealing, "Foreclosure," and Consumer Harm, 70 Antitrust L. J. 311, 362 (2002), citing cases at 325, n.85. Moreover, exclusive dealing arrangements covering even greater shares of the market have been routinely upheld if the contracts are relatively short-term, with a number of courts concluding that exclusive contracts covering one year or less are presumptively lawful. See, e.g., Roland Machinery v. Dresser Industries, 749 F.2d 380 (7th Cir. 1984); Omega Environmental Inc. v. Gilbarco, Inc., 127 F.3d 1157 (9th Cir. 1997); Concord Boat Corp. v. Brunswick Corp., 207 F.3d 1039 (8th Cir. 2000).
2. United States v. Microsoft Corp., 87 F. Supp. 2d 30 (D.D.C. 2000), 253 F.3d 34 (D.C. Cir. 2001) (en banc).
3. Microsoft, 87 F.Supp. 2d at 52. The D.C. Circuit signaled its disagreement, noting that "The District Court appears to have based its holding with respect to §1 upon a 'total exclusion test' rather than the 40% standard drawn from the case law." Microsoft, 253 F. 3d at 70. But the D.C. Circuit did not reverse the Section 1 ruling, which was not appealed by the plaintiffs.
4. "Microsoft's only explanation for its exclusive dealing is that it wants to keep developers focused upon its APIs [Windows application program interfaces]-- which is to say, it wants to preserve its power in the operating system market. _ That is not an unlawful end, but neither is it a procompetitive justification for the specific means here in question, namely exclusive dealing contracts with IAPs [Internet access providers]." Microsoft, 253 F. 3d at 71.
5. Id. at 70-71.
6. The D.C. Circuit defined "competition on the merits" as competition that "involves, for example, greater efficiency or enhanced consumer appeal." Id. at 59.
7. United States v. Dentsply Int'l, Inc. 277 F. Supp. 2d 387 (D. Del. Aug.6, 2003), 399 F.3d 181 (3d Cir. 2005), cert. denied, 126 S. Ct. 1023 (2006).
8. Dentsply, 277 F. Supp. 2d at 423.
9. For consistency of exposition throughout the paper we refer to the supplier of the product in question as "the manufacturer" and, in most cases, the purchaser of the manufacturer's product as "the dealer."
10. Id. at 453.
11. Dentsply, 399 F.3d at 193, 196-197. The Appeals Court discounted the possibility that rival manufacturers could compete effectively by relying on direct sales to dental laboratories. Id. at 193. Sales by Dentsply's two primary direct-selling competitors comprised only 8% of the market. Id. at 193. Moreover, the court concluded that, although dealers theoretically could terminate their relationship with Dentsply "at will" and switch to a competing line of artificial teeth, "the dealers have a strong economic incentive to continue carrying Dentsply's teeth." Id. at 193-194. The Appeals Court noted that "[a]lthough rivals could theoretically convince a dealer to buy their products and drop Dentsply's line, that has not occurred." In the more than ten years that Dentsply had employed exclusive dealing prior to the DOJ challenge, no dealer had dropped Dentsply in favor of rival brands of artificial teeth. Id. at 193-94. When a firm produces an essential product that makes rival entry difficult, the use of short-term contracts may not be a sufficient condition for the absence of an anticompetitive effect.
12. HowardP. Marvel, Exclusive Dealing, 25 J.L. & Econ. 1 (1982).
13. Dentsply, 277 F. Supp. 2d at 442-446.
14. This rationale was not presented by Marvel, but in Dentsply answers to interrogatories (GX 157 at Interrogatory Response No. 13) and by a Dentsply executive (D.I. 429 at 1719-20). Dentsply, 277 F. Supp. 2d at 440-441.
15. See infra notes, and .
16. Dentsply, 277 F. Supp. 2d at 441.
18. A number of other justifications for exclusive dealing are also shown to make economic sense within our more realistic economic framework. For example, Jacobson, supra note, at 357-360 lists nine economic justifications that have at times been accepted by courts for exclusive dealing contracts, with the prevention of free-riding and the creation of dedicated dealers as his first two justifications. However, a number of his other justifications are economic variants of these first two rationales. In particular, Jacobson lists quality assurance and the prevention of dealers passing off an inferior product in place of the manufacturer's product as his third justification for exclusive dealing. But the role of exclusive dealing in such cases is to prevent dealer free-riding on the manufacturer's brand name investment by switching buyers to rival products when the buyer is unaware that switching is occurring. (See discussion of this type of free-riding at infra note.) Jacobson also lists reducing the costs of monitoring dealer performance as a separate rationale for exclusive dealing, but the cause of dealer non-performance in these cases often involves the same types of free-riding we discuss, and exclusive dealing serves the same economic purpose of reducing the manufacturer's costs of detecting dealer nonperformance, discussed infra at note. In addition, Jacobson lists the role of exclusive dealing in decreasing dealer "out-of-stocks", which involves the same dealer-manufacturer incentive incompatibilities with regard to dealer supply of promotion and exclusive dealing solves this problem in the same way by creating dedicated dealers.
19. Marvel, supra note at 2, 6-8.
20. In re Beltone Elec. Corp., 100 F.T.C. 68 (1982).
21. Id. at 270. Beltone's exclusive territories were enforced in part by refusing to issue warranties submitted by dealers on sales to consumers outside their assigned territories. See Howard P. Marvel, Vertical Restraints in the Hearing Aids Industry in Impact Evaluations of Federal Trade Commission Vertical Restraint Cases, Ronald N. Lafferty, Robert H. Lande and John B. Kirkwood, eds., Federal Trade Commission, Washington, D.C. (1984), 270-384 at 280.
22. Beltone, 100 F.T.C. at 47, 201-202. The FTC began an investigation in 1970 of these distribution arrangements, used by a number of hearing aid manufacturers, that resulted in a series of actions brought in 1973 against several of the leading hearing aid manufacturers, including Dahlberg Electronics, Maico Hearing Instruments, Sonotone Corporation and Radioear Corporation, in addition to Beltone Electronics. All the other companies reached consent agreements between 1973 and 1976. Beltone, the largest manufacturer with approximately a 20percent market share, chose not to enter into a consent agreement and litigated to a successful conclusion in 1982.
23. Ryko Mfg. Co. v. Eden Servs., 823 F.2d 1215 (8th Cir. 1987).
24. Id. at 1218.
25. Id. at 1219-1220.
26. Id. at 1219-1220.
27. Id. at 1220.
28. Id. at 1219.
29. Id. at 1221, 1230.
30. Id. at 1221.
31. Jacobson, supra note , at 358.
32. In addition to the loss of profit to the manufacturer on any sales that are switched, in these circumstances there also is likely to be the cost to the manufacturer of a loss to its reputation (and reduced future sales) when the customer receives an inferior product that it believes is the manufacturer's product.
33. Marvel, supra note at 7.
34. Beltone, 100 F.T.C. at 285-287, 292. The FTC also found that Beltone's exclusive contracts did not foreclose distribution and that there was significant inter-brand competition. Id. at 290-291.
35. Ryko, 823 F.2d at 1234-1235, n.17. In addition to finding Ryko's exclusive contracts procompetitive, the court found that the exclusive did not foreclose competition since there was "no evidence suggesting that Ryko's exclusive dealing provisions generally prevent Ryko's competitors from finding effective distributors for (or other means of promoting and selling) their products." Id. at 1234.
36. Dentsply, 277 F. Supp. 2d at 445. The district court cited Marvel's article, stating that "[t]he term 'purely brand specific' is derived from Prof. Marvel's 1982 paper describing his theory, where he wrote: 'This argument does not apply if the promotional investment is purely brand specific. In such cases, the dealer will not be in a position to switch customers from brand to brand.'" Id.
37. Marvel used the example of Dentsply's promotion of "Portrait" and other new premium products. Dentsply , 277 F. Supp. 2d at 442.
38. Id. at 443-445 ("There are 'zero examples' in the record of these dealers steering customers from one brand to another").
39. Id. at 445-446.
40. Dentsply, 277 F. Supp. 2d at 441.
41. The following analysis is taken from Benjamin Klein & Joshua Wright, The Economics of Slotting Contracts, J. Law & Econ., forthcoming (August 2007), which is an extension of the original statement of the inadequate retailer incentive to promote (and an analysis of the role of various vertical restraints in solving this problem) presented in Benjamin Klein & Kevin M. Murphy, Vertical Restraints as Contract Enforcement Mechanisms, 31 J. L. & Econ. 265 (1988).
42. This is the inter-dealer free-riding problem described in Lester G. Telser, Why Should Manufacturers Want Fair Trade?, 3 J.L. & Econ. 86 (1960), the avoidance of which is the focus of the rationale for non-price vertical restraints in Continental T.V., Inc. v. GTE Sylvania Inc. 433 U.S. 36 (1977).
43. Marvel, supra note, at 4.
44. Marvel, supra note at 5. ("The free-rider problems facing exclusive and multiline dealers are identical.")
45. Dentsply, 277 F. Supp. 2d at 441; Marvel, supra note at 5.
46. This does not imply that these manufacturers possess any antitrust market power, in the sense of the ability to affect market prices. Almost every firm operating in the economy, except perhaps the wheat farmer described in introductory economics textbooks, faces a negatively sloped demand because it is producing a somewhat unique, differentiated product and, hence, charges a price greater than marginal cost. See Benjamin Klein, Market Power in Antitrust: Economic Analysis After Kodak, 3S. Ct. Econ. Rev. 43 (1993).
47. See Klein and Wright, supra note , where it is shown that, in equilibrium, the dealer's quantity response to its price reduction multiplied by the dealer's profit margin will equal the manufacturer's quantity response to a lower price multiplied by the manufacturer's profit margin (Klein and Wright, equation(5)). This implies that there will be close to the optimal amount of dealer price competition when there is intense competition at the dealer level.
48. Although we refer to dealer promotion of a manufacturer's products as inter-brand promotion because it primarily has inter-brand effects, such dealer promotion may also increase sales of a manufacturer without decreasing sales of its competitors (i.e. may increase total industry sales).
49. A dealer's provision of inter-brand promotion may have some inter-dealer effects. For instance, a basic product demonstration may be demanded by consumers and could have some inter-dealer effects. Consumers may not shop at dealers that do not demonstrate the product at all (unless the consumers engage in inter-dealer free riding). However, we refer to inter-brand promotional services as the increased demonstration time spent by a dealer on a manufacturer's product.
50. In fact, the common usage definition of promotion is something that is provided free of charge; if the dealer service were separately demanded and priced, it generally would not be called promotion. If consumers demanded these promotional services and were willing to pay for them, full-service dealers could prevent the inter-dealer free-riding currently emphasized in the economics and law (supra note) by charging consumers directly for the services, so that consumers would have no incentive to visit a full-service dealer and obtain the services free of charge before using these services to purchase the product at a low-service dealer. This notion seems strange because we do not observe such arrangements in the marketplace. However, it is not because of the transaction costs associated with separate pricing that we do not observe such arrangements, but because consumers generally would not pay for the services since they are promotional services aimed at marginal consumers.
51. Standard Fashion Co. v. Magrane-Houston Co., 258 U.S. 346 (1922).
52. Standard Fashion had a 40percent share of sales. Standard Fashion, 258 U.S. at 357. The case involved the attempt by Magrane-Houston (a dry-goods retailer in Boston) to substitute Standard Fashion's line of dress patterns for the products of another full-line pattern manufacturer, McCall. Standard's exclusive contracts were not an economic obstacle to McCall or other rival manufacturer competition for distribution since the contracts were two years in duration and the particular contract with Magrane-Houston had already been running for four years. Standard Fashion sued to enforce its exclusive contract solely because Magrane-Houston had failed to provide the contractually required three-months notice of termination. (Standard Fashion, 258 U.S. at 351-53.) Magrane-Houston's full compliance with the terms of the exclusive clearly would not have prevented its switch from Standard Fashion to McCall.
53. Standard Fashion, 258 U.S. at 351-352.
54. Standard Fashion, 258 U.S., Pls. Ex.7, Contract, Nov.25, 1914, R. at 131.
55. Butterick Publ'g Co. v. William G. Fisher, 203 Mass. 122, 131 (1909). Butterick was the owner of Standard Fashion.
56. Magrane-Houston committed to purchase and have on hand at all times $1,000 worth of Standard patterns, measured at net invoice prices which were 50percent of retail prices. This amounted to in excess of 10,000 patterns. Standard Fashion credited Magrane-Houston at 90percent of its cost for unsold, returned patterns that were exchanged for new stock. Standard Fashion, 258 U.S. at 352.
57. For instance, in Ryko, the court described that "the distributor's promotional efforts can be essential to the completion of individual [National Account Program (NAP)] sales [w]hile an oil company might designate Ryko an approved equipment supplier as the result of a national sales presentation, many NAP sales cannot be completed until the distributor has convinced the local purchaser that installing Ryko car-wash equipment at his location is a profitable idea." Ryko, 823 F.2d at 1220. In Beltone, dealers were supposed to follow up on sales leads with "a personal call upon the responding customer and to provide him or her with testing and information about hearing impairment and hearing aids. The dealer also requests that the person come to his shop for more thorough fitting of a suitable Beltone hearing aid." Beltone, 100 F.T.C. at. 202.
58. Stewart Macauley, Non-Contractual Relations in Business: A Preliminary Study, 28 Amer. Soc. Rev. 55 (1963) documents the frequent use of such a self-enforcement (that is, non-court enforced) mechanism to assure transactor performance in many contractual arrangements.
59. See, for example, Benjamin Klein and Keith B. Leffler, The Role of Market Forces in Assuring Contractual Performance, 89 J. Pol. Econ. 615 (1981).
60. Klein and Murphy, supra note.
61. Klein and Murphy, supra note. This role of resale price maintenance as a way for the manufacturer to provide sufficient compensation to dealers for the supply of increased promotion explains the use of resale price maintenance in the distribution of many products where the inter-dealer free-riding explanation described in Telser, supra note, and accepted by the Court in Sylvania, 433 U.S. at 55 as a procompetitive rationale for vertical restraints, does not appear applicable. See Benjamin Klein, Distribution Restrictions Operate By Creating Dealer Profits: Explaining the Use of Maximum Resale Price Maintenance in State Oil v. Kahn, 7 S.Ct. Econ. Rev. 1, 7-8 (1999).
62. A systematic historical survey of exclusive dealing contracts indicates that resale price maintenance frequently has been used in conjunction with exclusive dealing. See Thomas R. Overstreet, Jr., Resale Price Maintenance: Economic Theories and Empirical Evidence, Bureau of Economics Staff Report to the Federal Trade Commission, 84-101 (1983).
63. Standard Fashion, 258 U.S. at 352.
64. Marvel, supra note, justifies exclusivity in Standard Fashion as a way to protect Standard Fashion's intellectual property investments in the creation of dress pattern designs, which can easily be copied by rival manufacturers. While exclusive dealing does not prevent copying of a manufacturer's successful patterns by other full-line manufacturers, such as McCall, or by groups of limited-line manufacturers that together can supply retailers with a full-line of patterns, exclusive dealing prevents the retailer from switching consumers that have chosen a Standard Fashion design from a pattern catalog or pamphlet to the copy.
65. This is primarily because of the difficulty of more directly measuring and contracting for dealer promotional effort. If dealer promotion increases the dealer's sales a particular percentage, the usual vertical restraint contractual compensation arrangement provides a reasonable dealer compensation measure across dealers and over time.
66. Because dealers are compensated on total sales they also have an incentive to violate the implicit promotion contract by not following resale price maintenance. Dealers that supply the desired promotional services but lower price and increase sales to infra-marginal consumers who would purchase from other dealers are overcompensated for supplying the desired level of promotion. Meanwhile, other dealers will be under-compensated for supplying the desired promotion, leading them to reduce their promotional efforts. This incentive by dealers to engage in price competition exists even if all dealers are supplying the contracted for promotion and are not engaging in the Sylvania type of inter-dealer free-riding analyzed. See supra note.
67. Eden then challenged the termination by suing Ryko for its exclusive contract on Section 1 grounds. Supra note.
68. Beltone, 100 F.T.C. at 59; Dentsply, 277 F. Supp. 2d at 414-415, 420. Similar enforcement of exclusive dealing occurred in Roland Machinery, discussed infra note, where the manufacturer terminated its dealer, Roland Machinery, after it moved to non-exclusive dealing. Standard Fashion did not involve manufacturer termination of a dealer, but dealer termination of the manufacturer, supra note.
69. See Benjamin Klein & Lester Saft, The Law and Economics of Franchise Tying Contracts, 28 J. L. & Econ. 345 (1985) for a discussion of this reduced monitoring cost rationale for exclusive input requirements contracts in franchise arrangements.
70. The manufacturer cannot collect the higher profit premium that would have to be paid to dealers in the absence of exclusive dealing with an initial lump sum payment from the dealer because this would make it legally difficult for the manufacturer to terminate dealers for nonperformance (without returning the initial fee, which would eliminate the dealer sanction implied by termination). See Benjamin Klein, The Economics of Franchise Contracts, 2 J. Corp. Fin. 9 (1995) at 28-30.
71. Roland Machinery Co. v. Dresser Industries, 749 F.2d 380 (7th Cir. 1984).
72. Id. at 381.
73. Id. at 381-382.
74. The Appeals Court in Roland Machinery held that there was no evidence of an exclusive dealing agreement between Dresser and Roland (or any other distributor). Although there was a clear desire by Dresser for distributor exclusivity, the fact that Roland openly obtained a Komatsu dealership indicated to the Court that Roland did not believe it had made a commitment to exclusivity. While Dresser preferred exclusivity and, in fact, was extremely hostile to non-exclusive arrangements as evidenced by its termination of Roland, the Court held that there was no "meeting of minds" and, therefore, no agreement. (Roland Machinery, 749 F.2d at 392-3). In addition to denying the existence of an exclusive contract, the court concluded that, even if such a contract existed, it would not have been anticompetitive. Dresser Industries manufactured only 16 or 17percent of the construction equipment sold in Roland's territory of central Illinois. Id. at 382. The court also concluded that because Dresser's contracts were short-term and could not foreclose Komatsu, the second largest manufacturer of construction equipment in the world, from obtaining distribution. In fact, Komatsu had already become a major factor in the U.S. market. Id. at 393-395.
75. Roland Machinery, 749 F.2d at 395.
77. Joyce Beverages v. Royal Crown Cola Co., 555 F. Supp. 271 (S.D.N.Y. 1983).
78. Id. at 273-74.
79. Id. at 278. Royal Crown Cola bottlers were paid for extra promotional efforts, in part, with the grant of a valuable exclusive territory (id.) that would be lost if Royal Crown terminated a bottler for its insufficient efforts to "compete for shelf space, display racks, promotional rotations and the placement of feature advertising." Id. at 275. The exclusive dealing contract requirement clearly had nothing to do with anticompetitive foreclosure because Royal Crown Cola had only a 5percent share of U.S. cola sales. Id. at 273.
80. Hendricks Music Co. v. Steinway, Inc., 689 F. Supp. 1501 (N.D. Ill. 1988) at 1514, 1545-48.
81. Phillip Kotler, Marketing Management, 11th ed., 2003, at 513.
82. Marvel, supra note, at 3-5. See supra note.
83. "The free-rider problems facing exclusive and multiline dealers are identical." Marvel, supra note at 5.
84. Marvel, supra note, at 4. An indication of the impact of Marvel's analysis can be obtained by examining economics textbooks. An undivided loyalty rationale for exclusive dealing was included in some economics textbooks before Marvel's article. For instance, F.M. Scherer, Industrial Market Structure and Economic Performance (1980) at 586 states that "[f]or manufacturers, exclusive dealing arrangements are often appealing, because they ensure that their products will be merchandised with maximum energy and enthusiasm." This rationale for exclusive dealing was removed from later editions of the same textbook. See F.M. Scherer and David Ross, Industrial Market Structure and Economic Performance, Third Edition (1990).
85. See Dentsply, 277 F. Supp. 2d, GX. 157 at Interrogatory Response No. 13 (p. 12), cited in DOJ Proposed Findings of Fact at 332 ("In Dentsply's experience, the greater the number of competing tooth lines carried, the less likely that a dealer will be able to sustain all of the desired services and promotional elements at a high competitive level. In short, service and promotional support for a particular line is likely to suffer the greater the number of lines carried").
86. Dentsply, 277 F. Supp. 2d at 441.
88. Roland Machinery, 749 F.2d at 395.
91. Roland Machinery, 749 F.2d at 395, citing Sulmeyer v. Coca-Cola Co., 515 F.2d 835, 840 n.2 (5th Cir. 1975).
92. While many consumers may find it economic to shop multiple dealers before purchasing because of the significant size of the purchase, profitable after sale service (including warranty work) will often be supplied by the most conveniently located dealer. Therefore, the grant of a limited exclusive territory will be a profitable asset.
93. This is why instances where an automobile manufacturer permits a dual dealership (for example, in small towns), the manufacturer prefers that the two brands be in different segments of the market, so as not to be in direct competition with one another, and that the different brands have distinct sales staffs.
94. The shape of this curve is likely to be convex, which will be the case if, for example, there is declining effectiveness of promotional expenditures, p"(S) < 0, and the marginal cost of S is linear, so that C"(S) = 0.
95. "Dealers are an important conduit for supplier's promotional message." Dentsply, 277 F. Supp. 2d, United States' Brief in Support of its Proposed Findings of Fact and Conclusions of Law at 85 "Tooth dealers promote suppliers' products." Id. at 24. Because dealers assist laboratories in choosing teeth, they have significant ability to steer laboratories to a particular brand of artificial teeth. Id. Dealers can "assist suppliers in generating incremental business by promoting the manufacturer's product and providing these other services." Id. at 90. These conclusions were accepted by the Appeals Court. Dentsply, 277 F. Supp. 2d at 28.
96. Dentsply, 277 F. Supp. 2d at 445-446.
97. A "no economic sense" test is advocated by Gregory J. Werden, Identifying Exclusionary Conduct Under Section 2: The "No Economic Sense" Test, 73 Antitrust L. J. 413 (2006) and A. Douglas Melamed, Exclusive Dealing Agreements and Other Exclusionary Conduct- Are There Unifying Principles?, 73 Antitrust L. J. 375 (2006). Jonathan M. Jacobson & Scott A. Sher, "No Economic Sense" Makes No Sense For Exclusive Dealing, 73 Antitrust L.J. 779 (2006) draw the same implication about the inappropriateness of the "no economic sense" test for exclusive dealing contracts. However, Jacobson and Sher assume that Dentsply's claimed undivided dealer loyalty justification makes economic sense, a proposition that we have demonstrated is correct but which the Dentsply court rejected for what appeared to be good economic reasons that Jacobson and Sher do not analyze.
98. It is fairly obvious from both Appeals Court decisions that the Section 1 district court rulings would have been overturned if appealed by the DOJ, which perhaps failed to appeal because it did not want to risk further weakening of Section 1 exclusive dealing case law.
99. Additional, commonly unrecognized justifications for exclusive dealing include the prevention of buyer holdups of a seller that has made specific investments, as illustrated by Tampa Electric Co. v. Nashville Coal Co., 365 U.S. 320 (1961). See Benjamin Klein, The Economic Lessons of Fisher Body-General Motors, unpublished ms., 2006. (Jacobson, supra note, claims that the exclusive in Tampa Electric was used to provide assured supply to a buyer. However, an exclusive is a constraint on the buyer, not the seller. If it was the buyer, Tampa, that required assurances of supply rather than the seller, Nashville, that required assurances of demand, all that would have been needed would have been a requirements contract committing Nashville to supply, with perhaps some minimum purchase commitment by Tampa.) In addition, Benjamin Klein and KevinM. Murphy, Exclusive Dealing Intensifies Competitive Bidding for Distribution, unpublished working paper (2006), show that exclusive dealing may be used to increase competition by manufacturers for distribution when dealers effectively act as bargaining agents for ultimate consumers and lower price by efficiently internalizing each consumer's independent buying decision.