Program in Public Law

Eastman Kodak Co. v. Image Technical Services, et al.

From Professor John Bowen’s web page, “Supreme Court Antitrust Debates” at Ripon College.
Web site: http://www.ripon.edu/Faculty/bowenj/antitrust/kdakvits.htm

(View printer-friendly version)

(This document has been edited from the text of the original Supreme Court opinion. It is intended primarily for use in college and university courses in antitrust and public policy economics. While every effort has been made to assure that the edited version is faithful to the original, users should be aware that most references have been omitted and that certain editing liberties have been taken to enhance readability.)

Justice BLACKMUN delivered the opinion of the Court.

[1]This is yet another case that concerns the standard for summary judgmentin an antitrust controversy. The principal issue here is whether a defendant'slack of market power in the primary equipment market precludes -- as amatter of law -- the possibility of market power in derivative aftermarkets.

[2] Petitioner Eastman Kodak Company manufactures and sells photocopiersand micrographic equipment. Kodak also sells service and replacement partsfor its equipment. Respondents are 18 independent service organizations(ISOs) that in the early 1980s began servicing Kodak copying and micrographicequipment. Kodak subsequently adopted policies to limit the availabilityof parts to ISOs and to make it more difficult for ISOs to compete withKodak in servicing Kodak equipment.

[3] Respondents instituted this action in the United States DistrictCourt for the Northern District of California alleging that Kodak's policieswere unlawful under both Sections 1 and 2 of the Sherman Act. After truncateddiscovery, the District Court granted summary judgment for Kodak. The Courtof Appeals for the Ninth Circuit reversed. The appellate court found thatrespondents had presented sufficient evidence to raise a genuine issueconcerning Kodak's market power in the service and parts markets. It rejectedKodak's contention that lack of market power in service and parts mustbe assumed when such power is absent in the equipment market. Because ofthe importance of the issue, we granted certiorari.

[4] Because this case comes to us on petitioner Kodak's motion for summaryjudgment, ``the evidence of [respondents] is to be believed, and all justifiableinferences are to be drawn in [their] favor.'' Mindful that respondents'version of any disputed issue of fact thus is presumed correct, we beginwith the factual basis of respondents' claims.

[5] Kodak manufactures and sells complex business machines -- as relevanthere, high-volume photocopier and micrographics equipment.⁽¹⁾Kodak equipment is unique; micrographic software programs that operateon Kodak machines, for example, are not compatible with competitors' machines.Kodak parts are not compatible with other manufacturers' equipment, andvice versa. Kodak equipment, although expensive when new, has little resalevalue.

[6] Kodak provides service and parts for its machines to its customers.It produces some of the parts itself; the rest are made to order for Kodakby independent original-equipment manufacturers (OEMs). Kodak does notsell a complete system of original equipment, lifetime service, and lifetimeparts for a single price. Instead, Kodak provides service after the initialwarranty period either through annual service contracts, which includeall necessary parts, or on a per-call basis. It charges, through negotiationsand bidding, different prices for equipment, service, and parts for differentcustomers. Kodak provides 80% to 95% of the service for Kodak machines.

[7] Beginning in the early 1980s, ISOs began repairing and servicingKodak equipment. They also sold parts and reconditioned and sold used Kodakequipment. Their customers were federal, state, and local government agencies,banks, insurance companies, industrial enterprises, and providers of specializedcopy and microfilming services. ISOs provide service at a price substantiallylower than Kodak does. Some customers found that the ISO service was ofhigher quality. Some of the ISOs' customers purchase their own parts andhire ISOs only for service. Others choose ISOs to supply both service andparts. ISOs keep an inventory of parts, purchased from Kodak or other sources,primarily the OEMs.⁽²⁾

[8] In 1985 and 1986, Kodak implemented a policy of selling replacementparts for micrographic and copying machines only to buyers of Kodak equipmentwho use Kodak service or repair their own machines. As part of the samepolicy, Kodak sought to limit ISO access to other sources of Kodak parts.Kodak and the OEMs agreed that the OEMs would not sell parts that fit Kodakequipment to anyone other than Kodak. Kodak also pressured Kodak equipmentowners and independent parts distributors not to sell Kodak parts to ISOs.In addition, Kodak took steps to restrict the availability of used machines.Kodak intended, through these policies, to make it more difficult for ISOsto sell service for Kodak machines. It succeeded. ISOs were unable to obtainparts from reliable sources, and many were forced out of business, whileothers lost substantial revenue. Customers were forced to switch to Kodakservice even though they preferred ISO service.

[9] In 1987, the ISOs filed the present action in the District Court,alleging, inter alia, that Kodak had unlawfully tied the sale ofservice for Kodak machines to the sale of parts, in violation of Section1 of the Sherman Act, and had unlawfully monopolized and attempted to monopolizethe sale of service for Kodak machines, in violation of Section 2 of thatAct.⁽³⁾ Kodak fileda motion for summary judgment before respondents had initiated discovery.The District Court permitted respondents to file one set of interrogatoriesand one set of requests for production of documents, and to take six depositions.Without a hearing, the District Court granted summary judgment in favorof Kodak.

[10] The Court of Appeals for the Ninth Circuit, by a divided vote,reversed.

[11] A tying arrangement is ``an agreement by a party to sell one productbut only on the condition that the buyer also purchases a different (ortied) product, or at least agrees that he will not purchase that productfrom any other supplier.'' Such an arrangement violates Section 1 of theSherman Act if the seller has ``appreciable economic power'' in the tyingproduct market and if the arrangement affects a substantial volume of commercein the tied market.

[12] Kodak did not dispute that its arrangement affects a substantialvolume of interstate commerce. It, however, did challenge whether its activitiesconstituted a ``tying arrangement'' and whether Kodak exercised ``appreciableeconomic power'' in the tying market. We consider these issues in turn.

[13] For the respondents to defeat a motion for summary judgment ontheir claim of a tying arrangement, a reasonable trier of fact must beable to find, first, that service and parts are two distinct products,and, second, that Kodak has tied the sale of the two products.

[14] For service and parts to be considered two distinct products, theremust be sufficient consumer demand so that it is efficient for a firm toprovide service separately from parts. Evidence in the record indicatesthat service and parts have been sold separately in the past and stillare sold separately to self-service equipment owners. Indeed, the developmentof the entire high-technology service industry is evidence of the efficiencyof a separate market for service.

[15] Kodak insists that because there is no demand for parts separatefrom service, there cannot be separate markets for service and parts. Bythat logic, we would be forced to conclude that there can never be separatemarkets, for example, for cameras and film, computers and software, orautomobiles and tires. That is an assumption we are unwilling to make.``We have often found arrangements involving functionally linked productsat least one of which is useless without the other to be prohibited tyingdevices.''

[16] Kodak's assertion also appears to be incorrect as a factual matter.At least some consumers would purchase service without parts, because someservice does not require parts, and some consumers, those who self-servicefor example, would purchase parts without service. Enough doubt is caston Kodak's claim of a unified market that it should be resolved by thetrier of fact.

[17] Finally, respondents have presented sufficient evidence of a tiebetween service and parts. The record indicates that Kodak would sell partsto third parties only if they agreed not to buy service from ISOs.

[18] Having found sufficient evidence of a tying arrangement, we considerthe other necessary feature of an illegal tying arrangement: appreciableeconomic power in the tying market. Market power is the power ``to forcea purchaser to do something that he would not do in a competitive market.''<sup>(4)</sup>It has been defined as ``the ability of a single seller to raise priceand restrict output.'' The existence of such power ordinarily is inferredfrom the seller's possession of a predominant share of the market.

[19] Respondents contend that Kodak has more than sufficient power inthe parts market to force unwanted purchases of the tied market, service.Respondents provide evidence that certain parts are available exclusivelythrough Kodak. Respondents also assert that Kodak has control over theavailability of parts it does not manufacture. According to respondents'evidence, Kodak has prohibited independent manufacturers from selling Kodakparts to ISOs, pressured Kodak equipment owners and independent parts distributorsto deny ISOs the purchase of Kodak parts, and taken steps to restrict theavailability of used machines.

[20] Respondents also allege that Kodak's control over the parts markethas excluded service competition, boosted service prices, and forced unwillingconsumption of Kodak service. Respondents offer evidence that consumershave switched to Kodak service even though they preferred ISO service,that Kodak service was of higher price and lower quality than the preferredISO service, and that ISOs were driven out of business by Kodak's policies.Under our prior precedents, this evidence would be sufficient to entitlerespondents to a trial on their claim of market power.

[21] Kodak counters that even if it concedes monopoly share ofthe relevant parts market, it cannot actually exercise the necessary marketpower for a Sherman Act violation. This is so, according to Kodak,because competition exists in the equipment market.⁽⁵⁾Kodak argues that it could not have the ability to raise prices of serviceand parts above the level that would be charged in a competitive marketbecause any increase in profits from a higher price in the aftermarketsat least would be offset by a corresponding loss in profits from lowerequipment sales as consumers began purchasing equipment with more attractiveservice costs.

[22] Kodak does not present any actual data on the equipment, service,or parts markets. Instead, it urges the adoption of a substantive legalrule that ``equipment competition precludes any finding of monopoly powerin derivative aftermarkets.'' Kodak argues that such a rule would satisfyits burden as the moving party of showing ``that there is no genuine issueas to any material fact'' on the market power issue.

[23] Kodak contends that there is no need to examine the facts whenthe issue is market power in the aftermarkets. A legal presumption againsta finding of market power is warranted in this situation, according toKodak, because the existence of market power in the service and parts marketsabsent power in the equipment market ``simply makes no economic sense,''and the absence of a legal presumption would deter procompetitive behavior.

[24] Kodak bears a substantial burden in showing that it is entitledto summary judgment. It must show that despite evidence of increased pricesand excluded competition, an inference of market power is unreasonable.To determine whether Kodak has met that burden, we must unravel the factualassumptions underlying its proposed rule that lack of power in the equipmentmarket necessarily precludes power in the aftermarkets.

[25] The extent to which one market prevents exploitation of anothermarket depends on the extent to which consumers will change their consumptionof one product in response to a price change in another, i.e., the``cross-elasticity of demand.''<sup>(6)</sup>Kodak's proposed rule rests on a factual assumption about the cross-elasticityof demand in the equipment and aftermarkets: ``If Kodak raised its partsor service prices above competitive levels, potential customers would simplystop buying Kodak equipment. Perhaps Kodak would be able to increase shortterm profits through such a strategy, but at a devastating cost to itslong term interests.''⁽⁷⁾Kodak argues that the Court should accept, as a matter of law, this ``basiceconomic realit[y],'' that competition in the equipment market necessarilyprevents market power in the aftermarkets.⁽⁸⁾

[26] Even if Kodak could not raise the price of service and parts onecent without losing equipment sales, that fact would not disprove marketpower in the aftermarkets. The sales of even a monopolist are reduced whenit sells goods at a monopoly price, but the higher price more than compensatesfor the loss in sales. Kodak's claim that charging more for service andparts would be ``a short-run game,'' is based on the false dichotomy thatthere are only two prices that can be charged--a competitive price or aruinous one. But there could easily be a middle, optimum price at whichthe increased revenues from the higher-priced sales of service and partswould more than compensate for the lower revenues from lost equipment sales.The fact that the equipment market imposes a restraint on prices in theaftermarkets by no means disproves the existence of power in those markets.(``The existence of significant substitution in the event of furtherprice increases or even at the current price does not tell uswhether the defendant already exercises significant market power'').Thus, contrary to Kodak's assertion, there is no immutable physical law-- no ``basic economic reality''-- insisting that competition in the equipmentmarket cannot coexist with market power in the aftermarkets.

[27] We next consider the more narrowly drawn question: Does Kodak'stheory describe actual market behavior so accurately that respondents'assertion of Kodak market power in the aftermarkets, if not impossible,is at least unreasonable?

[28] To review Kodak's theory, it contends that higher service priceswill lead to a disastrous drop in equipment sales. Presumably, the theory'scorollary is to the effect that low service prices lead to a dramatic increasein equipment sales. According to the theory, one would have expected Kodakto take advantage of lower-priced ISO service as an opportunity to expandequipment sales. Instead, Kodak adopted a restrictive sales policy consciouslydesigned to eliminate the lower-priced ISO service, an act that would beexpected to devastate either Kodak's equipment sales or Kodak's faith inits theory. Yet, according to the record, it has done neither. Serviceprices have risen for Kodak customers, but there is no evidence or assertionthat Kodak equipment sales have dropped.

[29] Kodak and the United States attempt to reconcile Kodak's theorywith the contrary actual results by describing a ``marketing strategy ofspreading over time the total cost to the buyer of Kodak equipment.'' Inother words, Kodak could charge subcompetitive prices for equipment andmake up the difference with supracompetitive prices for service, resultingin an overall competitive price. This pricing strategy would provide anexplanation for the theory's descriptive failings -- if Kodak in fact hadadopted it. But Kodak never has asserted that it prices its equipment orparts subcompetitively and recoups its profits through service. Instead,it claims that it prices its equipment comparably to its competitors, andintends that both its equipment sales and service divisions be profitable.Moreover, this hypothetical pricing strategy is inconsistent with Kodak'spolicy toward its self-service customers. If Kodak were underpricing itsequipment, hoping to lock in customers and recover its losses in the servicemarket, it could not afford to sell customers parts without service. Insum, Kodak's theory does not explain the actual market behavior revealedin the record.

[30] Respondents offer a forceful reason why Kodak's theory, althoughperhaps intuitively appealing, may not accurately explain the behaviorof the primary and derivative markets for complex durable goods: the existenceof significant information and switching costs. These costs could createa less responsive connection between service and parts prices and equipmentsales.

[31] For the service-market price to affect equipment demand, consumersmust inform themselves of the total cost of the ``package''--equipment,service and parts--at the time of purchase; that is, consumers must engagein accurate lifecycle pricing. Lifecycle pricing of complex, durable equipmentis difficult and costly. In order to arrive at an accurate price, a consumermust acquire a substantial amount of raw data and undertake sophisticatedanalysis. The necessary information would include data on price, quality,and availability of products needed to operate, upgrade, or enhance theinitial equipment, as well as service and repair costs, including estimatesof breakdown frequency, nature of repairs, price of service and parts,length of ``down-time'' and losses incurred from down-time.⁽⁹⁾

[32] Much of this information is difficult -- some of it impossible-- to acquire at the time of purchase. During the life of a product, companiesmay change the service and parts prices, and develop products with moreadvanced features, a decreased need for repair, or new warranties. In addition,the information is likely to be customer-specific; lifecycle costs willvary from customer to customer with the type of equipment, degrees of equipmentuse, and costs of downtime.

[33] Kodak acknowledges the cost of information, but suggests, againwithout evidentiary support, that customer information needs will be satisfiedby competitors in the equipment markets. It is a question of fact, however,whether competitors would provide the necessary information. A competitorin the equipment market may not have reliable information about the lifecyclecosts of complex equipment it does not service or the needs of customersit does not serve. Even if competitors had the relevant information, itis not clear that their interests would be advanced by providing such informationto consumers.⁽¹⁰⁾

[34] Moreover, even if consumers were capable of acquiring and processingthe complex body of information, they may choose not to do so. Acquiringthe information is expensive. If the costs of service are small relativeto the equipment price, or if consumers are more concerned about equipmentcapabilities than service costs, they may not find it cost-efficient tocompile the information. Similarly, some consumers, such as the FederalGovernment, have purchasing systems that make it difficult to considerthe complete cost of the ``package'' at the time of purchase. State andlocal governments often treat service as an operating expense and equipmentas a capital expense, delegating each to a different department. Thesegovernmental entities do not lifecycle price, but rather choose the lowestprice in each market.

[35] As Kodak notes, there likely will be some large-volume, sophisticatedpurchasers who will undertake the comparative studies and insist, in returnfor their patronage, that Kodak charge them competitive lifecycle prices.Kodak contends that these knowledgeable customers will hold down the packageprice for all other customers. There are reasons, however, to doubt thatsophisticated purchasers will ensure that competitive prices are chargedto unsophisticated purchasers, too. As an initial matter, if the numberof sophisticated customers is relatively small, the amount of profits tobe gained by supracompetitive pricing in the service market could makeit profitable to let the knowledgeable consumers take their business elsewhere.More importantly, if a company is able to price-discriminate between sophisticatedand unsophisticated consumers, the sophisticated will be unable to preventthe exploitation of the uninformed. A seller could easily price-discriminateby varying the equipment/parts/service package, developing different warranties,or offering price discounts on different components.

[36] Given the potentially high cost of information and the possibilitya seller may be able to price-discriminate between knowledgeable and unsophisticatedconsumers, it makes little sense to assume, in the absence of any evidentiarysupport, that equipment-purchasing decisions are based on an accurate assessmentof the total cost of equipment, service, and parts over the lifetime ofthe machine. Indeed, respondents have presented evidence that Kodak practicesprice-discrimination by selling parts to customers who service their ownequipment, but refusing to sell parts to customers who hire third-partyservice companies. Companies that have their own service staff are likelyto be high-volume users, the same companies for whom it is most likelyto be economically worthwhile to acquire the complex information neededfor comparative lifecycle pricing.

[37] A second factor undermining Kodak's claim that supracompetitiveprices in the service market lead to ruinous losses in equipment salesis the cost to current owners of switching to a different product.⁽¹¹⁾If the cost of switching is high, consumers who already have purchasedthe equipment, and are thus ``locked-in,'' will tolerate some level ofservice-price increases before changing equipment brands. Under this scenario,a seller profitably could maintain supracompetitive prices in the aftermarketif the switching costs were high relative to the increase in service prices,and the number of locked-in customers were high relative to the numberof new purchasers.

[38] Moreover, if the seller can price-discriminate between its locked-incustomers and potential new customers, this strategy is even more likelyto prove profitable. The seller could simply charge new customers below-marginalcost on the equipment and recoup the charges in service, or offer packageswith life-time warranties or long-term service agreements that are notavailable to locked-in customers.

[39] Respondents have offered evidence that the heavy initial outlayfor Kodak equipment, combined with the required support material that worksonly with Kodak equipment, makes switching costs very high for existingKodak customers. And Kodak's own evidence confirms that it varies the packageprice of equipment/parts/service for different customers.

[40] In sum, there is a question of fact whether information costs andswitching costs foil the simple assumption that the equipment and servicemarkets act as pure complements to one another.⁽¹²⁾We conclude, then, that Kodak has failed to demonstrate that respondents'inference of market power in the service and parts markets is unreasonable,and that, consequently, Kodak is entitled to summary judgment. It is clearlyreasonable to infer that Kodak has market power to raise prices and driveout competition in the aftermarkets, since respondents offer direct evidencethat Kodak did so. It is also plausible, as discussed above, to infer thatKodak chose to gain immediate profits by exerting that market power wherelocked-in customers, high information costs, and discriminatory pricinglimited and perhaps eliminated any long-term loss. Viewing the evidencein the light most favorable to respondents, their allegations of marketpower ``mak[e] . . . economic sense.''

[41] Nor are we persuaded by Kodak's contention that it is entitledto a legal presumption on the lack of market power because, as in Matsushita,there is a significant risk of deterring procompetitive conduct. Plaintiffsin Matsushita attempted to prove the antitrust conspiracy ``throughevidence of rebates and other price-cutting activities.'' Because cuttingprices to increase business is ``the very essence of competition,'' theCourt was concerned that mistaken inferences would be ``especially costly,''and would ``chill the very conduct the antitrust laws are designed to protect.''But the facts in this case are just the opposite. The alleged conduct--higherservice prices and market foreclosure--is facially anticompetitive andexactly the harm that antitrust laws aim to prevent. In this situation,Matsushita does not create any presumption in favor of summary judgmentfor the defendant.

[42] Kodak contends that, despite the appearance of anticompetitiveness,its behavior actually favors competition because its ability to pursueinnovative marketing plans will allow it to compete more effectively inthe equipment market. A pricing strategy based on lower equipment pricesand higher aftermarket prices could enhance equipment sales by making iteasier for the buyer to finance the initial purchase.⁽¹³⁾It is undisputed that competition is enhanced when a firm is able to offervarious marketing options, including bundling of support and maintenanceservice with the sale of equipment. Nor do such actions run afoul of theantitrust laws. But the procompetitive effect of the specific conduct challengedhere, eliminating all consumer parts and service options, is far less clear.

[43] We need not decide whether Kodak's behavior has any procompetitiveeffects and, if so, whether they outweigh the anticompetitive effects.We note only that Kodak's service and parts policy is simply not one thatappears always or almost always to enhance competition, and therefore towarrant a legal presumption without any evidence of its actual economicimpact. In this case, when we weigh the risk of deterring procompetitivebehavior by proceeding to trial against the risk that illegal behaviorgo unpunished, the balance tips against summary judgment.

[44] For the foregoing reasons, we hold that Kodak has not met the requirementsof Fed. Rule Civ. Proc. 56(c). We therefore affirm the denial of summaryjudgment on respondents' Section 1 claim.⁽¹⁴⁾

[45] Respondents also claim that they have presented genuine issuesfor trial as to whether Kodak has monopolized or attempted to monopolizethe service and parts markets in violation of Section 2 of the ShermanAct. ``The offense of monopoly under Section 2 of the Sherman Act has twoelements: (1) the possession of monopoly power in the relevant market and(2) the willful acquisition or maintenance of that power as distinguishedfrom growth or development as a consequence of a superior product, businessacumen, or historic accident.''

[46] The existence of the first element, possession of monopoly power,is easily resolved. As has been noted, respondents have presented a triableclaim that service and parts are separate markets, and that Kodak has the``power to control prices or exclude competition'' in service and parts.Monopoly power under Section 2 requires, of course, something greater thanmarket power under Section 1. Respondents' evidence that Kodak controlsnearly 100% of the parts market and 80% to 95% of the service market, withno readily available substitutes, is, however, sufficient to survive summaryjudgment under the more stringent monopoly standard of Section 2.

[47] Kodak also contends that, as a matter of law, a single brand ofa product or service can never be a relevant market under the Sherman Act.We disagree. The relevant market for antitrust purposes is determined bythe choices available to Kodak equipment owners. Because service and partsfor Kodak equipment are not interchangeable with other manufacturers' serviceand parts, the relevant market from the Kodak-equipment owner's perspectiveis composed of only those companies that service Kodak machines. This Court'sprior cases support the proposition that in some instances one brand ofa product can constitute a separate market. The proper market definitionin this case can be determined only after a factual inquiry into the ``commercialrealities'' faced by consumers.

[48] The second element of a Section 2 claim is the use of monopolypower ``to foreclose competition, to gain a competitive advantage, or todestroy a competitor.'' If Kodak adopted its parts and service policiesas part of a scheme of willful acquisition or maintenance of monopoly power,it will have violated Section 2.

[49] As recounted at length above, respondents have presented evidencethat Kodak took exclusionary action to maintain its parts monopoly andused its control over parts to strengthen its monopoly share of the Kodakservice market. Liability turns, then, on whether ``valid business reasons''can explain Kodak's actions. Kodak contends that it has three valid businessjustifications for its actions: ``(1) to promote interbrand equipment competitionby allowing Kodak to stress the quality of its service; (2) to improveasset management by reducing Kodak's inventory costs; and (3) to preventISOs from free riding on Kodak's capital investment in equipment, partsand service.'' Factual questions exist, however, about the validity andsufficiency of each claimed justification, making summary judgment inappropriate.

[50] Kodak first asserts that by preventing customers from using ISOs,``it [can] best maintain high quality service for its sophisticated equipment''and avoid being ``blamed for an equipment malfunction, even if the problemis the result of improper diagnosis, maintenance or repair by an ISO.''Respondents have offered evidence that ISOs provide quality service andare preferred by some Kodak equipment owners. This is sufficient to raisea genuine issue of fact.

[51] Moreover, there are other reasons to question Kodak's profferedmotive of commitment to quality service; its quality justification appearsinconsistent with its thesis that consumers are knowledgeable enough tolifecycle price, and its self-service policy. Kodak claims the exclusive-servicecontract is warranted because customers would otherwise blame Kodak equipmentfor breakdowns resulting from inferior ISO service. Thus, Kodak simultaneouslyclaims that its customers are sophisticated enough to make complex andsubtle lifecycle-pricing decisions, and yet too obtuse to distinguish whichbreakdowns are due to bad equipment and which are due to bad service. Kodakhas failed to offer any reason why informational sophistication shouldbe present in one circumstance and absent in the other. In addition, becauseself-service customers are just as likely as others to blame Kodak equipmentfor breakdowns resulting from (their own) inferior service, Kodak's willingnessto allow self-service casts doubt on its quality claim. In sum, we agreewith the Court of Appeals that respondents ``have presented evidence fromwhich a reasonable trier of fact could conclude that Kodak's first reasonis pretextual.''

[52] There is also a triable issue of fact on Kodak's second justification--controllinginventory costs. As respondents argue, Kodak's actions appear inconsistentwith any need to control inventory costs. Presumably, the inventory ofparts needed to repair Kodak machines turns only on breakdown rates, andthose rates should be the same whether Kodak or ISOs perform the repair.More importantly, the justification fails to explain respondents' evidencethat Kodak forced OEMs, equipment owners, and parts brokers not to sellparts to ISOs, actions that would have no effect on Kodak's inventory costs.

[53] Nor does Kodak's final justification entitle it to summary judgmenton respondents' Section 2 claim. Kodak claims that its policies preventISOs from ``exploit[ing] the investment Kodak has made in product development,manufacturing and equipment sales in order to take away Kodak's servicerevenues.'' Kodak does not dispute that respondents invest substantiallyin the service market, with training of repair workers and investment inparts inventory. Instead, according to Kodak, the ISOs are free-ridingbecause they have failed to enter the equipment and parts markets. Thisunderstanding of free-riding has no support in our caselaw. To the contrary,as the Court of Appeals noted, one of the evils proscribed by the antitrustlaws is the creation of entry barriers to potential competitors by requiringthem to enter two markets simultaneously.

[54] None of Kodak's asserted business justifications, then, are sufficientto prove that Kodak is ``entitled to a judgment as a matter of law'' onrespondents' Section 2 claim.

III.

[55] In the end, of course, Kodak's arguments may prove to be correct.It may be that its parts, service, and equipment are components of oneunified market, or that the equipment market does discipline the aftermarketsso that all three are priced competitively overall, or that any anticompetitiveeffects of Kodak's behavior are outweighed by its competitive effects.But we cannot reach these conclusions as a matter of law on a record thissparse. Accordingly, the judgment of the Court of Appeals denying summaryjudgment is affirmed.

It is so ordered.

Justice SCALIA, with whom Justice O'CONNOR and Justice THOMAS join,dissenting.

[56]This is not, as the Court describes it, just "another case that concernsthe standard for summary judgment in an antitrust controversy." Rather,the case presents a very narrow--but extremely important--question of substantiveantitrust law: Whether, for purposes of applying our per se rulecondemning "ties," and for purposes of applying our exacting rules governingthe behavior of would-be monopolists, a manufacturer's conceded lack ofpower in the interbrand market for its equipment is somehow consistentwith its possession of "market," or even "monopoly," power in wholly derivativeaftermarkets for that equipment. In my view, the Court supplies an erroneousanswer to this question, and I dissent.

[57] Per se rules of antitrust illegality are reserved for thosesituations where logic and experience show that the risk of injury to competitionfrom the defendant's behavior is so pronounced that it is needless andwasteful to conduct the usual judicial inquiry into the balance betweenthe behavior's procompetitive benefits and its anticompetitive costs. "Thecharacter of the restraint produced by [behavior to which a per serule applies] is considered a sufficient basis for presuming unreasonablenesswithout the necessity of any analysis of the market context in which the[behavior] may be found." The per se rule against tying is justsuch a rule: Where the conditions precedent to application of the ruleare met, i.e., where the tying arrangement is backed up by the defendant'smarket power in the "tying" product, the arrangement is adjudged in violationof Section 1 of the Sherman Act, without any inquiry into the practice'sactual effect on competition and consumer welfare.

[58] Despite intense criticism of the tying doctrine in academic circles,see, e. g., R. Bork, The Antitrust Paradox 365-381 (1978), the statedrationale for our per se rule has varied little over the years.When the defendant has genuine "market power" in the tying product--thepower to raise price by reducing output--the tie potentially enables himto extend that power into a second distinct market, enhancing barriersto entry in each. In addition:

[59] "Tying arrangements may be used to evade price control in the tyingproduct through clandestine transfer of the profit to the tied product;they may be used as a counting device to effect price discrimination; andthey may be used to force a full line of products on the customer so asto extract more easily from him a monopoly return on one unique productin the line." For these reasons, "the law draws a distinction between theexploitation of market power by merely enhancing the price of the tyingproduct, on the one hand, and by attempting to impose restraints on competitionin the market for a tied product, on the other."

[60] Our Section 2 monopolization doctrines are similarly directed todiscrete situations in which a defendant's possession of substantial marketpower, combined with his exclusionary or anticompetitive behavior, threatensto defeat or forestall the corrective forces of competition and therebysustain or extend the defendant's agglomeration of power. Where a defendantmaintains substantial market power, his activities are examined througha special lens: Behavior that might otherwise not be of concern to theantitrust laws -- or that might even be viewed as procompetitive -- cantake on exclusionary connotations when practiced by a monopolist.

[61] The concerns, however, that have led the courts to heightened scrutinyboth of the "exclusionary conduct" practiced by a monopolist and of tyingarrangements subject to per se prohibition, are completely withoutforce when the participants lack market power. As to the former, "the [very]definition of exclusionary conduct," as practiced by a monopolist, ". .. [is] predicated on the existence of substantial market power." And withrespect to tying, we have recognized that bundling arrangements not coercedby the heavy hand of market power can serve the procompetitive functionsof facilitating new entry into certain markets, permitting "clandestineprice cutting in products which otherwise would have no price competitionat all because of fear of retaliation from the few other producers dealingin the market," assuring quality control, and, where "the tied and tyingproducts are functionally related, . . . reduc[ing] costs through economiesof joint production and distribution." "Accordingly, we have [only] condemnedtying arrangements [under the per se rule] when the seller has somespecial ability -- usually called `market power' -- to force a purchaserto do something that he would not do in a competitive market."

[62] The Court today finds in the typical manufacturer's inherent powerover its own brand of equipment -- over the sale of distinctive repairparts for that equipment, for example -- the sort of "monopoly power" sufficientto bring the sledgehammer of Section 2 into play. And, not surprisinglyin light of that insight, it readily labels single-brand power over aftermarketproducts "market power" sufficient to permit an antitrust plaintiff toinvoke the per se rule against tying. In my opinion, this makesno economic sense. The holding that market power can be found on the presentrecord causes these venerable rules of selective proscription to extendwell beyond the point where the reasoning that supports them leaves off.Moreover, because the sort of power condemned by the Court today is possessedby every manufacturer of durable goods with distinctive parts, the Court'sopinion threatens to release a torrent of litigation and a flood of commercialintimidation that will do much more harm than good to enforcement of theantitrust laws and to genuine competition. I shall explain, in Parts IIand III, respectively, how neither logic nor experience suggests,let alone compels, application of the per se tying prohibition andmonopolization doctrine to a seller's behavior in its single-brand aftermarkets,when that seller is without power at the interbrand level.

[63] On appeal in the Ninth Circuit, respondents, having waived their"rule of reason" claim, were limited to arguing that the record, construedin the light most favorable to them, supported application of the perse tying prohibition to Kodak's restrictive parts and service policy.As the Court observes, in order to survive Kodak's motion for summary judgmenton this claim, respondents bore the burden of proffering evidence on whicha reasonable trier of fact could conclude that Kodak possesses power inthe market for the alleged "tying" product.

[64] We must assume, for purposes of deciding this case, that petitioneris without market, much less monopoly, power in the interbrand marketsfor its micrographics and photocopying equipment. In the District Court,respondents did, in fact, include in their complaint an allegation whichposited the interbrand equipment markets as the relevant markets; in particular,they alleged a Section 1 "tie" of micrographics and photocopying equipmentto the parts and service for those machines. Though this allegation wasapparently abandoned in pursuit of Section 1 and 2 claims focused exclusivelyon the parts and service aftermarkets (about which more later), I thinkit helpful to analyze how that claim would have fared under the perse rule.

[65] Had Kodak -- from the date of its entry into the micrographicsand photocopying equipment markets -- included a lifetime parts and servicewarranty with all original equipment, or required consumers to purchasea lifetime parts and service contract with each machine, that bundlingof equipment, parts and service would no doubt constitute a tie. Nevertheless,it would be immune from per se scrutiny under the antitrust lawsbecause the tying product would be equipment, a market inwhich (we assume) Kodak has no power to influence price or quantity. Thesame result would obtain, I think, had Kodak -- from the date of its marketentry -- consistently pursued an announced policy of limiting parts salesin the manner alleged in this case, so that customers bought with the knowledgethat aftermarket support could be obtained only from Kodak. The foreclosureof respondents from the business of servicing Kodak's micrographics andphotocopying machines in these illustrations would be undeniably complete--ascomplete as the foreclosure described in respondents' complaint. Nonetheless,we would inquire no further than to ask whether Kodak's market powerin the equipment market effectively forced consumers to purchase Kodakmicrographics or photocopying machines subject to the company's restrictiveaftermarket practices. If not, that would end the case insofar as the perse rule was concerned. The evils against which the tying prohibitionis directed would simply not be presented. Interbrand competition wouldrender Kodak powerless to gain economic power over an additional classof consumers, to price discriminate by charging each customer a "system"price equal to the system's economic value to that customer, or to raisebarriers to entry in the interbrand equipment markets.

[66] I have described these illustrations as hypothetical, but in factthey are not far removed from this case. The record below is consistent--inlarge part--with just this sort of bundling of equipment on the one hand,with parts and service on the other. The restrictive parts policy, withrespect to micrographics equipment at least, was not even alleged to beanything but prospective. As respondents summarized their factual profferbelow:

[67] "Under this policy, Kodak cut off parts on new products to Kodakmicrographics ISOs. The effect of this, of course, was that as customersof Kodak micrographics ISOs obtained new equipment, the ISOs were unableto service the equipment for that customer, and, service for these customerswas lost by the Kodak ISOs. Additionally, as equipment became obsolete,and the equipment population became all "new equipment" (post April 1985models), Kodak micrographics ISOs would be able to service no equipmentat all."

[68] As to Kodak copiers, Kodak's restrictive parts policy had a broaderfoundation: Considered in the light most favorable to respondents, therecord suggests that, from its inception, the policy was applied to newand existing copier customers alike. But at least all post-1985 purchasersof micrographics equipment, like all post-1985 purchasers of new Kodakcopiers, could have been aware of Kodak's parts practices. The only thinglacking to bring all of these purchasers (accounting for the vast bulkof the commerce at issue here) squarely within the hypotheticals we havedescribed is concrete evidence that the restrictive parts policy was announcedor generally known. Thus, under the Court's approach the existence velnon of such evidence is determinative of the legal standard (the perse rule versus the rule of reason) under which the alleged tie is examined.In my judgment, this makes no sense. It is quite simply anomalous thata manufacturer functioning in a competitive equipment market should beexempt from the per se rule when it bundles equipment with parts-and-service,but not when it bundles parts with service. This vast difference in thetreatment of what will ordinarily be economically similar phenomena isalone enough to call today's decision into question.

[69] In the Court of Appeals, respondents sought to sidestep the impedimentposed by interbrand competition to their invocation of the per setying rule by zeroing in on the parts and service "aftermarkets" for Kodakequipment. By alleging a tie of parts to service, rather thanof equipment to parts-and-service, they identified a tying productin which Kodak unquestionably held a near-monopoly share: the parts uniquelyassociated with Kodak's brand of machines. The Court today holds that sucha facial showing of market share in a single-brand aftermarket is sufficientto invoke the per se rule. The existence of even vibrant interbrandcompetition is no defense.

[70] I find this a curious form of market power on which to premisethe application of a per se proscription. It is enjoyed by virtuallyevery manufacturer of durable goods requiring aftermarket support withunique, or relatively unique, goods. "Such reasoning makes every makerof unique parts for its own product a holder of market power no matterhow unimportant its product might be in the market." (emphasis added).⁽¹⁵⁾Under the Court's analysis, the per se rule may now be applied tosingle-brand ties effected by the most insignificant players in fully competitiveinterbrand markets, as long as the arrangement forecloses aftermarket competitorsfrom more than a de minimis amount of business. This seems to mequite wrong. A tying arrangement "forced" through the exercise of suchpower no more implicates the leveraging and price discrimination concernsbehind the per se tying prohibition than does a tie of the foremarketbrand to its aftermarket derivatives, which -- as I have explained -- wouldnot be subject to per se condemnation.⁽¹⁶⁾ As implemented, the Kodak arrangement challenged in this casemay have implicated truth-in-advertising or other consumer protection concerns,but those concerns do not alone suggest an antitrust prohibition.

[71] In the absence of interbrand power, a seller's predominant or monopolyshare of its single-brand derivative markets does not connote the powerto raise derivative market prices generally by reducing quantity.As Kodak and its principal amicus, the United States, point out,a rational consumer considering the purchase of Kodak equipment will inevitablyfactor into his purchasing decision the expected cost of aftermarket support."Both the price of the equipment and the price of parts and service overthe life of the equipment are expenditures that are necessary to obtaincopying and micrographic services." If Kodak set generally supracompetitiveprices for either spare parts or repair services without making an offsettingreduction in the price of its machines, rational consumers would simplyturn to Kodak's competitors for photocopying and micrographic systems.True, there are the occasional irrational consumers that consider onlythe hardware cost at the time of purchase (a category that regrettablyincludes the Federal Government, whose "purchasing system," we are told,assigns foremarket purchases and aftermarket purchases to different entities).But we have never before premised the application of antitrust doctrineon the lowest common denominator of consumer.

[72] The Court attempts to counter this theoretical point with theoryof its own. It says that there are "information costs"--the costs and inconvenienceto the consumer of acquiring and processing life-cycle pricing data forKodak machines--that "could create a less responsive connection betweenservice and parts prices and equipment sales." But this truism about thefunctioning of markets for sophisticated equipment cannot create "marketpower" of concern to the antitrust laws where otherwise there is none."Information costs," or, more accurately, gaps in the availability andquality of consumer information, pervade real-world markets; and becauseconsumers generally make do with "rough cut" judgments about price in suchcircumstances, in virtually any market there are zones within which otherwisecompetitive suppliers may overprice their products without losing appreciablemarket share. We have never suggested that the principal players in a marketwith such commonplace informational deficiencies (and, thus, bands of apparentconsumer pricing indifference) exercise market power in any sense relevantto the antitrust laws. "While [such] factors may generate `market power'in some abstract sense, they do not generate the kind of market power thatjustifies condemnation of tying."

[73] Respondents suggest that, even if the existence of interbrand competitionprevents Kodak from raising prices generally in its single-brandaftermarkets, there remain certain consumers who are necessarily subjectto abusive Kodak pricing behavior by reason of their being "locked in"to their investments in Kodak machines. The Court agrees; indeed, it goesfurther by suggesting that even a general policy of supracompetitiveaftermarket prices might be profitable over the long run because of the"lock-in" phenomenon. "A seller profitably could maintain supracompetitiveprices in the aftermarket," the Court explains, "if the switching costswere high relative to the increase in service prices, and the number oflocked-in customers were high relative to the number of new purchasers."In speculating about this latter possibility, the Court is essentiallyrepudiating the assumption on which we are bound to decide this case, viz.,Kodak's lack of any power whatsoever in the interbrand market. If Kodak'sgeneral increase in aftermarket prices were to bring the total "system"price above competitive levels in the interbrand market, Kodak would bewholly unable to make further foremarket sales -- and would find itselfexploiting an ever-dwindling aftermarket, as those Kodak micrographic andphotocopying machines already in circulation passed into disuse.

[74] The Court's narrower point, however, is undeniably true. Therewill be consumers who, because of their capital investment in Kodak equipment,"will tolerate some level of service-price increases before changing equipmentbrands,"; this is necessarily true for "every maker of unique partsfor its own product." But this "circumstantial" leverage created by consumerinvestment regularly crops up in smoothly functioning, even perfectly competitive,markets, and in most--if not all--of its manifestations, it is of no concernto the antitrust laws. The leverage held by the manufacturer of a malfunctioningrefrigerator (which is measured by the consumer's reluctance to walk awayfrom his initial investment in that device) is no different in kind ordegree from the leverage held by the swimming pool contractor when he discoversa 5-ton boulder in his customer's backyard and demands an additional sumof money to remove it; or the leverage held by an airplane manufacturerover an airline that has "standardized" its fleet around the manufacturer'smodels; or the leverage held by a drill press manufacturer whose customershave built their production lines around the manufacturer's particularstyle of drill press; the leverage held by an insurance company over itsindependent sales force that has invested in company-specific paraphernalia;or the leverage held by a mobile home park owner over his tenants, whoare unable to transfer their homes to a different park except at greatexpense. Leverage, in the form of circumstantial power, plays arole in each of these relationships; but in none of them is the leverageattributable to the dominant party's market power in any relevantsense. Though that power can plainly work to the injury of certain consumers,it produces only "a brief perturbation in competitive conditions--not thesort of thing the antitrust laws do or should worry about."

[75] The Court correctly observes that the antitrust laws do not permiteven a natural monopolist to project its monopoly power into anothermarket, i.e., to "`exploi[t] his dominant position in one marketto expand his empire into the next.'" However, when a manufacturer usesits control over single-branded parts to acquire influence in single-brandedservice, the monopoly "leverage" is almost invariably of no practical consequence,because of perfect identity between the consumers in each of the subjectaftermarkets (those who need replacement parts for Kodak equipment, andthose who need servicing of Kodak equipment). When that condition exists,the tie does not permit the manufacturer to project power over a classof consumers distinct from that which it is already able to exploit (andfully) without the inconvenience of the tie.

[76] We have never before accepted the thesis the Court today embraces:that a seller's inherent control over the unique parts for its own brandamounts to "market power" of a character sufficient to permit invocationof the per se rule against tying. As the Court observes, we haveapplied the per se rule to manufacturer ties of foremarketequipment to aftermarket derivatives--but only when the manufacturer'smonopoly power in the equipment, coupled with the use of derivative salesas "counting devices" to measure the intensity of customer equipment usage,enabled the manufacturer to engage in price discrimination, and therebymore fully exploit its interbrand power. That sort of enduring opportunityto engage in price discrimination is unavailable to a manufacturer--likeKodak--that lacks power at the interbrand level. A tie between two aftermarketderivatives does next to nothing to improve a competitive manufacturer'sability to extract monopoly rents from its consumers.⁽¹⁷⁾

[77] In the absence of interbrand power, a manufacturer's bundling ofaftermarket products may serve a multitude of legitimate purposes: It mayfacilitate manufacturer efforts to ensure that the equipment remains operableand thus protect the seller's business reputation; it may create the conditionsfor implicit consumer financing of the acquisition cost of the tying equipmentthrough supracompetitively-priced aftermarket purchases; and it may, throughthe resultant manufacturer control of aftermarket activity, "yield valuableinformation about component or design weaknesses that will materially contributeto product improvement,". Because the interbrand market will generallypunish intrabrand restraints that consumers do not find in their interest,we should not--under the guise of a per se rule--condemn such potentiallyprocompetitive arrangements simply because of the antitrust defendant'sinherent power over the unique parts for its own brand.

[78] I would instead evaluate the aftermarket tie alleged in this caseunder the rule of reason, where the tie's actual anticompetitiveeffect in the tied product market, together with its potential economicbenefits, can be fully captured in the analysis. Disposition of this casedoes not require such an examination, however, as respondents apparentlywaived any rule-of-reason claim they may have had in the District Court.I would thus reverse the Ninth Circuit's judgment on the tying claim outright.

[79] These considerations apply equally to respondents' Section 2 claims.An antitrust defendant lacking relevant "market power" sufficient to permitinvocation of the per se prohibition against tying a fortiorilacks the monopoly power that warrants heightened scrutiny of his allegedlyexclusionary behavior. Without even so much as asking whether the purposesof Section 2 are implicated here, the Court points to Kodak's control of"100% of the parts market and 80% to 95% of the service market," marketswith "no readily available substitutes," and finds that the proffer ofsuch statistics is sufficient to fend off summary judgment. But this showingcould easily be made, as I have explained, with respect to virtually anymanufacturer of differentiated products requiring aftermarket support.By permitting antitrust plaintiffs to invoke Section 2 simply upon theunexceptional demonstration that a manufacturer controls the supplies ofits single-branded merchandise, the Court transforms Section 2 from a specializedmechanism for responding to extraordinary agglomerations (or threatenedagglomerations) of economic power to an all-purpose remedy against run-of-the-millbusiness torts.

[80] In my view, if the interbrand market is vibrant, it is simply notnecessary to enlist Section 2's machinery to police a seller's intrabrandrestraints. In such circumstances, the interbrand market functions as aninfinitely more efficient and more precise corrective to such behavior,rewarding the seller whose intrabrand restraints enhance consumer welfarewhile punishing the seller whose control of the aftermarkets is viewedunfavorably by interbrand consumers. Because this case comes to us on theassumption that Kodak is without such interbrand power, I believe we arecompelled to reverse the judgment of the Court of Appeals. I respectfullydissent.

ENDNOTES:

⁽¹⁾Kodak's micrographic equipment includesfour different product areas. The first is capture products such as microfilmersand electronic scanners, which compact an image and capture it on microfilm.The second is equipment such as microfilm viewers and viewer/printers.This equipment is used to retrieve the images. The third is Computer OutputMicroform (COM) recorders, which are data-processing peripherals that recordcomputer-generated data onto microfilm. The fourth is Computer AssistedRetrieval (CAR) systems, which utilize computers to locate and retrievemicrographic images. See App. 156-158. 

⁽²⁾In addition to the OEMs, other sourcesof Kodak parts include (1) brokers who would buy parts from Kodak, or stripused Kodak equipment to obtain the useful parts and resell them, (2) customerswho buy parts from Kodak and make them available to ISOs, and (3) usedequipment to be stripped for parts. 

⁽³⁾Section 1 of the Sherman Act statesin relevant part: ``Every contract, combination in the form of trust orotherwise, or conspiracy, in restraint of trade or commerce among the severalStates, or with foreign nations, is declared to be illegal.'' 15 U.S.C.1.

Section 2 of the Sherman Act states: ``Every person who shall monopolize,or attempt to monopolize, or combine or conspire with any other personor persons, to monopolize any part of the trade or commerce among the severalStates, or with foreign nations, shall be deemed guilty of a felony, and,on conviction thereof, shall be punished by fine not exceeding one milliondollars if a corporation, or, if any other person, one hundred thousanddollars, or by imprisonment not exceeding three years, or by both saidpunishments, in the discretion of the court.'' 15 U.S.C. 2. 

⁽⁴⁾``[T]he essential characteristic ofan invalid tying arrangement lies in the seller's exploitation of its controlover the tying product to force the buyer into the purchase of a tied productthat the buyer either did not want at all, or might have preferred to purchaseelsewhere on different terms. When such `forcing' is present, competitionon the merits in the market for the tied item is restrained and the ShermanAct is violated.'' Jefferson Parish, 466 U.S., at 12. 

⁽⁵⁾In their brief and at oral argument,respondents argued that Kodak's market share figures for high-volume copymachines, computer-assisted retrieval systems, and micrographic-captureequipment demonstrate Kodak's market power in the equipment market.

In the Court of Appeals, however, respondents did not contest Kodak'sassertion that its market shares indicated a competitive equipment market.The Court of Appeals believed that respondents ``do not dispute Kodak'sassertion that it lacks market power in the [equipment] markets.'' Nordid respondents question Kodak's asserted lack of market power, althoughthey acknowledged that Kodak's entire case rested on its understandingthat respondents were not disputing the existence of competition in theequipment market. 

⁽⁶⁾What constrains the defendant's abilityto raise prices in the service market is ``the elasticity of demand facedby the defendant--the degree to which its sales fall . . . as its pricerises.''

Courts usually have considered the relationship between price in onemarket and demand in another in defining the relevant market. Because marketpower is often inferred from market share, market definition generallydetermines the result of the case. Kodak chose to focus on market powerdirectly rather than arguing that the relationship between equipment andservice and parts is such that the three should be included in the samemarket definition. Whether considered in the conceptual category of ``marketdefinition'' or ``market power,'' the ultimate inquiry is the same -- whethercompetition in the equipment market will significantly restrain power inthe service and parts markets. 

⁽⁷⁾The United States as Amicus Curiaein support of Kodak echoes this argument: ``The ISOs' claims are implausiblebecause Kodak lacks market power in the markets for its copier and micrographicequipment. Buyers of such equipment regard an increase in the price ofparts or service as an increase in the price of the equipment, and sellersrecognize that the revenues from sales of parts and service are attributableto sales of the equipment. In such circumstances, it is not apparent howan equipment manufacturer such as Kodak could exercise power in the aftermarketsfor parts and service.'' 

⁽⁸⁾It is clearly true, as the United Statesclaims, that Kodak ``cannot set service or parts prices without regardto the impact on the market for equipment.'' The fact that the cross-elasticityof demand is not zero proves nothing; the disputed issue is how much ofan impact an increase in parts and service prices has on equipment salesand on Kodak's profits. 

⁽⁹⁾In addition, of course, in order toprice accurately the equipment, a consumer would need initial purchaseinformation such as prices, features, quality, and available warranties,for different machinery with different capabilities, and residual valueinformation such as the longevity of product use and its potential resaleor trade-in value. 

⁽¹⁰⁾To inform consumers about Kodak,the competitor must be willing to forgo the opportunity to reap supracompetitiveprices in its own service and parts markets. The competitor may anticipatethat charging lower service and parts prices and informing consumers aboutKodak in the hopes of gaining future equipment sales will cause Kodak tolower the price on its service and parts, cancelling any gains in equipmentsales to the competitor and leaving both worse off. Thus, in an equipmentmarket with relatively few sellers, competitors may find it more profitableto adopt Kodak's service and parts policy than to inform the consumers.(Kodak, Xerox, and IBM together have nearly 100% of relevant market).

Even in a market with many sellers, any one competitor may not havesufficient incentive to inform consumers because the increased patronageattributable to the corrected consumer beliefs will be shared among othercompetitors. 

⁽¹¹⁾A firm can exact leverage wheneverother equipment is not a ready substitute. 

⁽¹²⁾The dissent disagrees based on itshypothetical case of a tie between equipment and service. "The only thinglacking" to bring this case within the hypothetical case, states the dissent,"is concrete evidence that the restrictive parts policy was . . . generallyknown." But the dissent's "only thing lacking" is the crucial thing lacking---evidence.Whether a tie between parts and service should be treated identically toa tie between equipment and service, as the dissent and Kodak argue, dependson whether the equipment market prevents the exertion of market power inthe parts market. Far from being "anomalous," requiring Kodak to provideevidence on this factual question is completely consistent with our priorprecedent. 

⁽¹³⁾It bears repeating that in this caseKodak has never claimed that it is in fact pursuing such a pricing strategy. 

⁽¹⁴⁾The dissent urges a radical departurein this Court's antitrust law. It argues that because Kodak has only an"inherent" monopoly in parts for its equipment, the antitrust laws do notapply to its efforts to expand that power into other markets. The dissent'sproposal to grant per se immunity to manufacturers competing inthe service market would exempt a vast and growing sector of the economyfrom antitrust laws. Leaving aside the question whether the Court has theauthority to make such a policy decision, there is no support for it inour jurisprudence or the evidence in this case.

Even assuming, despite the absence of any proof from the dissent, thatall manufacturers possess some inherent market power in the parts market,it is not clear why that should immunize them from the antitrust laws inanother market. The Court has held many times that power gained throughsome natural and legal advantage such as a patent, copyright, or businessacumen can give rise to liability if "a seller exploits his dominant positionin one market to expand his empire into the next." Moreover, on the occasionswhen the Court has considered tying in derivative aftermarkets by manufacturers,it has not adopted any exception to the usual antitrust analysis, treatingderivative aftermarkets as it has every other separate market. Our pastdecisions are reason enough to reject the dissent's proposal.

Nor does the record in this case support the dissent's proposed exemptionfor aftermarkets. The dissent urges its exemption because the tie here"does not permit the manufacturer to project power over a class of consumersdistinct from that which it is already able to exploit (and fully) withoutthe inconvenience of the tie." Beyond the dissent's obvious difficultyin explaining why Kodak would adopt this expensive tying policy if it couldachieve the same profits more conveniently through some other means, respondentsoffer an alternative theory, supported by the record, that suggests Kodakis able to exploit some customers who in the absence of the tiewould be protected from increases in parts prices by knowledgeable customers.

At bottom, whatever the ultimate merits of the dissent's theory, atthis point it is mere conjecture. Neither Kodak nor the dissent have providedany evidence refuting respondents' theory of forced unwanted purchasesat higher prices and price discrimination. While it may be, as the dissentpredicts, that the equipment market will prevent any harms to consumersin the aftermarkets, the dissent never makes plain why the Court shouldaccept that theory on faith rather than requiring the usual evidence neededto win a summary judgment motion. 

⁽¹⁵⁾That there exist innumerable partsand service firms in such industries as the automobile industry, does notdetract from this point. The question whether power to control an aftermarketexists is quite distinct from the question whether the power has been exercised.Manufacturers in some markets have no doubt determined that exclusionaryintrabrand conduct works to their disadvantage at the competitive interbrandlevel, but this in no way refutes the self-evident reality that controlover unique replacement parts for single-branded goods is ordinarily availableto such manufacturers for the taking. It confounds sound analysis to suggest,as respondents do, that the asserted fact that Kodak manufactures only10% of its replacement parts, and purchases the rest from original equipmentmanufacturers, casts doubt on Kodak's possession of an inherent advantagein the aftermarkets. It does no such thing, any more than Kodak's contractingwith others for the manufacture of all constituent parts included in itsoriginal equipment would alone suggest that Kodak lacks power in the interbrandmicrographics and photocopying equipment markets. The suggestion implicitin respondents' analysis -- that if a seller chooses to contract for themanufacture of its branded merchandise, it must permit the contractorsto compete in the sale of that merchandise -- is plainly unprecedented. 

⁽¹⁶⁾Even with interbrand power,I may observe, it is unlikely that Kodak could have incrementally exploitedits position through the tie of parts to service alleged here. Most ofthe "service" at issue is inherently associated with the parts, i.e.,that service involved in incorporating the parts into Kodak equipment,and the two items tend to be demanded by customers in fixed proportions(one part with one unit of service necessary to install the part). Whenthat situation obtains, "no revenue can be derived from setting a higherprice for the tied product which could not have been made by setting theoptimum price for the tying product." These observations strongly suggestthat Kodak parts and the service involved in installing them should notbe treated as distinct products for antitrust tying purposes. ("For productsto be treated as distinct, the tied product must, at a minimum, be onethat some consumers might wish to purchase separately without also purchasingthe tying product.") 

⁽¹⁷⁾The Court insists that the recordin this case suggests otherwise, i.e., that a tie between partsand service somehow does enable Kodak to increase overall monopolyprofits. Although the Court does not identify the record evidence on whichit relies, the suggestion, apparently, is that such a tie facilitates pricediscrimination between sophisticated, "high-volume" users of Kodak equipmentand their unsophisticated counterparts. The sophisticated users (who, theCourt presumes, invariably self-service their equipment) are permittedto buy Kodak parts without also purchasing supracompetitively-priced Kodakservice, while the unsophisticated are -- through the imposition of thetie -- compelled to buy both.

While superficially appealing, at bottom this explanation lacks coherence.Whether they self-service their equipment or not, rational foremarket consumers(those consumers who are not yet "locked in" to Kodak hardware) will bedriven to Kodak's competitors if the price of Kodak equipment, togetherwith the expected cost of aftermarket support, exceeds competitive levels.This will be true no matter how Kodak distributes the total system priceamong equipment, parts, and service. Thus, as to these consumers, Kodak'slack of interbrand power wholly prevents it from employing a tie betweenparts and service as a vehicle for price discrimination. Nor does a tiebetween parts and service offer Kodak incremental exploitative power overthose consumers -- sophisticated or not -- who have the supposed misfortuneof being "locked in" to Kodak equipment. If Kodak desired to exploit itscircumstantial power over this wretched class by pressing them up to thepoint where the cost to each consumer of switching equipment brands barelyexceeded the cost of retaining Kodak equipment and remaining subject toKodak's abusive practices, it could plainly do so without the inconvenienceof a tie, through supracompetitive parts pricing alone. Since the locked-insophisticated parts purchaser is as helpless as the locked-in unsophisticatedone, I see nothing to be gained by price discrimination in favor of theformer. If such price discrimination were desired, however, it would nothave to be accomplished indirectly, through a tie of parts to service.Section 2(a) of the Robinson-Patman Act would prevent giving lower partsprices to the sophisticated customers only "where the effect of such discriminationmay be substantially to lessen competition or tend to create a monopolyin any line of commerce, or to injure, destroy, or prevent competitionwith any person who either grants or knowingly receives the benefit ofsuch discrimination, or with customers of either of them . . . ." Thatprohibited effect often occurs when price-discriminated goods are soldfor resale (i.e., to purchasers who are necessarily in competitionwith one another). ("Secondary-line injury arises [under the Robinson-PatmanAct] when a powerful firm buying supplies at favorable prices thereby gainsa decisive advantage over its competitors that are forced to pay higherprices for their supplies"). It rarely occurs where, as would be the casehere, the price-discriminated goods are sold to various businesses forconsumption.