Opinion
13-cv-2680 (AJN)
09-30-2016
Maxon Hyundai Mazda, et al., Plaintiffs, v. Carfax, Inc., Defendant.
OPINION & ORDER :
Plaintiffs, 469 dealers of used cars, bring this antitrust suit against Carfax, Inc. ("Carfax"), the leading provider of vehicle history reports. Plaintiffs claim that Carfax's exclusive dealing agreements with car manufacturers and used car websites violate Sections 1 and 2 of the Sherman Act, 15 U.S.C. §§ 1 & 2. Before the Court is Carfax's motion for partial summary judgment on the question of whether the exclusive dealing agreements foreclosed a sufficient portion of the market to implicate the Sherman Act. For the reasons that follow, that motion is granted. I. BACKGROUND
The following facts are taken primarily from the parties' Local Rule 56.1 statements—i.e., Defendant's Statement of Material Facts as to Which There is No Genuine Issue to Be Tried ("Carfax 56.1"), Plaintiffs' Rule 56.1 Response to Defendant's Statement of Material Facts as to Which There is No Genuine Issue to Be Tried ("Pls. 56.1 Resp."), and Defendant's Reply to Plaintiffs' Rule 56.1 Response to Defendant's Rule 56.1 Statement of Facts ("Carfax 56.1 Reply"). The Court accepts as true any properly supported statement where the opposing party has either admitted the facts described in the statement or has failed to rebut those facts with citations to admissible evidence. See Giannullo v. City of New York, 322 F.3d 139, 140 (2d Cir. 2003). Where appropriate, the Court also relies on the evidence submitted to the Court for this motion. The Court declines Plaintiff's invitation to strike Carfax's 56.1 Reply, see Dkt. No. 164, particularly in light of the fact that both parties' responses to the other's 56.1 statements have ignored Rule 56.1's admonition to be "plain and concise."
There are 469 used car dealers currently named as Plaintiffs in this case. Pls.' 56.1 Resp. ¶ 1. That number includes both "franchise" dealers, who are associated with a specific car manufacturer, and independent dealers who are not. Id. At issue here is a series of agreements that allegedly harmed Plaintiffs by adversely affecting competition in the market for vehicle history reports ("VHRs"). A VHR consists of data about a particular vehicle, including data provided to authorized sellers by the National Motor Vehicle Title Information System ("NMVTIS"). Id. ¶¶ 10-10.1.
Carfax describes itself as "the first company to collect vehicle history information from multiple sources and compile it into vehicle history reports." Id. ¶¶ 2-2.1. Founded in 1984, Carfax now has more than 15 billion records of vehicle history data. Id. Carfax generally sells VHRs in one of three ways: individually, in packages of five, or through its "Advantage program," which provides dealers with access to unlimited VHRs after paying a price that is generally based on the average number of used cars in the dealer's inventory. Id. ¶ 3. Carfax has sold VHRs to many of the Plaintiffs in this case. Id.
Carfax and its principal competitor, Experian, are the largest sellers of VHRs in the United States. Pls. 56.1 Resp. ¶ 8-8.1. Experian sells VHRs under the brand name "AutoCheck." Id. Although the parties disagree over the difference in size and market power between Carfax and AutoCheck, they have stipulated for purposes of this motion that Carfax has a 90% market share in the VHR market and that it has monopoly power. Dkt. No. 151 at 2; see also Second Am. Compl. ("SAC") ¶¶ 543-45. Aside from AutoCheck, there are at least fourteen other companies that supply reports that include vehicle data and identify as Carfax's competitors. Pls. 56.1 Resp. ¶¶ 9-9.1. Whether all of those reports qualify as "VHRs" is in issue and is discussed infra.
One source of demand for VHRs is what are known as "Certified Pre-Owned," or "CPO" programs. Id. ¶¶ 1, 4. CPO programs are run by manufacturers, so only franchise dealers (associated with a particular manufacturer) sell used cars that have been certified under these programs. Id. Cars that are certified pursuant to a CPO program tend to be particularly desirable to consumers, as they are usually newer, have fewer miles, and have passed a mechanical inspection. Id. ¶ 4. Some manufacturers require that a car's CPO "qualifications" be established by data contained in VHRs. Id. ¶¶ 5-5.1. Accordingly, dealers that participate in a CPO program (or the manufacturer administering such a program) will often run a VHR for each car the dealer wants certified. Id. Additionally, many manufacturers require dealers to provide a VHR to each customer interested in a CPO car. Id. Although dealers can command a premium for CPO cars, the sale of CPO cars constitutes only about 4.5 to 6 percent of the total number of used cars sold in a given year. Id. ¶ 6. In keeping with those numbers, the VHRs that Carfax provided for vehicles in CPO programs constituted 3% of Carfax's total VHR sales for the years 2012 to 2014. Id. ¶ 15.
Carfax has agreements with a number of manufacturers that have CPO programs. Id. ¶ 16. These agreements—the "CPO Agreements"—generally require that a franchise dealer, or the manufacturer itself, purchase a Carfax VHR for each vehicle in the manufacturer's program. Id. ¶¶ 16-16.1, 17. The CPO Agreements also have provisions that affect co-promotional and co-marketing activity. Id. ¶¶ 16-16.1. The agreements do not require dealers to purchase all of their VHRs from Carfax, id. ¶ 18, but whether they create an incentive to do so is subject to dispute. Carfax entered into these CPO Agreements both before and during the period relevant to this litigation (i.e., the period from 2008 through 2013). Carfax 56.1 Reply ¶¶ 69-70.1.
As Plaintiffs note, the statute of limitations allows Plaintiffs to seek damages dating back to April 23, 2009. See 15 U.S.C. § 15b; see also Opp. Br. 1 n.1. Because some of the agreements at issue here were in place before 2009, and carried into the limitations period, it is appropriate to consider their effects in the context of this partial summary judgment motion. Cf. Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263, 296 (2d Cir. 1979) ("[A] purchaser suing a monopolist for overcharges paid within the previous four years may satisfy the conduct prerequisite to recovery by pointing to anticompetitive actions taken before the limitations period.").
During the relevant period, Carfax also had agreements with two key used car listing websites, Autotrader and Cars.com. Pls. 56.1 Resp. ¶¶ 54-55.1. These agreements—the "Website Agreements"—placed various restrictions on the ads that Autotrader and Cars.com could run with respect to VHR providers and on the information that dealers that posted on those sites could include in their listings. Carfax and Autotrader have had an agreement in place since at least 2005. Id. ¶¶ 54-54.1. They entered a new agreement in December 2008, and that agreement was in place (subject to two amendments) until 2013. Id. Prior to 2013, the agreements with Autotrader required, with one narrow exception, that dealers could link only to Carfax VHRs in used car listings they posted to the site. Id. The pre-2013 agreements also made Carfax the only VHR provider that Autotrader would advertise via promotional links around the site. Id. Carfax and Cars.com have likewise had agreements in place since at least 2005. Id. ¶¶ 55-55.1. They reached a new agreement in October 2010, and that agreement was in place until January 2014. Id. The pre-2014 agreements had similar terms to the exclusive arrangement with Autotrader. See id. The most recent agreements with both websites, which were entered into after this lawsuit was filed, have ended many of the exclusive benefits that Carfax enjoyed. See id. ¶¶ 54-55.1
Plaintiffs filed this suit on April 23, 2016. See Dkt. No. 1. Carfax moved to dismiss the second amended complaint on December 17, 2013. Dkt. No. 29. On September 29, 2014, the Court issued an order granting in part and denying in part the motion to dismiss. Dkt. No. 59. The Court dismissed Plaintiffs' claim that Carfax's exclusive dealing agreements violated Section 3 of the Clayton Act, 15 U.S.C. § 14. The Court declined, however, to dismiss the claims that Plaintiffs brought under Sections 1 and 2 of the Sherman Act. Carfax filed an answer on October 14, 2014, Dkt. No. 61, and the case proceeded to discovery on a lengthy schedule. On December 1, 2015, the parties jointly proposed permitting Carfax to file an expedited motion for partial summary judgment after a period of limited discovery. Dkt. No. 151. The proposed motion for partial summary judgment was to be limited to "the issue of whether Carfax's allegedly exclusive agreements with third parties, as alleged in the Second Amend Complaint, foreclose competition in the sale of VHRs in violation of Section 1 or 2 of the Sherman Act." Id. at 2. For purposes of the motion, the parties agreed to assume "that the issues of the relevant market and market or monopoly power as alleged in the Second Amended Complaint are resolved in Plaintiff's favor, as the non-movants." Id. The parties also agreed to a schedule for expert witness reports, in which each party was limited to "one expert report and one reply report." Id. at 5. The Court approved the parties' proposal on December 6, 2015, subject to the condition that a grant of partial summary judgment would lead to dismissal of the case (on the theory that Plaintiffs could not prevail on any of their claims if they could not prove substantial foreclosure of the VHR market) and that a denial of partial summary judgment would not trigger additional general summary judgment motions. Dkt. No. 152.
Carfax filed its motion for partial summary judgment on February 26, 2016. Dkt. No. 154. In addition to the parties' briefing on the motion, there are four expert reports before the court—an initial report and a reply report from Carfax's expert, Dr. Willig, and an initial report and rebuttal report from Plaintiffs' expert, Dr. Singer. Neither side has moved to exclude the other's expert. The briefing and exhibits on the partial summary judgment motion were filed under seal, and the parties have proposed a lengthy set of redactions. In light of the allegedly confidential business information discussed in this Opinion & Order, the Court is temporarily sealing this Opinion & Order so that the parties may submit any proposed redactions. The Court will rule on the proposed redactions to this Opinion & Order and the underlying materials simultaneously.
The parties' stipulation in advance of this partial summary judgment practice contemplated that both parties' experts would submit two reports and that all the briefs would be "a part of the record for summary judgment purposes." Dkt. No. 151 at 5. Accordingly, the Court rejects Plaintiffs' suggestion that it should strike the reply report of Carfax's expert for relying on new arguments and data. See Dkt. No. 164. The arguments and data to which Plaintiffs object were produced to address the methodologies and conclusions of Plaintiffs' expert. Similarly, the Court rejects Carfax's suggestion that the Court should not consider Plaintiffs' rebuttal report, see Dkt. No. 163, given that Carfax's expert did introduce new data in his reply report, which makes the rebuttal report contemplated in the parties' stipulation all the more appropriate.
II. LEGAL STANDARD
Summary judgment is appropriate when, after reviewing the parties' submissions in the light most favorable to the non-moving party, "there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law." Fed. R. Civ. P. 56(a). A fact is "material" if it "might affect the outcome of the suit under the governing law," and is genuinely in dispute if "the evidence is such that a reasonable jury could return a verdict for the nonmoving party." Roe v. City of Waterbury, 542 F.3d 31, 35 (2d Cir. 2008) (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986)). In seeking summary judgment, the initial "burden is upon the moving party to demonstrate that no genuine issue respecting any material fact exists." Gallo v. Prudential Residential Servs., 22 F.3d 1219, 1223 (2d Cir. 1994). If the movant "demonstrates the absence of a genuine issue of material fact, the opposing party must come forward with specific evidence demonstrating the existence of a genuine dispute of material fact" to survive summary judgment. Brown v. Eli Lilly & Co., 654 F.3d 347, 358 (2d Cir. 2011) (internal quotation marks and citations omitted). The opposing party cannot make such a showing "merely by making assertions that are conclusory or based on speculation." Major League Baseball Props., Inc. v. Salvino, Inc., 542 F.3d 290, 310 (2d Cir. 2008) (citations omitted); see also Anderson, 477 U.S. at 249-50 (summary judgment "may be granted" if the opposing evidence is "merely colorable" or "not significantly probative"). In ruling on a motion for summary judgment, "all ambiguities must be resolved and all inferences drawn in favor of the party against whom summary judgment is sought." Gallo, 22 F.3d at 1223. In the specific context of antitrust litigation, the Second Circuit has recognized that summary judgment "may be particularly important . . . to prevent lengthy and drawn-out litigation that has a chilling effect on competitive market forces." Capital Imaging Assocs., P.C. v. Mohawk Valley Med. Assocs., Inc., 996 F.2d 537, 541 (2d Cir. 1993). Accordingly, "the range of inferences that may be drawn from ambiguous evidence is limited." Id.
III. PLAINTIFFS' SECTION 1 CLAIM
Section 1 of the Sherman Act prohibits "[e]very contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce." 15 U.S.C. § 1. Despite the breadth of this language, Section 1 prohibits "only unreasonable restraints of trade." In re Publ'n Paper Antitrust Litig., 690 F.3d 51, 61 (2d Cir. 2012) (emphasis added) (quoting Bus. Elecs. Corp. v. Sharp Elecs. Corp., 485 U.S. 717, 723 (1988)). Certain restraints on trade, like horizontal price-fixing agreements, "are unlawful per se." MacDermid Printing Sols. LLC v. Cortron Corp., No. 15-589-CV, 2016 WL 4204795, at *5 (2d Cir. Aug. 10, 2016). Others, like the exclusive dealing that Plaintiffs allege here, "must be evaluated under the so-called 'rule of reason.'" Id.; see also Xerox Corp. v. Media Scis. Int'l, Inc., 511 F. Supp. 2d 372, 389 (S.D.N.Y. 2007) (exclusive dealing arrangements are subject to the rule of reason). Under the rule of reason, a restraint is unreasonable if its "anticompetitive effects outweigh its procompetitive effects." E & L Consulting, Ltd. v. Doman Indus. Ltd., 472 F.3d 23, 29 (2d Cir. 2006) (quoting Atl. Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 342 (1990)). To reach that conclusion, courts engage in a three-step inquiry: First, "the plaintiffs bear an initial burden to demonstrate the defendants' challenged behavior had an actual adverse effect on competition as a whole in the relevant market." Geneva Pharm. Tech. Corp. v. Barr Labs. Inc., 386 F.3d 485, 506-07 (2d Cir. 2004) (emphasis in original) (internal quotation marks and citation omitted). Second, "[i]f the plaintiffs satisfy their initial burden, the burden shifts to the defendants to offer evidence of the pro-competitive effects of their agreement." Id at 507. And third, "[a]ssuming defendants can provide such proof, the burden shifts back to the plaintiffs to prove that any legitimate competitive benefits offered by defendants could have been achieved through less restrictive means." Id. The factfinder must ultimately "engage in a careful weighing of the competitive effects of the agreement—both pro and con—to determine if the effects of the challenged restraint tend to promote or destroy competition." Id.
Exclusive dealing arrangements, like those at issue here, are "presumptively legal" under the Sherman Act. CDC Techs., Inc. v. IDEXX Labs., Inc., 186 F.3d 74, 80 (2d Cir. 1999) (citation omitted). That is because exclusive dealing can have a number of pro-competitive virtues, including "assuring steady supply, affording protection against price fluctuations, reducing selling expenses, and promoting stable, long-term business relationships." Geneva Pharm., 386 F.3d at 508. But exclusive dealing threatens competition if, for instance, it "allow[s] one supplier of goods or services unreasonably to deprive other suppliers of a market for their goods." Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 45 (1984) (O'Connor, J., concurring), abrogated on other grounds by Ill. Tool Works Inc. v. Indep. Ink, Inc., 547 U.S. 28 (2006). Such exclusive dealing arrangements can impair the ability of rival firms to expand, thereby preventing them from becoming better competitors or perhaps even forcing them to exit the market altogether. 11 Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 1802d5, at 84 (3d ed. 2011).
Determining whether a particular exclusive dealing arrangement is unlawful requires a careful analysis of the "the competitive characteristics of the relevant market." Geneva Pharm., 386 F.3d at 508. In particular, the Supreme Court has long held that "the competition foreclosed" by such an arrangement "must be found to constitute a substantial share of the relevant market" in order to violate the antitrust laws. Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 328 (1961) (emphasis added); see also Geneva Pharm., 386 F.3d at 508 ("Exclusive dealing is an unreasonable restraint of trade and a § 1 violation only when the agreement freezes out a significant fraction of buyers or sellers from the market."). Requiring plaintiffs who bring exclusive dealing claims to prove substantial foreclosure "serves a useful screening function" because "an exclusive deal affecting a small fraction of a market clearly cannot have the requisite harmful effect upon competition." United States v. Microsoft Corp., 253 F.3d 34, 69 (D.C. Cir. 2001). Drawing a bright line with respect to "substantial foreclosure" is difficult. Nonetheless, courts have generally held that "foreclosure levels are unlikely to be of concern where they are less than 30 or 40 percent." Discover Fin. Servs. v. Visa U.S.A. Inc., 598 F. Supp. 2d 394, 406 (S.D.N.Y. 2008) (quoting Stop & Shop Supermarket Co. v. Blue Cross & Blue Shield of R.I., 373 F.3d 57, 68 (1st Cir. 2004)); see also Areeda & Hovenkamp, supra, ¶ 1821c, at 190 ("Picking the correct number is somewhat arbitrary, but single-firm foreclosure percentages of less than 30 or 40 percent . . . would seem to be harmless to competition."). This Court agrees with this authority.
Other factors courts consider in assessing substantial foreclosure include "the duration and terminability of the arrangement," as "the shorter an agreement's term and the easier it is to terminate, the more likely that it will be upheld." Am. Express Travel Related Servs. Co. v. Visa U.S.A., No. 04-CIV-8967 (BSJ), 2005 WL 1515399, at *3 (S.D.N.Y. June 23, 2005). Short-term or at-will exclusive dealing contracts generally do not threaten competition because when they expire, a firm's rivals can bid to take over the contract. See CDC Techs., 186 F.3d at 81; Balaklaw v. Lovell, 14 F.3d 793, 799 (2d Cir. 1994). But the duration of exclusivity appearing on the face of a contract may not tell the whole story if switching to a competitor would prove costly for other reasons. See Areeda & Hovencamp, supra, ¶ 1802g3, at 101-03; see also Maxon Hyundai Mazda v. Carfax, Inc., No. 13-CV-2680 (AJN), 2014 WL 4988268, at *11 (S.D.N.Y. Sept. 29, 2014). Courts also consider whether "competitors can reach the ultimate consumers of the product by employing existing or potential alternative channels of distribution." CDC Techs., 186 F.3d at 80-81 (citation omitted). If so, then the exclusive dealing has not truly foreclosed competitors from the marketplace. Id. Finally, courts ask whether "there are significant barriers to entry into [the] market," which heightens the risk posed by exclusive dealing. Id.
A. Overview of Section 1 Analysis
As outlined above, the parties agreed to limit this motion for partial summary judgment to the issue of whether the exclusive agreements described in the Second Amended Complaint "foreclose competition in the sale of VHRs" in violation of the Sherman Act. Dkt. No. 151 at 2; see also Dkt. No. 152. In order to limit the briefing to that particular issue, the parties stipulated to certain allegations in the Second Amended Complaint for purposes of this motion. Specifically, they agreed that "the issues of the relevant market and market or monopoly power as alleged in the Second Amended Complaint are resolved in Plaintiffs' favor." Dkt. No. 151 at 2. In light of those stipulations, the Court assumes that the relevant market is "[t]he publication and sale of VHRs in the United States." SAC ¶ 539. The Court adopts the Second Amended Complaint's definition of VHRs as including, "at a minimum," five key pieces of information required by the NMVTIS. Id. ¶ 542. And the Court assumes that Carfax "possess a 90 market share" in the relevant market and submarkets, and that it "possesses monopoly power" in those markets. Id. ¶¶ 543-44. "Monopoly power" is "the power to control prices or exclude competition." Geneva Pharm., 386 F.3d at 500 (citation omitted).
Carfax's motion for partial summary judgment seeks to establish that "Carfax's alleged conduct could not implicate enough of the market to support a Sherman Act claim." Carfax Br. 2. In other words, Carfax argues that no reasonably jury could find that Carfax's agreements with classified websites and manufacturers foreclosed a substantial share of the market for VHRs. The question of "substantial foreclosure" goes only to step one of the rule of reason analysis—i.e., to Plaintiffs' initial burden to prove that Carfax's exclusive deals "had an actual adverse effect on competition as a whole in the relevant market." Geneva Pharm., 386 F.3d at 506-07 (emphasis in original). Carfax has not sought summary judgment with respect to steps two or three—i.e., where the burden shifts to Carfax "to offer evidence of . . . pro-competitive effects" and then back to Plaintiffs "to prove that any legitimate competitive benefits offered by defendants could have been achieved through less restrictive means." Id. at 507. Accordingly, the Court limits its consideration of Plaintiffs' Section 1 claim to the sole issue of whether Carfax's alleged exclusive dealing arrangements substantially foreclosed competition in the overall market for VHRs.
To be sure, Carfax makes a brief argument as to one pro-competitive benefit of its CPO Agreements, see Carfax Br. 21, but it declines to respond to Plaintiffs' claim that any pro-competitive justification is "beyond the scope of this Motion." Opp. Br. 33.
Within the confines of that overarching issue, the parties focus their briefing on two principal questions. First, are the alleged exclusive agreements the kind of agreements that are capable of raising antitrust concerns—i.e., are they truly "exclusive" and do they foreclose competition? Second, if the agreements do foreclose competition, is that foreclosure "substantial?" The Court considers each question in turn. It ultimately concludes that no reasonable jury could find that Carfax's agreements foreclosed a substantial share of the market for VHRs and that Carfax is accordingly entitled to summary judgment on that issue.
B. A Reasonable Jury Could Conclude that Some, Though not All, of the Challenged Agreements Were "Exclusive" and Foreclosed Competition
Carfax argues that partial summary is appropriate because the agreements that Plaintiffs challenge—the CPO Agreements and the Website Agreements—could not have unlawfully foreclosed competition. Rather, Carfax contends, the specific characteristics of the agreements, such as their relatively short duration and the fact that they did not directly prohibit dealers from buying VHRs from competitors, render the agreements reasonable as a matter of law. In other words, Carfax claims that the Court need not decide how much of the market for VHRs was foreclosed because Carfax's competitors were not truly prevented from competing in the market for VHRs. The Court addresses Carfax's arguments with respect to each category of agreements.
1. CPO Agreements
Carfax contends that the majority of the CPO Agreements it entered into with manufacturers could not have unlawfully foreclosed competition for two reasons. First, it claims that the agreements were not truly exclusive, and second, it argues that the agreements were not of sufficient duration to raise competitive concerns. With respect to the first reason, Carfax notes that the CPO Agreements only required exclusivity in one situation: "when the dealer was attempting to certify a vehicle or selling a CPO vehicle under that [manufacturer's] CPO program." Carfax Br. 18. But dealers and manufacturers were not prohibited, Carfax maintains, from purchasing another VHR for the same vehicle. Id. This latter contention is not in dispute. See Pls. 56.1 Resp. ¶ 20.1 (acknowledging the prospect of "pull[ing]" both "AutoCheck and Carfax reports"). Nor is the fact that at least some dealers used both Carfax and AutoCheck—in fact, at least half of Plaintiffs bought at least one VHR from AutoCheck during the period relevant to this litigation. See id. ¶ 21-21.1. Nonetheless, by requiring manufacturers or dealers to purchase a Carfax VHR for a CPO vehicle, see Carfax 56.1 ¶ 16, the CPO Agreements, Plaintiffs argue, decreased the need for a manufacturer or dealer to purchase a VHR for that same car from AutoCheck or another competitor. Plaintiffs offer little evidence in support of this possibility—their expert merely asserts that "[a]lthough a dealer could purchase an additional VHR from a rival for a given CPO vehicle, such a purchase would be redundant." Singer Rep. ¶ 38. Nevertheless, as this Court acknowledged in resolving the motion to dismiss, it is reasonable to infer that "a dealer that has already purchased a VHR from [Carfax] under compulsion from a manufacturer is highly unlikely to purchase a second VHR containing the same NMVTIS data." Maxon Hyundai Mazda, 2014 WL 4988268, at *11. Bearing in mind that the "practical effect" is what matters in assessing an exclusive agreement, Tampa Elec., 365 U.S. at 326, a reasonable jury could that find that the CPO Agreements had the effect of creating exclusivity.
Carfax's second argument—that the CPO Agreements were not long enough to be anticompetitive—is stronger at this stage. When exclusive agreements are "easily terminable on short notice," they tend not to foreclose competition. CDC Techs., 186 F.3d at 81. To the contrary, short-term exclusive deals can actually encourage competitors "to improve the [services] and prices they offer in order to secure the exclusive positions." Balaklaw, 14 F.3d at 799. As with other aspects of exclusive dealing cases, drawing a bright line with respect to contract duration is difficult. Generally speaking though, exclusive deals covering "periods of less than one year" should presumptively "be approved, particularly where switching appears to face no substantial impediments." Areeda & Hovenkamp, supra, ¶ 1821d3, at 200 & n.60 (citing cases); cf. Roland Mach. Co. v. Dresser Indus., Inc., 749 F.2d 380, 395 (7th Cir. 1984) ("Exclusive-dealing contracts terminable in less than a year are presumptively lawful under section 3 [of the Clayton Act]."). Some courts have gone even further and held that particular exclusive deals of longer duration were permissible as a matter of law. See, e.g., Spinelli v. Nat'l Football League, 96 F. Supp. 3d 81, 117 (S.D.N.Y. 2015) (holding that contracts with "exclusivity periods of no more than three years . . . do not foreclose competition and are not anticompetitive as a matter of law" where the contracts were subject to re-bidding at the end of each term).
Despite this general presumption against condemning short-term exclusive deals, courts have been careful to note that the length of an exclusive agreement "is not dispositive of whether it violates § 1," even if the contract "is terminable at will." Am. Express Travel, 2005 WL 1515399, at *6. In order to find that short-term deals foreclose competition, however, there must be other factors that could lead a reasonable jury to conclude that the practical effect of the agreement was to prolong exclusivity. For instance, in Minnesota Mining & Manufacturing. Co. v. Appleton Papers, Inc. ("3M"), a case relied on by Plaintiffs, the court held that there were "genuine issues of fact as to whether [the defendant's] agreements [were] actually terminable at will." 35 F. Supp. 2d 1138, 1144 (D. Minn. 1999). The agreements at issue involved "sole-source relationships" between the defendant, which supplied carbonless paper, and paper merchants. Id. at 1140. In reaching its holding that those agreements might not be terminable at will, the court pointed to evidence that (1) some of the agreements contained incentives that tied up merchants' paper inventory for several years, (2) there were high costs for merchants to switch from one brand to another, (3) merchants who did switch could no longer use the defendant, which possessed high market share and a well-known brand, as a supplier, and (4) "major distributors" had agreed among themselves to continue to distribute the defendant's product, irrespective of competitors' price or quality. Id. at 1144. Impediments along these or similar lines are necessary to permit the inference that a short-term exclusive agreement nonetheless foreclosed competition.
Applying these principles to the CPO Agreements, the Court concludes that at least a majority of the agreements did not foreclose competition as a matter of law. Carfax has divided the CPO Agreements into two groups: (1) the agreements that were only "terminable in more than a year"—meaning agreements that both lasted more than a year and could not be terminated without cause on short notice—and (2) the agreements that could be ended more easily (the "short-term" CPO Agreements). Carfax Br. 17. At least 21 manufacturers had CPO Agreements with Carfax at some point from April 23, 2009, through 2014. See Pls. 56.1 Resp. ¶¶ 22-37. Of all the CPO Agreements Carfax entered into with those manufacturers, there is evidence that only agreements with ten manufacturers were terminable in more than a year. See id. Confronted with that evidence, Plaintiffs have admitted the relevant allegations as to contract length and terminability. See id. In other words, there is no genuine dispute that the majority of Carfax's CPO Agreements were subject to terms that courts generally find not to foreclose competition.
And of those ten, at least four were terminable within less than a year after the first year of the agreement. See Pls. 56.1 Resp. ¶¶ 23-23.1, 26-27.1, 29-29.1.
The question, then, is whether there are other characteristics of these short-term CPO Agreements, or of the market for VHRs generally, that would allow a reasonable jury to conclude that the agreements had the practical effect of prolonging the period of exclusivity. Plaintiffs claim that the CPO Agreements had the effect of "lock[ing] in franchise dealers for extended periods of time." Opp. Br. 28. But a reasonable jury could not reach that conclusion based on this record. First, Plaintiffs emphasize that terminating the CPO Agreements was not "costless." Id. at 29 (emphasis omitted); see also Singer Rep. ¶ 101. But the existence of transition costs is not enough for a reasonable jury to find that a short-term contract was actually a long-term exclusive deal. That is because it is generally the role of competitors in a marketplace to compete for the contract by offering incentives to get customers to switch—for instance, as one treatise notes, if one supplier offers a discount to customers who deal exclusively, "a rival can steal a customer at any time simply by matching the discount." Areeda & Hovenkamp, supra, ¶ 1821d3, at 201. Here, Plaintiffs claim that Carfax "imposed a steep penalty price for rejecting exclusivity." Opp. Br. 29. But the only benefits of exclusivity that Plaintiffs identify are: (1) the ability to use the Carfax brand, (2) the amount of support from Carfax (non-exclusive CPO programs got less), and (3) the pre-screening of cars in the CPO program. Id.; see also Singer Rep. ¶ 101. Plaintiffs point to no evidence that these benefits are so substantial that they would have the practical effect of locking a dealer or manufacturer into staying with Carfax.
By way of comparison, in United States v. Dentsply Int'l, Inc.—another case cited by Plaintiffs—the Third Circuit treated Dentsply's sales of teeth to dealers as exclusive dealing arrangements even though the transactions were "technically only a series of independent sales." 399 F.3d 181, 193 (3d Cir. 2005). In reaching that conclusion, the court highlighted Dentsply's track record of enforcing a policy against dealers that required the dealers not to offer additional tooth lines if they wanted to continue to offer Dentsply products. See id. at 185. Given Dentsply's large market share, this "realistically ma[de] the arrangements . . . as effective as those in written contracts." Id. at 193; see also PNY Techs., Inc. v. SanDisk Corp., No. 11-CV- 04689 (WHO), 2014 WL 2987322, at *8 (N.D. Cal. July 2, 2014) (emphasizing that Dentsply's threat to stop supplying its products to dealers was the salient factor in finding that the arrangement in that case was not truly "at will"). Here, by contrast, Plaintiffs point to no evidence in the record of conduct on Carfax's part that could be deemed the kind of all-or-nothing threat that Dentsply made to its distributors. Nor do Plaintiffs identify evidence that the market for VHRs created the kinds of pressures that distributors faced if they rejected exclusivity with Dentsply. To be sure, Plaintiffs do point to an Experian internal strategy memo that describes Carfax as the "incumbent" with most major classified sites, CPO programs, and dealers, and that describes dealers as [Redacted] Ex. 234 at Maxon-Experian000585. But again, while this is evidence that Carfax reaped the benefits of incumbency, and that switching away from Carfax was not "costless," it is not affirmative evidence that the costs of switching were so high that the usual methods competitors employ to dislodge an incumbent would have been ineffective in the VHR market. See Omega Envtl., Inc. v. Gilbarco, Inc., 127 F.3d 1157, 1164 (9th Cir. 1997) ("We agree with the unremarkable proposition that a competitor with a proven product and strong reputation is likely to enjoy success in the marketplace, but reject the notion that this is anticompetitive. It is the essence of competition."); cf. Eisai, Inc. v. Sanofi Aventis U.S., LLC, 821 F.3d 394, 407 (3d Cir. 2016) ("[T]he threat of a lost discount is a far cry from the anticompetitive conduct at issue in . . . Dentsply."). Plaintiffs would have to make some affirmative showing of significant barriers to switching in order for any reasonable jury to conclude that the practical effect of the CPO Agreements was to prolong exclusivity.
Second, Plaintiffs emphasize the fact that it is the manufacturers—not the dealers who participate in the manufacturers' CPO programs—who decide whether to terminate an exclusive arrangement with Carfax. Opp. Br. 29-30. Plaintiffs claim that manufacturers "have little incentive to seek the best prices or the most favorable terms regarding VHRs for all participating dealers." Id. at 30. But Plaintiffs point to no evidence in the record that manufacturers had an incentive to ignore the interests of their CPO dealers—after all, these were franchise dealers who used the manufacturers' brands and applied the "Certified Pre-Owned" label to the manufacturers' cars. This lack of evidence is in contrast with the one case Plaintiffs cite on this point, Burlington Drug Co. v. VHA, Inc., No. 2:95-CV-15, 1998 U.S. Dist. LEXIS 17787 (D. Vt. Sept. 3, 1998). In Burlington, the court declined to grant summary judgment for the defendant on an exclusive dealing claim, even though the agreements at issue could have been terminated in 90 days, without cause. Id. *22, *28. The defendant in Burlington—a cooperative of about 1,600 hospitals and health care organizations—used its purchasing power to obtain discounts on certain drugs from pharmaceutical wholesalers. Id. at *2-3. In order for an individual hospital to actually have access to the lower prices, that hospital had to agree to purchase all of its requirements for certain drugs from the same wholesaler. Id. at *3. But only the hospital cooperative, not the hospitals themselves, had the ability to cancel any contract with the pharmaceutical wholesalers. Id. at *29-30. In rejecting the defendant's argument that these contracts were too short to be exclusive dealing arrangements, the court found that there was sufficient evidence for a jury to conclude that the interests of the hospital cooperative and the individual hospitals were not the same. Id. at *30. Again, Plaintiffs have identified no such evidence here.
Moreover, Plaintiffs also have not identified evidence that AutoCheck and other VHR providers were incapable of competing for exclusive deals with the manufacturers themselves, which is what Plaintiffs would have to prove for a reasonable jury to conclude that manufacturers were somehow locked in to short-term CPO Agreements. See Pls. 56.1 Resp. ¶ 50.1 (declining to challenge evidence that Carfax's cash payments to manufacturers were modest). All told, Plaintiffs have failed to point to any of the kind of evidence that courts have relied on in cases like 3M and Dentsply to establish that the short-term CPO Agreements could have had the practical effect of prolonging the period of exclusivity. Given this absence in the record, no reasonable jury could find that the short-term CPO Agreements foreclosed competition. And as explained below, infra at III(C)(2), a reasonable jury would therefore not include the VHRs that were provided as part of any short-term CPO Agreements in assessing the extent of any foreclosure in the market for VHRs.
As will also be discussed infra at III(C)(2), the Court accepts, for purposes of this motion, that the remaining agreements with ten manufacturers were of sufficient duration to potentially foreclose competition. A reasonable jury could therefore rely on those CPO Agreements—but not the short-term CPO Agreements—in assessing how much of the VHR market was foreclosed.
2. Website Agreements
Carfax's principal contention with respect to the Website Agreements is that, because Autotrader and Cars.com are "advertising platforms," Carfax's agreements with them cannot have foreclosed competition in the VHR market. Carfax Br. 23. Before turning to the specifics of Carfax's argument on this point, it is helpful to understand how the Website Agreements affected dealers and consumers who used these two sites. The Website Agreements had three components that are relevant here. First, if a dealer listing a car on Autotrader or Cars.com wanted to include a link to a VHR in the listing, the VHR had to be from Carfax. See Pls. 56.1 Resp. ¶¶ 54-55.1. Second, the Website Agreements also provided that Carfax would be the only VHR provider advertised on Autotrader and Cars.com. Id. This advertising took at least two forms—first, generalized ads on the sites (i.e., "banner ads"), and second, referral links to Carfax on individual listings, if that listing did not already include a link to a VHR. Id. Third, Carfax was the only VHR provider that would be offered to private owners who wanted to a) list their car on one of the websites and b) include a link to a VHR in their listing. Id.; see also Koenig Decl. ¶ 68. In other words, someone shopping for a used car on Autotrader or Cars.com would have encountered one of three situations when browsing a particular listing. First, if it was a dealer listing that included a link to a VHR, then that VHR was from Carfax (except in the narrow situation described supra). Second, if it was a listing from a private owner who had chosen to include a link to a VHR, then the VHR was from Carfax. And third, if it was a listing that did not include a VHR, regardless of whether it was posted by a dealer or a private owner, then there would be a referral link to allow the consumer to purchase a VHR, and that link would direct the consumer to Carfax. With that framework in mind, the Court considers Carfax's contention that the Website Agreements cannot form the basis for a Section 1 claim because they did not foreclose competition.
There was one narrow exception to this requirement. If a dealer participated in a CPO program where it was required to use a VHR provider other than Carfax, then the dealer could list cars from that CPO program on Autotrader or Cars.com with a link to a non-Carfax VHR. See Koenig Decl. ¶¶ 66, 72. There is no genuine dispute that these agreements affected a small share of the market for VHRs purchased through CPO programs, see Carfax 56.1 Reply ¶¶ 71-71.1, and therefore an even smaller share of the market for VHRs for cars that were listed on Autotrader or Cars.com.
As an initial matter, Carfax is right that some components of the Website Agreements could not have foreclosed competition. Specifically, the parts of the agreements pertaining to advertisements that offered consumers on Autotrader and Cars.com the option to purchase a VHR—i.e., the banner ads and the referral links in listings that did not already include a VHR—did not contribute to any foreclosure. These components of the Website Agreements did not foreclose competition because no reasonable jury could find that they prevented Carfax's competitors from reaching consumers. As long as "competitors can reach the ultimate consumers of [a] product by employing existing or potential alternative channels of distribution, it is unclear whether exclusive dealing arrangements with distributors foreclose from competition any part of the relevant market." CDC Techs., 186 F.3d at 80 (emphasis in original) (alterations omitted). Applying that principle to exclusive advertising agreements, the court in Wellnx Life Sciences Inc. v. Iovate Health Scis. Research Inc., held at summary judgment that where the plaintiff "ha[d] alternative channels of advertising to the market," the defendant's exclusive deals with advertisers did not unlawfully foreclose competition. 516 F. Supp. 2d 270, 295 (S.D.N.Y. 2007). Here, there is no dispute that AutoCheck had a variety of ways to advertise its VHRs to individual consumers shopping for used cars. Those methods included AutoCheck's website and paid search-and-display advertising—i.e., the ads one sees when searching for used cars on Google. Pls. 56.1 Resp. ¶¶ 63-63.2. Accordingly, a reasonable jury could not conclude that these components of the Website Agreements foreclosed competition.
A reasonable jury could conclude, however, that one component of the Website Agreements did foreclose competition—the requirement that dealers link only to Carfax VHRs in their listings. If dealers wanted to include a link to a VHR when they posted a car on Autotrader or Cars.com, this component of the Website Agreements made purchasing a Carfax VHR a precondition of advertising on those sites. As such, a reasonable jury could find that the Website Agreements had the "practical effect" of locking those dealers into buying Carfax VHRs. See Tampa Elec., 365 U.S. at 326. The advertising agreement at issue in Wellnx, by contrast, did not have such effects. There, a manufacturer of bodybuilding supplements challenged an agreement that merely gave one of its rivals preferential advertising treatment from the publisher of bodybuilding magazines. 516 F. Supp. 2d at 280-81. In dismissing the plaintiff's claims, the Wellnx court emphasized that consumers did not buy bodybuilding supplements directly from magazines and that the plaintiff's products could still reach consumers through distributors and retailers. Id. at 295. In other words, there was no prospect that the exclusive advertising agreement locked in any particular consumer to purchasing supplements from the plaintiff's rival. Here, by contrast, a jury could find that the Website Agreements did have such an effect on would-be consumers—i.e., the dealers who wanted to include a VHR in their Autotrader and Cars.com listing. As with the CPO Agreements, it is perhaps true that AutoCheck could have competed for the business of these dealers by trying to sell them a second VHR for the car they were listing. But as the Court has already explained, a reasonable jury could find that the value of a second VHR would have been significantly diminished. That is particularly true here, given that the Website Agreements prevented dealers from listing a second (i.e., a non-Carfx) VHR. Accordingly, a reasonable jury could find that the Website Agreements foreclosed competition, but only to the extent they affected dealers who sought to list cars on Autotrader or Cars.com.
C. The Website Agreements and CPO Agreements Could Not, as a Matter of Law, Have Collectively Foreclosed a Substantial Share of the VHR Market
Having determined that a reasonable jury could find that the Website Agreements and some of the CPO Agreements foreclosed competition, the remaining question is whether a reasonable jury could find that foreclosure to be substantial. Ultimately, and as explained in greater depth below, the answer to that question depends on more than just a numerical assessment of market foreclosure. But the percentage of the market for VHRs that was foreclosed is certainly an important factor in evaluating Plaintiffs' Section 1 claim. Accordingly, the Court first addresses how to determine the "market for VHRs" and then turns to the question of what percentage of that market was potentially foreclosed by the CPO Agreements and the Website Agreements. Finally, the Court considers whether the evidence would be sufficient for a reasonable jury to conclude that the agreements substantially foreclosed competition in the market for VHRs.
1. Defining the Market for VHRs
As explained previously, the parties stipulated to the definition of the relevant market for this partial summary judgment motion. See Dkt. No. 151 at 2; see also SAC ¶ 539 (defining the relevant market as "[t]he publication and sale of VHRs in the United States"). Despite this stipulation, Plaintiffs have attempted to tweak the market definition in their summary judgment submissions, which helped expose the weakness of their argument. For example, Plaintiffs' expert, Dr. Singer, defined the market for VHRs as the market for reports that provide information comparable to the reports supplied by Carfax and Autocheck. See Singer Rep. ¶ 47. Such reports, Dr. Singer argues, are "distinct from the reports offered by the National Motor Vehicle Title Information System (NMVTIS)." Id. ¶ 48. Dr. Singer explains that the information contained in NMVTIS reports is limited to that which can be "gleaned from governmental records"—namely, "(1) current title and last title date, (2) brand history, (3) odometer reading, (4) total loss history, (5) and salvage history." Id. But "VHRs," according to Dr. Singer, include additional information, like details on accident and repair history (if the accident resulted in less than a total loss), emissions data, airbag deployments, and auction history. Id. Dr. Singer therefore concludes that "NMVTIS report are better understood as inputs to VHRs," rather than competitive substitutes. Id.
The problem with this conclusion, as Carfax notes, is that it is inconsistent with the parties' stipulation. See Reply Br. 9. The portion of the Second Amended Complaint that defines the relevant market is titled "Monopoly Power in Relevant Market and Submarkets." SAC at 111. Under that heading, and just a few paragraphs after alleging that the relevant market is "[t]he publication and sale of VHRs in the United Sates," Plaintiffs explain that VHRs "are required by the NMVTIS to provide, at a minimum, information on five key indicators of the quality and condition" of a used car. Id. ¶¶ 539, 542. In other words, a report that provides information on those indicators meets the Second Amended Complaint's "minimum" definition of a VHR. The five indicators outlined in the Second Amended Complaint are the same ones that Dr. Singer identified as characteristics of NMVTIS reports: "(1) date of last title and name of state titling agency; (2) 'brand' history applied by state titling agency, including 'junk', 'salvage', and 'flood'; (3) odometer reading; (4) total loss history; and (5) salvage history." Id. ¶ 542. Allegations from earlier in the Second Amended Complaint (outside the section that actually defines the market) reaffirm that reports providing NMVTIS data are alleged to be VHRs. Plaintiffs allege that "[t]he NMVTIS web site . . . encourages consumers shopping for used cars to obtain privately prepared VHRs containing all information required by NMVTIS [and] supplied by any of ten NMVTIS-approved" companies. Id. ¶ 486 (emphases added). Plaintiffs then go on to list "[t]he ten approved vendors of NMVTIS VHRs." Id. ¶¶ 488. The fact that those vendors are defined by NMVTIS approval, and not by whether they provide information beyond NMVTIS data, supports the conclusion that reports providing NMVTIS data were part of the market for VHRs that was alleged in the Second Amended Complaint. While Plaintiffs may now have concluded that VHRs should be defined more narrowly, allowing them to rely on that narrower definition for purposes of this motion would negate the purpose of stipulating to a market definition in the first place. The Court therefore holds, for purposes of this motion and consistent with the parties' stipulation, that the market for VHRs includes reports that contain the data required by NMVTIS. The Court next assesses the possible foreclosure in that market.
2. Measuring the Percentage of Foreclosure
Summary judgment in exclusive dealing cases is appropriate when a plaintiff fails to offer evidence that exclusive agreements foreclosed a large enough share of the market to raise a reasonable inference that the agreements harmed competition. See, e.g., Discover Fin. Servs., 598 F. Supp. 2d at 406 (granting defendant's summary judgment motion on exclusive dealing claim and explaining that "foreclosure levels are unlikely to be of concern where they are less than 30 or 40 percent" (citation omitted)). Accordingly, Carfax argues that its motion for partial summary judgment should be granted because there is insufficient evidence in the record for a reasonable jury to conclude that the CPO Agreements and the Website Agreements foreclosed a high enough percentage of the market for VHRs to raise an anticompetitive inference. Plaintiffs make two principal arguments in response: first, that Carfax has failed to carry its initial burden at summary because it failed to introduce evidence of the foreclosure share for several of the years relevant to this case; and second, that there is sufficient evidence of market foreclosure, in the form of Plaintiffs' expert report. The Court considers each argument in turn.
First, Plaintiffs claim that the Court cannot grant partial summary judgment to Carfax because it failed to offer evidence of the degree of foreclosure in the VHR market for any year before 2012. Opp. Br. 11. Specifically, Plaintiffs note that Carfax's expert economist conducted a foreclosure analysis from 2012 to 2014, but did not attempt to quantify the amount of foreclosure for previous years. Id. Plaintiffs' contend that this alleged failure means that Carfax has not met its initial burden as the party seeking summary judgment. Id.; see also Gallo, 22 F.3d at 1223 ("[T]he burden is upon the moving party to demonstrate that no genuine issue respecting any material fact exists."). The Court disagrees. Because Plaintiffs will bear "the burden of proof at trial, summary judgment should be granted if [Carfax] can point to an absence of evidence to support an essential element of the nonmoving party's claim." Gemmink v. Jay Peak Inc., 807 F.3d 46, 48 (2d Cir. 2015) (internal quotation marks omitted), cert. denied, 136 S. Ct. 1684 (2016). The Court agreed to allow this partial summary judgment motion because the issue of substantial foreclosure in an essential element of Plaintiffs' exclusive dealing claims. Accordingly, Carfax can carry, and has carried, its burden by pointing to a lack of admissible evidence on the issue of foreclosure for the years prior to 2012. But more to the point, Carfax has produced evidence that supports a lack of foreclosure during those years. It points to evidence that the CPO Agreements were sufficiently short-term and/or terminable that they could not have foreclosed competition, see Carfax Br. 17 (citing Carfax 56.1 ¶¶ 22-31), and that the Website Agreements did not foreclose competition because AutoCheck could reach consumers through other methods, see id. at 29. To be sure, the Court has already concluded that a reasonable jury could disagree with Carfax's interpretation of those pieces of evidence, but the evidence certainly satisfies any initial burden of production that Carfax had. Finally, Carfax argued in its opening brief that although it did not calculate a particular foreclosure share for 2009-2011 (apparently because it did not "have the same data available" for those years), there was reason to believe that the foreclosure share calculated in 2012 (23%) was higher than the foreclosure share could have been for preceding years because "Autotrader's listings were growing" during that time. Carfax Br. 25 n.23. All told, the fact that Carfax did not provide market foreclosure data for 2009-2011 does not mean Plaintiffs automatically defeat Carfax's summary judgment motion for those years.
The one Second Circuit case that Plaintiffs cite on this point, Giannullo v. City of New York, is not to the contrary. In Giannullo, the district court had granted summary judgment to the defendants on plaintiff's claim that he was arrested without probable cause. See 322 F.3d 139, 140 (2d Cir. 2003). The Second Circuit reversed on the grounds that the "record [did] not support the district court's determination that defendants were entitled to summary judgment as a matter of law, either on the basis of probable cause or qualified immunity." Id. at 143. Both of those defenses, however, required the defendants to point to affirmative evidence in the record. It is therefore irrelevant that the Giannullo court emphasized that "summary judgment must be denied" where "the movant fails to fulfill its initial burden" because the defendants had an initial burden of identifying evidence in the record in that case. Id. at 140-41 (internal quotation marks omitted).
Plaintiffs' second argument is that, even if Carfax carried its initial summary judgment burden, Plaintiffs have nonetheless carried their own burden because their expert, Dr. Singer, found that Carfax foreclosed a substantial share of the market for VHRs. See Opp. Br. 11-12. The key conclusion from Dr. Singer's report is that Carfax foreclosed on average 30% of the VHR market through its exclusive agreements from 2008 through 2013 (the last year Carfax's exclusive agreements with Autotrader and Cars.com were in effect). Singer Rep. ¶ 50 & Table 1. That Dr. Singer reached this conclusion does not, standing alone, defeat Carfax's partial summary judgment motion. After all, "[t]he fact that expert testimony is admissible and even rigorous does not necessarily entail that it will create a fact issue." 2 Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 309c2, at 214 (4th ed. 2014). Rather, if "an expert opinion is not supported by sufficient facts to validate it in the eyes of the law, or when indisputable record facts contradict or otherwise render the opinion unreasonable, it cannot support a jury's verdict." Brooke Grp. Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 242 (1993); cf Capital Imaging, 996 F.2d at 541 (emphasizing importance of summary judgment in antitrust cases "to prevent lengthy and drawn-out litigation that has a chilling effect on competitive market forces"). Such is the case here. For the reasons that follow, the Court holds that a reasonable jury could not adopt Dr. Singer's conclusion that Carfax's exclusive agreements foreclosed 30% of the VHR market during the relevant years.
Moreover, and as explained infra, even reaching the 30% foreclosure threshold would not necessarily be sufficient for Plaintiffs to defeat Carfax's partial summary judgment motion.
There are two reasons why no reasonable jury could rely on Dr. Singer's 30% foreclosure finding. First, Dr. Singer's attempt to estimate the foreclosure share for the VHR market is missing key data from 2008 through 2011. The foreclosure share for the VHR market, according to Dr. Singer, is "the quotient obtained by dividing (1) the number of Carfax VHRs implicated by its exclusive agreements with Autotrader and Cars.com and its exclusive agreements with [manufacturers] regarding their CPO programs in a given year by (2) the total number of VHRs sold by all parties in that year." Singer Rep. ¶ 50. In other words, the numerator is the number of VHRs allegedly foreclosed by exclusive dealing, and the denominator is the number of VHRs in the overall market. To arrive at his figure for the denominator, Dr. Singer used an approach that relied on "Carfax's total VHR sales" for a given year and Carfax's share of annual revenue in the VHR market. Id. ¶ 57. The exact mechanics of the approach are not important, but what does matter is that if Dr. Singer had used a larger number for "Carfax's total VHR sales," then the denominator in his quotient would have been larger—i.e., because Carfax's VHR sales were used to estimate the overall size of the market, using a larger figure for those sales would have resulted in a larger estimate of the market. As Dr. Willig points out in his reply report, however, the data Dr. Singer used to calculate Carfax's total VHR sales in 2012 and 2013 is different from the data he used to calculate Carfax's total VHR sales for 2008 to 2011. Willig Reply Rep. ¶ 34. Specifically, Dr. Singer only used Carfax's VHR sales to dealers in 2008 through 2011, whereas he used Carfax's VHR sales to dealers, consumers, and banking, insurance, and other non-dealer customers ("BIG" customers) in 2012 and 2013. Dr. Singer does not dispute this characterization of his methods in his rebuttal report. See Singer Reply Rep. ¶¶ 32-33. Rather, he notes that the consumer and BIG data for 2008 through 2010 had not previously been produced to him when he generated his initial report. Id. ¶ 33. That may be so, but it does not change the undisputed fact that Dr. Singer's foreclosure estimate was based on an artificially small figure for Carfax's VHR sales, and therefore an artificially small estimate of the overall size of the market.
Moreover, if Carfax's data for the missing years is correct, then the overall foreclosure share would appear to drop significantly—from 30% to, at most, 24% according to Dr. Willig. See Willig Rep. Br. ¶ 35 & Table IV. Dr. Singer does not challenge Dr. Willig's methodology in re-calculating the foreclosure share with this new data. See Singer Reply Rep. ¶ 33. Instead, he argues that the data for consumer and BIG sales for 2008 to 2010 are implausibly high and inconsistent with Carfax's financial statements. Id. ¶¶ 34-35. But even if that were true, Dr. Singer provides no estimate for what a realistic figure for consumer and BIG sales during the missing year would actually be, and he does not attempt to recalculate his own foreclosure estimate to take into account Carfax VHR sales that he now acknowledges he did not include. See id. ¶ 36 ("I am not persuaded to alter my foreclosure estimates based on this eleventh-hour data"). Although the consumer and BIG data for 2008 to 2010 may have been late-arriving, the underlying methodological flaw of using artificially low numbers for Carfax's VHR sales should have been apparent from the outset. And in the absence of any revised estimate from Dr. Singer, no reasonable jury could rely on Dr. Singer's conclusion that Carfax's exclusive agreements foreclosed 30% of the VHR market. See Brooke Grp., 509 U.S. at 242.
The second reason why no reasonable jury could rely on Dr. Singer's 30% foreclosure share finding is that Dr. Singer based his estimate of the market for VHRs—again, the denominator in the equation outlined above—on a market that did not include all of the relevant VHRs. Specifically, he omitted any VHRs that are based primarily on just NMVTIS data ("NMVTIS reports"). See Singer Reply Rep. ¶¶ 18-19 (acknowledging exclusion of "NMVTIS reports (and NMVTIS resellers)"); see also Singer Rep. ¶¶ 49, 57. In other words, he declined to factor in VHRs that the parties expressly stipulated were part of the VHR market for purposes of this motion. The Court will hold the parties to the stipulations they made when they asked the Court to permit Carfax to file this motion for partial summary judgment. Accordingly, Dr. Singer's foreclosure share estimate was artificially high because his estimate of the market for VHRs was artificially low.
To the extent Plaintiffs would argue that their stipulation to Carfax's market share (90%) turns out to be inconsistent with the market share that would result from stipulating that NMVTIS reports count as VHRs, the Court notes that that is one of the dangers of making stipulations before all of the facts have been uncovered in discovery. In any event, the Court's holding does not hinge on Dr. Singer's failure to include the NMVTIS reports, given the other reasons his analysis was flawed.
It is undisputed that factoring in the NMVTIS reports would further lower Dr. Singer's estimate of the percentage of the VHR market that was foreclosed. As with Dr. Singer's failure to include consumer and BIG data for certain years though, there is a genuine dispute as to precisely how much of an effect including the NMVTIS reports would have had on Dr. Singer's estimate of the foreclosure share. Compare Willig Reply Rep. ¶ 40 & Table IV, with Singer Reply Rep. ¶¶ 24-29. But as with the omission of the consumer and BIG data, Dr. Singer declines to provide an estimate of what a corrected foreclosure share—i.e., one that includes NMVTIS reports in the market for VHRs—might look like. See Singer Reply Rep ¶ 29. Without a revised estimate or some other evidence that correcting for the omission of the NMVTIS reports would have only a minimal impact on Dr. Singer's initial analysis, no reasonable jury could rely on Dr. Singer's conclusion that Carfax's exclusive agreements were responsible for a 30% foreclosure share in the market for VHRs.
Finally, there is an additional flaw in Dr. Singer's foreclosure conclusion—namely, that Dr. Singer assumed that all of the "exclusive" CPO Agreements foreclosed competition. See Singer Rep. ¶ 55. As this Court has held, however, no reasonable jury could find that the majority of the CPO Agreements were of a sufficient duration to foreclose competition. Accordingly, no reasonable jury would include VHR sales resulting from these short-term CPO Agreements in its assessment of the share of the VHR market that the CPO Agreements foreclosed. While the Court cannot say at summary judgment precisely how much correcting for this error would affect Dr. Singer's analysis, a reasonable jury would have to conclude that this omission artificially inflated the 30% foreclosure estimate he provided.
Relatedly, this error minimizes any inference of foreclosure that a jury could reasonably draw from the part of Dr. Singer's report that discusses what he terms the "spillover" effects of the CPO Agreements. See Singer Rep. ¶¶ 67-70. In essence, Dr. Singer argues that the CPO Agreements created "spillover" because they foreclosed the sale of VHRs for non-CPO vehicles. The CPO Agreements allegedly had this effect because, by requiring dealers to pay for enough VHRs to cover the vehicles that were part of a CPO program, Carfax made the prospect of purchasing an Advantage plan—which provided unlimited VHRs—more attractive. See id. Once a dealer had opted into the Advantage plan, the dealer had less incentive to buy VHRs from Carfax's competitors, since the dealer could receive all its VHRs from Carfax. See id. These missed sales are the alleged spillover effects. The parties do not appear to dispute the underlying economic theory driving the possibility of spillover effects. See, e.g., Willig Rep. ¶ 70. Rather, they disagree over the extent of any actual spillover caused by the CPO Agreements. Having reviewed the reports of both experts, the Court concludes that the record contains a genuine dispute of fact as to whether the CPO Agreements resulted in any meaningful amount of spillover. But even if such spillover existed, Plaintiffs have pointed to no evidence that would allow a jury to understand how much of an effect it had on the market for VHRs. Instead, Dr. Singer limits his discussion of the issue to arguing that the spillover exists. See Singer Rep. ¶¶ 71-75. Additionally, because the source of the alleged spillover is the CPO Agreements, only the CPO Agreements that could have actually foreclosed competition—i.e., only the ones with ten manufacturers—are relevant in determining whether the spillover foreclosed competition. Because all of the CPO Agreements put together represent only about 3% of Carfax's VHR sales, Pls. 56.1 Resp. ¶ 15, no reasonable jury could find that any spillover caused by the subset of CPO Agreements that potentially foreclosed competition was significant.
For all the foregoing reasons, no reasonable jury would adopt Dr. Singer's estimate that the CPO Agreements and the Website Agreements collectively foreclosed 30% of the market for VHRs.
This conclusion is not altered by two kinds of data that Dr. Singer apparently omitted. First, Plaintiffs note that Dr. Singer's foreclosure share estimate does not include data for CPO Agreements from 2008 to 2011. Opp. Br. 12; see also Singer Rep. ¶ 67. Second, Plaintiffs explain that the estimate does not include an assessment of exclusive agreements that Carfax had with other classified websites, such as AOL, Vehix, and Yahoo Autos. Opp. Br. 12; see also Singer Rep. ¶ 42; Carfax 56.1 Reply ¶¶ 88-88.1 & n.445. Both corrections would result in a more Plaintiff-friendly estimate of the share of the VHR market that was foreclosed. Opp. Br. 12. But Plaintiffs point to no evidence that these corrections would make anything more than a marginal difference in Dr. Singer's foreclosure estimate. Even if they had, such evidence could not rehabilitate Dr. Singer's findings in light of the foregoing discussion.
3. Determining Whether the Foreclosure is Substantial
The final task in assessing Carfax's motion for partial summary judgment is to determine whether, on the basis of the evidence before the Court, a reasonable jury could find that Carfax's exclusive agreements foreclosed a substantial share of the market for VHRs. For several reasons, a reasonable jury could not reach that conclusion. First, in light of the shortcomings just discussed in Dr. Singer's analysis, there is insufficient evidence to find that Carfax's exclusive agreements implicated a large enough share of the market—i.e., at least 30 or 40 percent—to generate harm to competition. See Discover Fin. Servs., 598 F. Supp. 2d at 406. Accordingly, this is a case where any foreclosure actually caused by the exclusive deals "would seem to be harmless to competition." Areeda & Hovenkamp, supra, ¶ 1821c, at 190. Put differently, because Plaintiffs have not plausibly shown that Carfax locked up at least 30 or 40 percent of the VHR market, the only reasonable inference is that AutoCheck, or any other purveyor of VHRs, could have competed effectively. Cf. Tops Markets, Inc. v. Quality Markets, Inc., 142 F.3d 90, 96 (2d Cir. 1998) ("The Sherman Act protects competition as a whole in the relevant market.").
Second, a consideration of the "the competitive characteristics of the relevant market," Geneva Pharm., 386 F.3d at 508, confirms that no reasonable jury could find that any foreclosure caused by Carfax's exclusive deals was substantial enough to harm competition. Plaintiffs have pointed to no "evidence of high barriers to entry," which would suggest that "potential suppliers could not easily enter the market" and compete. Id. at 509; see also CDC Techs, 186 F.3d at 80 (highlighting plaintiff's failure to show existence of significant barriers to entry in granting summary judgment in exclusive dealing case). Nor have they pointed to evidence that the agreements at issue were particularly long-term. See Am. Express, 2005 WL 1515399, at *3 ("[T]he shorter an agreement's term and the easier it is to terminate, the more likely that it will be upheld." (internal quotation marks omitted)). In fact, even the Website Agreements—which appear to have been for longer terms than most of the CPO Agreements—lasted for no longer than three to five years. Relatedly, Plaintiffs do not dispute that AutoCheck negotiated with Autotrader and Cars.com during the period of alleged exclusivity, or that AutoCheck was able to obtain a partnership with another auto website, KBB.com, during the relevant period. Pls. 56.1 Resp. ¶¶ 56--57.1.
The only competitive characteristic of the VHR market that might raise concerns is the fact—stipulated to for this motion—that Carfax possessed a 90% market share and monopoly power. But even with that kind of market power, "there must be other grounds to believe that the defendant's behavior will harm competition market-wide." CDC Techs., Inc., 186 F.3d at 81 (internal quotation marks omitted). Plaintiffs have failed to identify such grounds. Compare Dentsply, 399 F.3d at 188-89, 193 (finding anti-competitive effect where defendant had "dominant share" of the market for more than ten years and used that share to "effectively choke[] off the market for artificial teeth, leaving only a small sliver for competitors"). Plaintiffs do cite evidence that Carfax intended for its agreements to be exclusive, and that it hoped to benefit from such exclusivity. See, e.g., Opp. Br. 35 (quoting document from Cars.com that notes: [Redacted] ). Such intent, however, "is not by itself sufficient to meet the adverse-effect requirement." K.M.B. Warehouse Distribs., Inc. v. Walker Mfg. Co., 61 F.3d 123, 130 (2d Cir. 1995). Rather, "[w]ithout some evidence of an adverse impact on competition," the fact that a supplier secures an exclusive deal "in order to limit competition does not satisfy a plaintiff's initial burden under § 1." Id. (emphasis added).
In light of Plaintiffs' failure to identify evidence that would allow a reasonable jury to conclude that Carfax's exclusive agreements foreclosed at least 30 or 40 percent of the VHR market, and absent any evidence as to how either the structure of the VHR market or the particulars of Carfax's exclusive agreements enabled Carfax to substantially foreclose competition, Plaintiffs cannot prevail on their Section 1 exclusive dealing claim. Accordingly, Carfax's motion for partial summary judgment with respect to Section 1 is granted.
In supplemental briefing requested by the Court after the Second Circuit decided MacDermid Printing Solutions LLC v. Cortron Corp., No. 15-589-CV, 2016 WL 4204795 (2d Cir. Aug. 10, 2016), Plaintiffs appear to raise the prospect that even if they cannot prove substantial foreclosure, they can continue to press their Section 1 claim if they present evidence of harm to consumers. Pls. Supp. Br. 2. This is incorrect as a matter of antitrust law and an apparent attempt to walk back the pledge that led the Court to permit a motion for partial summary judgment in the first place. With respect to the law, direct evidence of harm to competition is generally sufficient in Section 1 cases, but it is not sufficient in exclusive dealing cases. That is because "[e]xclusive dealing is an unreasonable restraint of trade and a § 1 violation only when the agreement freezes out a significant fraction of buyers or sellers from the market." Geneva Pharm., 386 F.3d at 508 (emphasis added). If the rule were otherwise, it would vitiate the presumption that exclusive deals, which are hallmarks of many competitive industries, are legal and pro-competitive. See CDC Techs., 186 F.3d at 80. Accordingly, a showing of substantial foreclosure is necessary. That was clearly the parties' understanding when they stipulated that a motion for partial summary judgment—on the specific issue of foreclosure—would resolve the case if the Court granted Carfax's motion. See Dkt. No. 151. To the extent Plaintiffs suggest an argument otherwise in their supplemental briefing, that argument is waived by virtue of their express representations to this Court. --------
IV. PLAINTIFFS' SECTION 2 CLAIMS
Section 2 of the Sherman Act makes it illegal to "monopolize, or attempt to monopolize, . . . any part of the trade or commerce among the several States." 15 U.S.C. § 2. To prevail on a claim for unlawful monopolization, a plaintiff must prove, first, "that the defendant possessed monopoly power in the relevant market," and second, "that [the defendant] willfully acquired or maintained that power." New York ex rel. Schneiderman v. Actavis PLC, 787 F.3d 638, 651 (2d Cir. 2015). With respect to this second element, "the possession of monopoly power will not be found unlawful unless it is accompanied by an element of anticompetitive conduct." Verizon Commc'ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 407 (2004) (emphasis in original). For an attempted monopolization claim, "the plaintiff must prove: '(1) that the defendant has engaged in predatory or anticompetitive conduct with (2) a specific intent to monopolize and (3) a dangerous probability of achieving monopoly power.'" Schneiderman, 787 F.3d at 651 (quoting Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447, 456 (1993)). Both monopolization and attempted monopolization claims therefore have "anticompetitive conduct" as one of their elements. See Microsoft, 253 F.3d at 58 ("A firm violates § 2 only when it acquires or maintains, or attempts to acquire or maintain, a monopoly by engaging in exclusionary conduct."). Put differently, in order "to be condemned as exclusionary, a monopolist's act must have an 'anticompetitive effect.' That is, it must harm the competitive process and thereby harm consumers. In contrast, harm to one or more competitors will not suffice." Id. (emphases in original).
Plaintiffs' Section 2 claim rests on the same theory of anticompetitive conduct as their Section 1 claim—that Carfax's exclusive deals foreclosed a substantial share of the market for VHRs. And as this Court has already held, no reasonable jury could find substantial foreclosure of the VHR market on these facts. Accordingly, the conclusion that Carfax's conduct did not harm competition for purposes of Section 1 applies with full force in the context of Plaintiffs' Section 2 claim. See Williams v. I.B. Fischer Nevada, 999 F.2d 445, 448 (9th Cir. 1993) ("As the no-switching agreement is not anticompetitive and thus does not establish a section 1 claim, it cannot form the basis of a section 2 claim."). To be sure, exclusive dealing "may give rise to a § 2 violation even though the contracts foreclose less" than the share of the market "usually required in order to establish a § 1 violation." Microsoft, 253 F.3d at 70. But the Court's conclusion as to the lack of anticompetitive harm in this case does not turn on a precise foreclosure threshold. Rather, it rests on an assessment of the competitive characteristics of the VHR market, including the Court's finding that Plaintiffs have not offered evidence that would allow a reasonable jury to conclude that Carfax used its market share to substantially foreclose competition. Plaintiffs have not identified any evidence from which a reasonable jury could find that the CPO Agreements and Website Agreements did harm the competitive process under Section 2, but not Section 1. The Court therefore grants Carfax's motion for partial summary judgment with respect to Section 2.
V. CONCLUSION
For the foregoing reasons, Carfax's motion for partial summary judgment is GRANTED. The parties shall have one week from the date of this Opinion & Order to meet and confer and submit a joint letter proposing any redactions to the Opinion & Order, along with a justification for why such redactions are appropriate under this circuit's law. See Lugosch v. Pyramid Co. of Onondaga, 435 F.3d 110 (2d Cir. 2006). Although the Court will publicly docket an order announcing its bottom-line conclusions, it will refrain from publicly docketing this Opinion & Order until it has ruled on the proposed redactions. The Court will do so expeditiously.
Upon resolution of any redactions, the Court will close the case in light of the parties' stipulation, see Dkt. No. 151, and representations to the Court, see Tr. of Nov. 23, 2015 Conf. This resolves Docket Nos. 154 and 164.
SO ORDERED. Dated: September 30, 2016
New York, New York
/s/_________
ALISON J. NATHAN
United States District Judge