Richard RODNEY, Jr., on behalf of himself and all others similarly situated, Plaintiff-Appellant, v. NORTHWEST AIRLINES, INC.; and Northwest Airlines Corp., Defendants-Appellees, 146 F. App'x 783

BATCHELDER, Circuit Judge. Plaintiff-Appellant Richard Rodney appeals the district court’s order denying his motion to certify his antitrust suit as a class action. Because the district court did not abuse its discretion in concluding that questions of law or fact common to the members of the class do not predominate over any questions affecting only Rodney, we will AFFIRM the district court’s order denying class certification. I. On March 6, 2001, Rodney filed a two-count complaint against Northwest Airlines, Inc. and Northwest Airlines Corp. (“Northwest”) seeking damages and injunctive relief under Sections 4 and 16 of the Clayton Act, 15 U.S.C. §§ 15 and 26 and

Section 2 of the Sherman Act, 15 U.S.C. § 2. The complaint alleges that Northwest made efforts “to illegally dominate and control the market for passenger air travel into and out of Memphis Airport (the ‘Memphis Hub’), Detroit Metropolitan Airport (the ‘Detroit Hub’), and Minneapolis/St. Paul Airport (the ‘Minneapolis Hub’).... ” Rodney claims that he was harmed by Northwest’s monopolistic practices in 1996, when he took a Northwest flight from the Minneapolis Hub to Los Angeles. On July 25, 2001, Rodney moved the court for class certification on behalf of a damages class pursuant to Fed. R. Crv. P. 23(

a) and (b)(3) and an injunctive relief class pursuant to Fed. R. Crv. P. 23(

a) and (b)(2). Rodney’s motion asked the court to name himself and Phillip Sax, another antitrust plaintiff who was represented by the same counsel, as lead plaintiffs.

The motion defined the class as, All members of Northwest Airlines’ WorldPerks frequent flyer program who purchased and used (i.e., flew those routes) airline tickets from Northwest Airlines or through its authorized agents on either a “Y” fare, or other unrestricted or fully refundable fare, and/or any other airline ticket purchased 14 days or less prior to departure, for non-stop travel into or out of Detroit-Wayne County International Airport [ ], Minneapolis-St.

Paul International Airport [], and/or Memphis International Airport [ ], on one-way tickets or round trip tickets, on [] “supracompetitve price” routes.... On March 31, 2004, the district court issued an order and memorandum denying the motion for class certification. Because Rodney spent little energy addressing the certification of the injunctive relief class and did not identify a specific activity for the court to enjoin, the court refused to certify the class under Rule 23(b)(2); Rodney does not appeal this decision.

The district court denied Rodney’s request for certification of a damages class under Rule 23(b)(3) on two independent grounds. First, the court held that the plaintiffs failed to demonstrate that questions common to the class members predominate over any questions affecting only individual members. In reaching this conclusion, the court stated that individualized evidence could well predominate over common evidence on the issues of monopoly power, antitrust injury, and damages.

Alternatively, the district court ruled that Rodney could not adequately represent the class because he flew only one of the many allegedly monopolized routes. The court also held that Sax was not a proper member of the class. Rodney appealed and on June 25, 2004, a panel of this court granted permission to take an interlocutory appeal pursuant to Rule 23(f). Rodney’s case relies almost exclusively on a report prepared by two professors, Dr. Clinton Oster, Jr. and Dr. John S. Strong, both of whom are experts in the economics of the airline industry.

The report concludes that Northwest exercises monopoly power for air services originating in each of its three hubs. On appeal, Rodney contends that Northwest monopolizes 74 separate routes as a result of its domination of the three hubs. II. We review for abuse of discretion a district court’s decision to deny class certification. Alkire v. Irving, 330 F.3d 802, 810 (6th Cir.2003). An abuse of discretion occurs when the district court relies upon clearly erroneous findings of fact, improperly applies the proper legal standard, or employs an erroneous legal standard. Miami Univ. Wrestling Club v.

Miami Univ., 302 F.3d 608, 613 (6th Cir.2002). We will not find an abuse of discretion without “a definite and firm conviction that the trial court committed a clear error of judgment.” Id. The party seeking certification of a class under Rule 23(

a) must show that, (1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class. In addition to the prerequisites of Rule 23(a), a party seeking class certification must show that the class action is maintainable under Rule 23(b).

Rodney contends that his class action is maintainable under Rule 23(b)(3) because “questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy.” “There are no hard and fast rules ... regarding the suitability of a particular type of antitrust case for class action treatment. Rather, the unique factors of each case will generally be the determining factor governing certification.” Bell Atlantic Corp. v.

AT&T, 339 F.3d 294, 301 (5th Cir.2003) (internal quotation omitted). Rule 23 does not give the court “any authority to conduct a preliminary inquiry into the merits of a suit in order to determine whether it may be maintained as a class action.” Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 177, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974). A district court cannot deny certification based upon its belief that the plaintiff cannot prevail on the merits. Castano v. American Tobacco Co., 84 F.3d 734, 744 (5th Cir.1996).

However, a court is allowed to look beyond the pleadings on a class certification motion to determine what type of evidence will be presented by the parties. See General Telephone Co. v. Falcon, 457 U.S. 147, 160, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982). “Under Rule 23(f), this court can review the merits of an appeal only insofar as they bear upon the propriety of class certification, that is, whether the proposed class satisfies the prerequisites of Rule 23.” In re Lorazepam & Clorazepate Antitrust Litigation, 289 F.3d 98, 105 (D.C.Cir.2002).

Rule 23(b)(3)’s “predominance” inquiry tests “whether the proposed classes are sufficiently cohesive to warrant adjudication by representation.” Amchem Products, Inc. v. Windsor, 521 U.S. 591, 623, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997). In analyzing the predominance requirement, courts “take care to inquire into the substance and structure of the underlying claims without passing judgment on their merits.” Robinson v.

Texas Automobile Dealers Assoc., 387 F.3d 416, 421 (5th Cir.2004). “Determining whether the plaintiffs can clear the predominance hurdle set by Rule 23(b)(3) also requires us to consider how a trial on the merits would be conducted if a class were certified.” Bell Atlantic, 339 F.3d at 302 (internal quotation omitted). Though not a determination on the merits, the Rule 23(b)(3) analysis helps “prevent [] the class from degenerating into a series of individual trials.” Id. (internal quotation omitted).

To prevail in an antitrust action, a plaintiff must prove several elements, including that: 1) the defendant possessed monopoly power in a properly defined market; 2) the defendant engaged in anti-competitive or exclusionary conduct with respect to each properly defined market; 3) the plaintiff was actually injured; 4) the defendant’s conduct caused antitrust injury; and 5) the plaintiff suffered damages. See Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 334-35, 110 S.Ct. 1884, 109 L.Ed.2d 333 (1990); Brunswick Corp. v.

Pueblo Bowl-O-Mat, 429 U.S. 477, 485-91, 97 S.Ct. 690, 50 L.Ed.2d 701 (1977); Conwood Co., L.P. v. U.S. Tobacco Co., 290 F.3d 768, 782 (6th Cir.2002) (“[a] claim under § 2 of the Sherman Act requires proof of two elements: 1) the possession of monopoly power in a relevant market; and 2) the willful acquisition, maintenance, or use of that power by anti-competitive or exclusionary means.... ”).

We will first examine Rodney’s argument that a court analyzing whether common issues will predominate under Rule 23(b)(3) is confined to the evidence contained in the plaintiff’s case-in-chief and then consider, in turn, whether each element of an antitrust action would be established with proof common to the class.

Rodney argues that the district court’s application of Rule 23(b)(3) was erroneous because “in deciding whether common issues predominate, a court should assess only the plaintiffs proffered method of proving his case-in-chief.” We have held, however, that “the parties should be afforded an opportunity to present evidence on the maintainability of the class action.” In re American Medical Systems, Inc., 75 F.3d 1069, 1079 (6th Cir.1996) (emphasis added) (quotation omitted) (considering “evidence in the record presented by the nonmoving party” in reversing the district court’s order granting class certification); see also Waste Mgmt.

Holdings v. Mowbray, 208 F.3d 288, 295 (1st Cir.2000) (“we regard the law as settled that affirmative defenses should be considered in making class certification decisions”). Indeed, we have noted, albeit in an unpublished opinion, that “the Advisory Committee Notes to Rule 23(b)(3) advise against class certification where a defendant has a defense to liability that will vary with each individual class member.” Butler v. Sterling, Inc., No. 98-3223, 2000 WL 353502 at *6 (6th Cir. March 31, 2000).

Because a defendant’s evidence may be probative of class cohesiveness and may be such as to cause the class to degenerate into a series of individual trials, we hold that a court performing a “predominance” inquiry under Rule 23(b)(3) may consider not only the evidence presented in the plaintiffs case-in-chief but the defendant’s likely rebuttal evidence. Considering both Rodney’s evidence and the evidence that Northwest is likely to present in response to Rodney’s claims, we proceed to our examination of whether each element of Rodney’s antitrust case would be established with proof common to the class.

Market Definition Establishing relevant markets is the first step in every monopolization claim. Conwood Co., L.P., 290 F.3d at 782. Rodney defines the market as all non-stop scheduled flights into and out of each of Northwest’s Hubs.

At issue then, is whether questions of law or fact common to the members of the class predominate in determining whether Rodney has properly defined the market. “In considering what is the relevant market for determining the control of price and competition, no more definite rule can be declared than that commodities reasonably interchangeable by consumers for the same purposes make up that ‘part of the trade or commerce’, monopolization of which may be illegal.” Worldwide Basketball and Sports Tours, Inc. v. NCAA, 388 F.3d 955, 961 (6th Cir.2004) (quoting United States v.

E.I. du Pont de Nemours & Co., 351 U.S. 377, 395, 76 S.Ct. 994, 100 L.Ed. 1264 (1956)). This “reasonable interchangeability” standard, which is the essential test for determining the relevant product market, “may be gauged by (1) the product uses, ie., whether the substitute products or services can perform the same function, and/or (2) consumer response (cross-elasticity); that is, consumer sensitivity to price levels at which they elect substitutes for the defendant’s product or service.” Worldwide Basketball, 388 F.3d at 961 (quoting White & White, Inc. v. American Hosp.

Supply Corp., 723 F.2d 495, 500 (6th Cir.1983)). The plaintiff carries the burden of proving that no such substitutes are available. Int’l Logistics Group v. Chrysler Corp., 884 F.2d 904, 908 (6th Cir.1989). We think that individual questions predominate on the issue of market definition because cross-elasticity analyses will be specific to each of the 74 routes. For example, an analysis of whether bus travel from Detroit to Toledo is reasonably interchangeable with a flight between those two cities will not help to define the market for travel between Minneapolis and Los Angeles. Drs.

Strong and Oster’s analysis of whether a competing airline’s flight offerings differs from Northwest’s offerings suggests that individual issues will predominate over the question of market definition. The report states, The extent to which a flight with different characteristics is a substitute for another airline’s flight offerings depends on both the nature of the trip and the nature of the traveler.

For a short-distance trip, a one-stop flight or a connecting flight is likely to have a substantial time and circuity penalty compared to a nonstop flight, so that a traveler will be less likely to such flight offerings as acceptable substitutes, even if they are priced somewhat lower. On a long-distance trip, the time penalty and added circuity are less noticeable and one-stop or connecting flights are more likely to be considered substitutes. Dr.

Oster testified that factors including flight frequencies, flight times, size of airports, the existence of a layover, and duration of the flight are all factors that must be considered in an analysis of whether substitute products can perform the same function as the Northwest flights. He further testified that this analysis would have to be performed on a route-by-route basis. When asked whether markets with fewer than 20,000 passengers per year are natural monopolies, Dr. Oster responded “[n]ot necessarily,” and later elaborated that this analysis would have to be performed on a market-by-market basis.

Shortly thereafter, the following exchange occurred, Question: And would you agree with me that even with high market shares that there are other factors that must be examined to determine whether monopoly power exists? Answer: Yes. Question: And would you agree with me that the factors that must be examined have to be examined on a route-specific basis? Answer: Yes. We also note that the report of Northwest’s expert witness indicates that the company would present route-specific evidence in support of its argument that Rodney has failed to define the market.

Because market definition will involve an analysis of the alternatives available in each of the 74 different routes, this factor weighs against certification. Even assuming that questions of law or fact common to the members of the class predominate over the issue of whether Rodney has properly defined the market, we think that his definition of the market as all non-stop scheduled flights into and out of each of Northwest’s hubs fails as a matter of law. At the outset we must consider whether Rule 23(

f) allows a court to deny class certification based upon an antitrust plaintiffs failure to define the market. While our circuit has not spoken to this issue, in Unger v. Amedisys Inc., the Fifth Circuit vacated the district court’s certification of a securities fraud class action based on the “fraud on the market” theory, which requires proof that the security at issue is traded in an “efficient market.” 401 F.3d 316, 322 (5th Cir.2005) (quotation omitted).

Stating that “[questions of market efficiency cannot be treated differently from other preliminary certification issues,” the court of appeals vacated the class certification and remanded because the district court devoted insufficient attention to evaluating the market efficiency factors. Id. at 325. Similarly, the Fourth Circuit has held that “while an evaluation of the merits to determine the strength of plaintiffs’ case is not part of a Rule 23 analysis, the factors spelled out in Rule 23 must be addressed through findings, even if they overlap with issues on the merits.” Gariety v.

Grant Thornton LLP, 368 F.3d 356, 366 (4th Cir.2004) (vacating the certification of a class on the ground that a district court does not comply with the procedural requirements of Rule 23 by accepting a plaintiffs allegation that an “efficient market” exists). We think that the reasoning of the Unger and Gariety courts, which we find to be persuasive, applies with equal force in instances where an antitrust plaintiff, who has not properly defined the market, seeks class certification.

Accordingly, we hold that a court may consider an antitrust plaintiffs failure to define the market as part of its class certification analysis without violating the procedural requirements of Rule 23. As explained above, interchangeable commodities are part of the same market, Worldwide Basketball, 388 F.3d at 961, and an antitrust plaintiff carries the burden of proving that no reasonable substitutes are available. International Logistics Group, Ltd. v. Chrysler Corp., 884 F.2d 904, 908 (6th Cir.1989). Drs.

Oster and Strong’s report does not contain a cross-elasticity analysis of whether bus or train travel is a reasonable substitute for flights on Northwest and Dr. Oster himself testified that he did not analyze whether ground transportation is a substitute for air transportation. Moreover, Dr. Oster testified that he did not do a specific analysis to determine whether connecting service is a substitute for non-stop service. Rodney’s failure to properly define a market also weighs against a finding that certification is proper.

Monopoly Power To make a successful claim under § 2 of the Sherman Act, Rodney must prove that Northwest possessed monopoly power in the relevant market. Re/Max Intern., Inc. v. Realty One, Inc., 173 F.3d 995, 1016 (6th Cir.1999). A plaintiff can do this either directly, by “showing the exercise of actual control over prices or the actual exclusion of competitors,” or circumstantially, by showing that the defendant has a high market share within a defined market. Id. (internal quotation omitted).

To prove that Northwest exercised monopoly power, Rodney’s experts composed a series of “data screens” designed to identify those routes that were monopolized.

The district court held that individualized information would predominate over the monopoly power element because Northwest would probably impeach Rodney’s methodology “by showing that the data screens do not credibly produce routes where Northwest had monopoly power.” “This type of evidence,” wrote the district court “could easily involve voluminous individualized information for each specific route, rather than just evidence tending to discredit plaintiffs method as a whole.” We are not firmly convinced that the district court was wrong.

Rodney’s use of “data screens” does little to assuage our concern that proving monopoly power will cause the class action to degenerate into a series of mini-trials inasmuch as Rodney’s own experts describe the “data screens” as a “Market-By-Market Analysis of Market Power.” Even assuming that Rodney’s “data screens” prove that Northwest has a disproportionately high market share in a manner that will not result in a series of mini-trials, an analysis of other factors probative of monopoly power, like the absence of entry barriers, see Am. Council of Certified Podiatric Physicians and Surgeons v. Am.

Board of Podiatric Surgery, 185 F.3d 606, 623 (6th Cir.1999), will necessarily be done on a route-by-route basis. “Entry barriers are ‘additional long-run costs that were not incurred by incumbent firms but must be incurred by new entrants,’ or ‘factors in the market that deter entry while permitting incumbent firms to earn monopoly returns.’ ” Rebel Oil Co., Inc. v. Atlantic Richfield Co. 51 F.3d 1421, 1439 (9th Cir.1995) (quoting Los Angeles Land Co. v. Brunswick Corp., 6 F.3d 1422, 1427-28 (9th Cir.1993)).

Sources of entry barriers include licensing agreements, buyer preference for established brands, and capital market evaluations imposing higher capital costs on new entrants. Rebel Oil, 51 F.3d at 1439. The report of Janusz A. Ordover, the defense's expert, indicates that Northwest will rebut Rodney’s data screens by citing these sorts of entry barriers, the presence or absence of which will be specific to each of the 74 individual routes.

Because the presence of entry barriers is unique to each route, proof that Northwest exercised monopoly power over the route that Rodney took does not establish that Northwest exercised monopoly power over the other 73 routes. This factor weighs against granting class certification.

Anti-competitive Conduct and Injury In Fact The district court thought that the elements of anti-competitive conduct and injury in fact weighed in favor of granting class certification and we are inclined to agree. “Anticompetitive conduct is conduct designed to destroy competition, not just to eliminate a competitor.” Richter Concrete v. Hilltop Concrete, 691 F.2d 818, 823 (6th Cir.1982). Rodney seeks to satisfy this element not through proof of specific anti-competitive acts, but rather through the “predation by reputation” theory.

According to this theory, Northwest’s reputation for aggressive competition at its hubs deterred entry into those markets, and evidence to support this theory would seem to span all of the 74 markets and apply to all of the class members. We do not think that such proofs are likely to cause the class action to degenerate into a series of mini-trials.

Antitrust Injury In addition to showing anti-competitive conduct and cause in fact, an antitrust plaintiff must also show an “antitrust injury,” which is “(1) ‘injury of the type that the antitrust laws were intended to prevent’ and (2) injury ‘that flows from that which makes defendants’ acts unlawful.’ ” In re Cardizem, 332 F.3d 896, 909 (6th Cir.2003) (applying § 1 of the Sherman Act) (quoting Brunswick Corp., 429 U.S. at 489, 97 S.Ct. 690); see also Atlantic Richfield, 495 U.S. at 342-43, 110 S.Ct. 1884.

This requirement “ensures that a plaintiff can recover only if the loss stems from a competition-reducing aspect or effect of the defendant’s behavior.” Cardizem, 332 F.3d at 910 (quoting Atlantic Richfield, 495 U.S. at 342-43, 110 S.Ct. 1884).

Rodney argues that his “data screens provide a common method of proving widespread injury to the class.” He adds that “in selecting the damaged routes, Plaintiffs experts applied a set of data screens designed to filter out any discrepancy not caused by anti-competitive conduct.” Even assuming that Rodney’s “data screens” do filter out injuries not caused by anti-competitive conduct, we do not think that they could do so in a manner that would preserve class cohesion in view of his experts’ statement that the data screens would have to be applied on a “Route-By-Route” basis.

Nor do we agree that Rodney’s use of the “predation by reputation” theory to prove that passengers on each of the 74 different routes suffered an antitrust injury would avoid the degeneration of the class action into a series of mini-trials. As the district court correctly observed, “it is not readily apparent how a ‘reputation for predation’ directly damages individual passengers.” In any event, Northwest would likely be forced to rebut Rodney’s claims with evidence that competing carriers chose not to enter particular routes for reasons other than Northwest’s reputation.

According to Michael Mooney, an executive employed by Midwest Express who testified for Rodney, factors that influence a competing carrier’s decision to enter a market include the profitability of the route and the strategic fit of the route vis-a-vis the carrier’s business plan. Dr. Strong also testified that a competing carrier’s pilot contracts may prevent that carrier from entering a specific market.

In short, the record reveals that there are any number of reasons, other than Northwest’s “reputation,” why Northwest’s competitors declined to enter a given route and these reasons would likely be different from one route to the next. If we were to certify Rodney’s class, Northwest would be forced to prove that competitors declined to enter the route that Rodney actually took for reasons other than Northwest’s reputation and then make the same showing for each of the 73 other routes at issue. Clearly, issues unique to each of the 74 different routes predominate on the question of antitrust injury.

Damages A plaintiff seeking class certification must present a damages model that functions on a class-wide basis. See Bell Atlantic, 339 F.3d at 303. “Class treatment, however, may not be suitable where the calculation of damages is not susceptible to a mathematical or formulaic calculation, or where the formula by which the parties propose to calculate individual damages is clearly inadequate.” Id. at 307. A plaintiff need not calculate a specific damage figure so long as he proposes an acceptable method for calculating damages.

See In re Visa Check/MasterMoney Antitrust Litigation, 280 F.3d 124, 138-39 (2d Cir.2001). The district court found that the damages factor counseled against certifying the class because Rodney’s experts “have admitted that they have not yet identified any methodology for computing damages.” Drs. Oster and Strong’s report contains a

section titled “methodology for the calculation of damages.” According to Rodney’s brief, “[t]his methodology compares the prices paid by class members who paid for tickets (the ‘fared average fare’) in the 74 Northwest ‘damage’ markets ... to benchmark prices that have been charged by all other ‘major’ or ‘national’ airlines in similar markets (the ‘base fare’), which are constructed from the U.S. Department of Transportation’s Databank 1A.” Dr.

Oster testified, however, that this calculation is not a methodology for computing damages but is instead a mechanism for identifying the markets in which to calculate damages. When asked “[s]o you have not identified a methodology for computing damages other than to identify the markets; is that correct?” Dr. Oster responded “[tjhat’s’ correct because we’ve not been asked to calculate the damages in those markets.” Dr.

Oster also testified that “if our task were to calculate damages for all passengers in the market, this would be one way you could proceed.” He added that “[wjhether we would use some method [to calculate damages] that resulted in the involved fared average fare or not, I really don’t want to say because [Dr.

Strong] and I have not sat down and said, okay, if we’re going to calculate damages, how should we do it, because we haven’t been given that task.” In recognition of this testimony, Rodney argues that his experts’ methodology could be used to calculate damages and that such a calculation would satisfy the requirements of Rule 23(b)(3). At first blush, it seems that Rodney cannot prove that class issues predominate the damages element because his experts have not agreed on a formula for computing damages.

Even assuming that Rodney’s experts would use the model appearing in his experts’ report, the element of damages doesn’t favor certification. The different variables that would be plugged into that formula are specific to the individual routes insofar as the formula requires us to calculate a “base fare,” which is the fare charged by Northwest’s competitors for a route of similar mileage. Moreover, as evidenced by the Report of Janusz A.

Ordover, Northwest’s expert, Northwest intends to rebut Rodney’s evidence by arguing that the routes that Northwest flies are not comparable to the “benchmark” prices used in Rodney’s proposed damage formula. This would involve analysis of the distinct characteristics of various Northwest routes. The damages element certainly does not favor class certification.

Conclusion Considering all of the elements of Rodney’s antitrust case, we do not have a definite and firm conviction that the district court committed a clear error of judgment when it concluded that issues of fact common to the class would predominate over any questions affecting only Rodney. If Rodney’s class were certified, the proofs would include evidence specific to each of 74 different routes and the class action would inevitably degenerate into 74 mini-trials. Accordingly, we AFFIRM the order of the district court denying Rodney’s motion to certify the class.

Because we affirm the district court on the ground that Rodney has failed to satisfy the requirements of Rule 23(b)(3), we need not, and will not, address the district court’s alternative holding that Rodney is not an adequate representative of the class as required by Rule 23(a)(3). . We note that our conclusion that Rodney has not defined the market is only relevant to the issue of whether a class should be certified and does not preclude the factfinder from ultimately concluding otherwise.

Gariety, 368 F.3d at 366 ("The juiy or factfinder can be given free hand to find all of the facts required to render a verdict on the merits, and if its finding of any fact differs from a finding made in connection with class action certification, the ultimate factfinder’s finding on the merits will govern the judgment”). . Because we affirm the district court’s order on the ground that Rodney has failed to show that common issues of law and fact will predominate, we do not address Rodney’s claim that "predation by reputation” theory can give rise to antitrust liability.