Marcus v. BMW of N. Am., LLC, No. 11-1192 (3d Cir. 2012)

Annotate this Case
Justia Opinion Summary

In 2007, Marcus leased a 2007 BMW from a dealership in New Jersey. Marcus suffered four “flat” tires during his three-year lease. Each time, he drove his car to a BMW dealership in New York and had the tire replaced. BMW billed Marcus between $350 to $390 for parts, labor, fees, and taxes. In each instance, the run-flat tires (RFT) worked as intended. Marcus sued Bridgestone, asserting consumer fraud, breach of warranty, and breach of contract claims. He claims that Bridgestone RFTs are “defective” because they: are highly susceptible to flats, punctures and bubbles, and fail at a significantly higher rate than radial tires or other run-flat tires; cannot be repaired, only replaced, in the event of a small puncture; and are “exorbitantly priced.” He claimed RFT-equipped BMWs cannot be retrofitted to operate with conventional tires, and that they are difficult to replace. The district court certified the suit under FRCP 23(b)(3) as an opt-out class action on behalf of all purchasers and lessees of certain model-year BMWs equipped with Bridgestone RFTs sold or leased in New Jersey with tires that have gone flat and been replaced. The Third Circuit vacated. Marcus’s claims do not satisfy the numerosity and predominance requirements.

Download PDF
PRECEDENTIAL UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT _______________ Nos. 11-1192 & 11-1193 _______________ JEFFREY MARCUS, Individually and On Behalf of All Others Similarly Situated. v. BMW of NORTH AMERICA, LLC; BRIDGESTONE AMERICAS TIRE OPERATIONS, LLC, f/k/a BRIDGESTONE FIRESTONE NORTH AMERICAN TIRE, LLC; BRIDGESTONE CORPORATION BRIDGESTONE AMERICAS TIRE OPERATIONS, LLC, f/k/a BRIDGESTONE FIRESTONE NORTH AMERICAN TIRE, LLC; BRIDGESTONE CORPORATION, Appellants (No. 11-1192) BMW OF NORTH AMERICA, LLC, Appellant (No. 11-1193) _______________ On Appeal from the United States District Court For the District of New Jersey (D.C. Civil Action No. 2-08-cv-05859) District Judge: Honorable Katharine S. Hayden _______________ Argued September 20, 2011 _______________ Before: AMBRO, CHAGARES, and ALDISERT, Circuit Judges (Opinion filed August 07, 2012) Hugh R. Whiting, Esq. Dustin B. Rawlin, Esq. Jones Day North Point 901 Lakeside Avenue Cleveland, OH 44114 [ARGUED] Counsel for Appellant, Bridgestone Corporation Susan T. Dwyer, Esq. Herrick Feinstein LLP 2 Park Avenue New York, NY 10016 Ronald J. Levine, Esq. David R. King, Esq. 2 Herrick Feinstein LLP 210 Carnegie Center Princeton, NJ 08540 Counsel for Appellant, Bridgestone Americas Tire Operations, LLC Rosemary J. Bruno, Esq. Christopher J. Dalton, Esq. [ARGUED] Buchanan Ingersoll & Rooney 550 Broad Street, Suite 810 Newark, NJ 07102 Counsel for Appellant, BMW of North America, LLC Karin E. Fisch, Esq. [ARGUED] Orin Kurtz, Esq. Abbe, Spanier, Rodd & Abrams LLP 212 East 39th Street New York, NY 10016 Alan E. Sash, Esq. Steven J. Hyman, Esq. McLaughlin & Stern, LLP 260 Madison Avenue New York, NY 10016 Counsel for Appellee, Jeffrey Marcus 3 _______________ OPINION OF THE COURT _______________ AMBRO, Circuit Judge Table of Contents I. Factual and Procedural Background ............................... 6 II. Jurisdiction and Standard of Review ............................. 10 III. Preliminary Matters ....................................................... 10 A. The Class Definition and the Claims to be Given Class Treatment ................................................................... 12 B. Ascertainability ........................................................... 14 IV. Rule 23(a) ...................................................................... 18 A. Numerosity ................................................................. 18 B. Commonality .............................................................. 24 C. Typicality .................................................................... 25 V. Rule 23(b)(3): Predominance....................................... 29 A. The Common Law Claims ......................................... 30 1. Common Proof of Susceptibility to Road Hazard Damage ................................................................. 32 2. Common Proof of Proximate Causation .............. 36 B. The New Jersey Consumer Fraud Act Claims ........... 39 VI. Conclusion .................................................................... 55 This class action involves run-flat tires ( RFTs ). As their name suggests, they can run while flat. Even if an RFT suffers a total and abrupt loss of air pressure from a puncture or other road damage, the vehicle it is on remains 4 stable and can continue driving for 50 to 150 miles at a speed of up to 50 miles per hour. Jeffrey Marcus leased a BMW convertible equipped with four Bridgestone RFTs. Unfortunately, he experienced four flat tires during his three-year lease. In each case, the RFT worked as intended. Even though the tire lost air pressure, Marcus was able to drive his car to a BMW dealer to have the tire replaced. Unsatisfied nonetheless, Marcus sued Bridgestone Corporation, Bridgestone Americas Tire Operations, LLC ( BATO ) (together Bridgestone ), and BMW of North America, LLC ( BMW ), asserting consumer fraud, breach of warranty, and breach of contract claims. Among other things, he claims that Bridgestone RFTs are defective because they: (1) are highly susceptible to flats, punctures and bubbles, and . . . fail at a significantly higher rate than radial tires or other run-flat tires; (2) cannot be repaired, only replaced, in the event of a small puncture; and (3) are exorbitantly priced. J.A. 91, 100, 102. He also claims RFT-equipped BMWs cannot be retrofitted to operate with conventional, non-run-flat tires, and that many service stations do not sell Bridgestone RFTs, making them difficult to replace. J.A. 91, 92. He faults BMW and Bridgestone for failing to disclose these defects. The District Court certified Marcus s suit under Federal Rule of Civil Procedure 23(b)(3) as an opt-out class action brought on behalf of all purchasers and lessees of certain model-year BMWs equipped with Bridgestone RFTs sold or leased in New Jersey with tires that have gone flat and been replaced. For the reasons that follow, we part from the District Court. Among other problems, on the record before us, Marcus s claims do not satisfy the numerosity and predominance requirements. We thus vacate the District Court s certification order and remand for proceedings consistent with this opinion. 5 I. FACTUAL AND PROCEDURAL BACKGROUND In July 2007, Jeffrey Marcus (a New York resident) leased a 2007 BMW 328ci from an authorized BMW dealership in Ramsey, New Jersey. The convertible first caught his eye at another dealership in South Hampton, New York. He saw the car on the showroom floor and, interest piqued, picked up a brochure. Aside from visiting the dealership, picking up the brochure, and riding in a friend s 328ci as a passenger, Marcus claims he absolutely [did] not do any other research on BMW vehicles or RFTs before leasing his car. J.A. 875. As noted, Marcus suffered four flat 1 tires during his three-year lease. Each time he experienced a flat, he drove his car to a BMW dealership in New York and had the tire replaced. BMW then billed Marcus between $350 to $390 for parts, labor, fees, and taxes. See J.A. 407-11. After his first flat, Marcus purchased a road-hazard warranty for about $400, which covered at least some of the replacement costs for flat tires two through four. See J.A. 880. Marcus s first two flat tires were not available for inspection in this lawsuit. Dealer records show that a nail punctured the first tire and the second was replaced due to a blown out bubble. J.A. 407-08. Marcus s third tire was replaced because he ran over a chunk of metal the size of a finger, according to his own expert, and his fourth because he ran over another sharp object that tore and gouged the tire and damaged the sidewall. J.A. 300. The parties experts agree that the third and fourth tires could not have been 1 As noted in our introduction, RFTs do not go flat in the conventional sense. When discussing RFTs, we use the term flat to mean a tire that has suffered a loss of air pressure and been replaced. 6 repaired, and that any tire (run-flat or conventional) would have been damaged, if not destroyed, under the circumstances. See J.A. 300, 414-15. They also agree on two other, more general propositions: (1) a tire can go flat or fail for a wide variety of reasons and not be a defective tire; and (2) to determine properly and accurately the cause of any particular tire failure, a careful and thorough examination of that tire is necessary. See J.A. 305, 399, 1476-77. The parties dispute what Marcus knew about RFTs and RFT-equipped BMWs before leasing his car and, more importantly, what other purchasers and lessees could have known. To provide a market and consumer perspective on RFTs and BMWs, BMW presented the expert report and testimony of William Pettit. See J.A. 1761-76. He concluded that [a]n abundance of [RFT] information exists in the public domain extolling the safety and convenience benefits and discussing potential downsides. J.A. 1769. He pointed to BMW and Bridgestone documents (e.g., their press releases and marketing brochures), as well as information from the public domain (e.g., articles in publications like Consumer Reports, BusinessWeek, The Wall Street Journal, and The New York Times). To take but a few examples, BMW boasts in its brochures that should you get a flat, you can still travel up to 150 miles at 50 mph, thanks to [the 2007 BMW 3 Series Convertible s] standard run-flat tires, J.A. 329, so you can drive off a busy highway, out of a dangerous area, or just continue on your journey on a rainy night, J.A. 1108. In several other places, its brochures mention that the car comes equipped with RFTs and that, [d]ue to low-profile tires, wheels, tires and suspension parts are more susceptible to road hazard and consequential damages. J.A. 366, 374, 377, 380. A warning about road hazard susceptibility also appears in Bridgestone s tire warranty low aspect ratio tires, with 7 reduced sidewall height, may be more susceptible to damage from potholes, road hazards, and other objects such as curbs as well as in BMW press releases from July 2006 and March 2007. See J.A. 577, 579, 580, 993, 1449. In addition, the BMW Owner s Manual for the 328i warns drivers that, [f]or safety reasons, BMW recommends that damaged RunFlat Tires be replaced rather than repaired. J.A. 1558. BMW s Approved Tires brochure similarly advises consumers that [w]hile some tire manufacturers will allow tire repairs, BMW only recommends replacement of damaged tires. J.A. 1092. BMW and Bridgestone also highlight information from the public domain. An April 2006 BusinessWeek article says, Run-flat tires aren t cheap. Four Bridgestone Blizzak Run-Flats cost about $1,200, compared with perhaps $500 to $800 for comparable conventional tires. J.A. 762. Sounding a similar note, an article from the June 2007 Consumer Reports mentions that some RFT owners have complained about limited replacement choices and high replacement costs. J.A. 961. However, this report concludes that [d]espite the disadvantages and inconveniences of run-flat tires for many, Consumer Reports believes that the safety benefits can outweigh the downsides. Id. Marcus disputes BMW s and Bridgestone s openness about RFTs, the information available in the public domain, and whether this information is sufficient to provide consumers with notice about the downsides of Bridgestone RFTs. He offers internal BMW and Bridgestone documents such as emails, research reports and marketing surveys to show that the defendants were aware that many other consumers were as displeased with RFTs as he was. After his second flat tire, Marcus sued BMW and Bridgestone. In his Amended Class Action Complaint, he asserts claims against BMW and Bridgestone: (1) on behalf of 8 a nationwide class, for breach of implied warranty in violation of the Magnuson Moss Warranty Act ( MMWA ); and (2) on behalf of a New Jersey sub-class for (a) consumer fraud in violation of the New Jersey Consumer Fraud Act ( NJCFA ), (b) breach of implied warranty, (c) breach of contract, and (d) breach of the implied covenant of good faith and fair dealing. He also asserts a claim on behalf of the New Jersey sub-class for breach of express warranty against BMW. The District Court denied Marcus s motion for class certification with respect to the nationwide class, but granted the motion with respect to the New Jersey sub-class. See Marcus v. BMW of N. Am., LLC, No. 08-5859 (KSH), 2010 WL 4853308 (D.N.J. Nov. 19, 2010). The Court did not provide a definition of the class it certified, though it crossreferenced the docket entry for Marcus s amended notice of motion for class certification. That notice sought certification of a New Jersey state subclass consisting of [any] and [a]ll current and former owners and lessees of 2006, 2007, 2008, and 2009 BMW vehicles equipped with run-flat tires manufactured by Bridgestone and/or BATO and sold or leased in [New Jersey] whose Tires have gone flat and been replaced (excluded from the Class and the Sub-Class are Defendants, as well as Defendants affiliates, employees, officers and directors, including franchised dealers, and any person who has experienced physical injury as a result of the defects at issue in this litigation) . . .. J.A. 198. 9 II. JURISDICTION AND STANDARD OF REVIEW The District Court had jurisdiction under 28 U.S.C. § 1332(d). Having granted BMW s and Bridgestone s petitions for leave to appeal under Federal Rule of Civil Procedure 23(f), we have appellate jurisdiction under 28 U.S.C. § 1292(e). We review a class certification order for abuse of discretion, which occurs if the district court s decision rests upon a clearly erroneous finding of fact, an errant conclusion of law or an improper application of law to fact. In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 312 (3d Cir. 2009) (quotation marks omitted). [W]hether an incorrect legal standard has been used is an issue of law to be reviewed de novo. Id. III. PRELIMINARY MATTERS The class action is an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only. Wal-Mart Stores v. Dukes, 131 S. Ct. 2541, 2550 (2011) (quotation marks omitted). To invoke this exception, every putative class action must satisfy the four requirements of Rule 23(a) and the requirements of either Rule 23(b)(1), (2), or (3). See Fed. R. Civ. P. 23(a)-(b). To satisfy Rule 23(a), (1) the class must be so numerous that joinder of all members is impracticable (numerosity); (2) there must be questions of law or fact common to the class (commonality); (3) the claims or defenses of the representative parties must be typical of the claims or defenses of the class (typicality); and (4) the named plaintiffs must fairly and adequately protect the interests 10 of the class (adequacy of representation, or simply adequacy). In re Cmty. Bank of N. Va., 622 F.3d 275, 291 (3d Cir. 2010) (quoting Fed. R. Civ. P. 23). Rule 23(b)(3), the basis for certification here, requires that (i) common questions of law or fact predominate (predominance), and (ii) the class action is the superior method for adjudication (superiority). Id. [T]he requirements set out in Rule 23 are not mere pleading rules. Hydrogen Peroxide, 552 F.3d at 316; see also Dukes, 131 S. Ct. at 2551. The party seeking certification bears the burden of establishing each element of Rule 23 by a preponderance of the evidence. See Hydrogen Peroxide, 553 F.3d at 307. Echoing the Supreme Court, we have repeatedly emphasize[d] that [a]ctual, not presumed[,] conformance with Rule 23 requirements is essential. Id. at 326 (quoting Newton v. Merril Lynch, Pierce, Fenner & Smith, Inc., 259 F.3d 154, 167 (3d Cir. 2001) (quoting Gen. Tel. Co. of the Sw. v. Falcon, 457 U.S. 147, 160 (1982))). To determine whether there is actual conformance with Rule 23, a district court must conduct a rigorous analysis of the evidence and arguments put forth. Id. at 316 (quoting Falcon, 457 U.S. at 161). When doing so, the court cannot be bashful. It must resolve all factual or legal disputes relevant to class certification, even if they overlap with the merits including disputes touching on elements of the cause of action. Id. at 307. 2 Rule 23 gives no license to shy away 2 As a practical matter, the certification decision is typically a game-changer, often the whole ballgame, for plaintiffs and plaintiffs counsel. See Newton, 259 F.3d at 167 ( [D]enying or granting class certification is often the defining moment in class actions (for it may sound the death knell of the 11 from making factual findings that are necessary to determine whether the Rule s requirements have been met. Before considering Rule 23(a) and Rule 23(b)(3), we address two preliminary matters: (1) whether the District Court clearly defined the parameters of the class and the claims to be given class treatment, as required by Rule 23(c)(1)(B); and (2) whether the class must be (and, if so, is in fact) objectively ascertainable. A. The Class Definition and the Claims to be Given Class Treatment An order that certifies a class action must define the class and the class claims, issues, or defenses . . . . Fed. R. Civ. P. 23(c)(1)(B). Specifically, the text of the order or an incorporated opinion must include (1) a readily discernible, clear, and precise statement of the parameters defining the class or classes to be certified, and (2) a readily discernible, clear, and complete list of the claims, issues or defenses to be treated on a class basis. Wachtel v. Guardian Life Ins. Co., 453 F.3d 179, 187 (3d Cir. 2006). Clearly delineating the contours of the class along with the issues, claims, and defenses to be given class treatment serves several important purposes, such as providing the parties with clarity and assisting class members in understanding their rights and making informed opt-out decisions. Id. litigation on the part of plaintiffs, or create unwarranted pressure to settle nonmeritorious claims on the part of defendants) . . . . ). Appropriately, attentiveness to the potential for unwarranted settlement pressure coincides with a court s obligation to conduct a rigorous analysis of whether a plaintiff has demonstrated actual conformance with Rule 23. Hydrogen Peroxide, 552 F.3d at 310. 12 The definition of the class certified here is not clear and precise. Rather than set out its own definition, the District Court noted in its certification order that [c]ertification of the New Jersey sub-class is granted [D.E. 144], cross-referencing the docket entry for Marcus s amended notice of motion for class certification. That notice sought certification of a New Jersey state subclass consisting of [any] and [a]ll current and former owners and lessees of 2006, 2007, 2008, and 2009 BMW vehicles equipped with run-flat tires manufactured by Bridgestone and/or BATO and sold or leased in the United States whose Tires have gone flat and been replaced (excluded from the Class and the Sub-Class are Defendants, as well as Defendants affiliates, employees, officers and directors, including franchised dealers, and any person who has experienced physical injury as a result of the defects at issue in this litigation) . . .. J.A. 198. Although Marcus s notice of motion defines the New Jersey subclass in terms of vehicles sold or leased in the United States, the Court mentions in its accompanying opinion that the subclass is limited to those who bought or leased their cars in New Jersey. Marcus, 2010 WL 4853308 at *1. Even with this correction, however, the parameters of Marcus s class definition are far from clear. As written, the definition seems to include not only (1) owners and lessees who bought or leased a new or used BMW from a New Jersey BMW dealership, but also (2) subsequent owners and lessees who bought or leased a used BMW in New Jersey from anyone, not just BMW dealers, and even (3) subsequent 13 owners and lessees who bought or leased a used BMW anywhere in the country from anyone whose BMW was initially bought or leased in New Jersey. When confronted with this ambiguity at oral argument, Marcus s counsel clarified that the intent was to include only owners and lessees of new BMWs purchased or leased in New Jersey from BMW dealerships with original-equipment Bridgestone RFTs. See Oral Arg. Tr. 20:19-:24. Even if the District Court shared counsel s understanding of the class definition, counsel s post hoc clarification is no substitute for a readily discernible, clear, and precise statement of the parameters defining the class . . . to be certified in either the certification order or accompanying opinion. Wachtel, 453 F.3d at 187. In addition, the certification order does not define the claims, issues, or defenses to be treated on a class basis at all. The District Court s opinion does address Marcus s claims and the issues presented, but we cannot find a readily discernible, clear, and complete list. Id.; see also id. at 189 (noting that an opinion that forces us to comb the entirety of its text and cobble together the various statements in search of isolated statements that may add up to a partial list of class claims, issues, or defenses falls short of the readily discernible and complete list of class claims, issues, or defenses required by the Rule ). Rule 23(c)(1)(B) requires more, and thus a remand. If Marcus continues to attempt to certify a class, he needs to provide the Court with a more clearly defined class and set of claims, issues, or defenses to be given class treatment. B. Ascertainability Many courts and commentators have recognized that an essential prerequisite of a class action, at least with respect to actions under Rule 23(b)(3), is that the class must be 14 currently and readily ascertainable based on objective criteria. See, e.g., John v. Nat. Sec. Fire & Cas. Co., 501 F.3d 443, 445 (5th Cir. 2007); In re Initial Pub. Offerings Sec. Litig., 471 F.3d 24, 30 (2d Cir. 2006); Crosby v. Social Sec. Admin. of the U.S., 796 F.2d 576, 580 (1st Cir. 1986); see also Chiang v. Veneman, 385 F.3d 256, 271 (3d Cir. 2004) (holding that defining a class by reference to those who believe they were discriminated against undermines the validity of the class by introducing a subjective criterion into what should be an objective evaluation ), abrog. on other grounds by Hydrogen Peroxide, 552 F.3d at 318 n.18; Johnson v. Geico Cas. Co., 673 F.Supp.2d 255, 268 (D. Del. 2009); Agostino v. Quest Diagnostics, Inc., 256 F.R.D. 437, 478 (D.N.J. Feb. 11, 2009); 1 William B. Rubenstein et al., Newberg on Class Actions, § 3:2-3:3 (5th ed.); 7A Charles Alan Wright et al., Federal Practice and Procedure § 1760 (3d ed. 2005). If class members are impossible to identify without extensive and individualized fact-finding or minitrials, then a class action is inappropriate. Some courts have held that where nothing in company databases shows or could show whether individuals should be included in the proposed class, the class definition fails. See Clavell v. Midland Funding LLC, No. 10-3593, 2011 WL 2462046, at *4 (E.D. Pa. June 21, 2011); Sadler v. Midland Credit Mgmt, Inc., No.06-C-5045, 2008 WL 2692274, at *5 (N.D. Ill. July 3, 2008); In re Wal-Mart Stores, Inc. Wage & Hour Litig., No. C 06-2069 SBA, 2008 WL 413749, at *8 (N.D. Cal. Feb. 13, 2008); Deitz v. Comcast Corp., No. C 06-06352 WHA, 2007 WL 2015440, at *8 (N.D. Cal. July 11, 2007). The ascertainability requirement serves several important objectives. First, it eliminates serious administrative burdens that are incongruous with the efficiencies expected in a class action by insisting on the easy identification of class members. Sanneman v. Chrysler Corp., 191 F.R.D. 441, 446 (E.D. Pa. Mar. 2, 2000). Second, 15 it protects absent class members by facilitating the best notice practicable under Rule 23(c)(2) in a Rule 23(b)(3) action. See Manual for Complex Litigation, § 21.222 (4th ed. 2004). Third, it protects defendants by ensuring that those persons who will be bound by the final judgment are clearly identifiable. See Xavier v. Philip Morris USA, Inc., 787 F. Supp. 2d 1075, 1089 (N.D. Cal. 2011) ( Ascertainability is needed for properly enforcing the preclusive effect of final judgment. The class definition must be clear in its applicability so that it will be clear later on whose rights are merged into the judgment, that is, who gets the benefit of any relief and who gets the burden of any loss. If the definition is not clear in its applicability, then satellite litigation will be invited over who was in the class in the first place. ); 1 William B. Rubenstein et al., Newberg on Class Actions, § 3:1 (5th ed.). The proposed class raises serious ascertainability issues. In its briefs, BMW claims that it may be able to identify current and former original owners and lessees of BMW vehicles factory-equipped with Bridgestone RFTS which were initially purchased or leased from New Jersey dealerships. BMW Br. at 38 (emphasis in original); see also BMW Reply Br. 8. At oral argument, BMW s counsel was even less optimistic. He suggested that BMW could not know which of the vehicles that fit the class definition had Bridgestone RFTs because [t]hey re made in Germany by a different company, and BMW does not have a parts manifest. See Oral Arg. Tr. 16:3-16:8. To complicate matters further, not every car that comes onto a dealership lot with Bridgestone RFTs necessarily leaves the lot with the same tires. According to BMW, dealers might change the tires at a customer s request. In any event, says BMW, even if the proper cars with the proper tires could be identified, defendants records would not indicate whether all potential class members Bridgestone RFTs have gone flat and been 16 replaced, as the class definition requires, because the class is not limited to those persons who took their vehicles to BMW dealers to have their tires replaced. If Marcus attempts to certify a class on remand, the District Court adjusting the class definition as needed must resolve the critical issue of whether the defendants records can ascertain class members and, if not, whether there is a reliable, administratively feasible alternative. We caution, however, against approving a method that would amount to no more than ascertaining by potential class members say so. For example, simply having potential class members submit affidavits that their Bridgestone RFTs have gone flat and been replaced may not be proper or just. See Xavier, 787 F.Supp.2d at 1089-90 (rejecting plaintiffs ascertainability proposal to have potential class members submit affidavits about their smoking histories). BMW and Bridgestone will be able to cross-examine Marcus at trial about whether and why his tires have gone flat and been replaced. Forcing BMW and Bridgestone to accept as true absent persons declarations that they are members of the class, without further indicia of reliability, would have serious due process implications. 3 3 The class definition and ascertainability problems spill into the predominance inquiry as well. According to BMW and Bridgestone, the District Court could not have properly conducted a choice-of-law analysis because the class definition includes unidentified persons from jurisdictions unknown. See BMW Br. at 45; see also BMW Reply Br. at 19 ( [T]he nebulous class definition . . . makes the choice-oflaw analysis difficult if not impossible. ). If the District Court finds on remand an ascertainable class, and clarifies its definition accordingly, then the defendants can renew their 17 IV. RULE 23(A)4 A. Numerosity Rule 23(a)(1), which requires that a class be so numerous that joinder of all members is impracticable, promotes three core objectives. First, it ensures judicial economy. It does so by freeing federal courts from the onerous rule of compulsory joinder inherited from the English Courts of Chancery and the law of equity. See 1 Rubenstein, supra, § 1:12. Courts no longer have to conduct a single, administratively burdensome action with all interested parties compelled to join and be present. Id. The impracticability of joinder, or numerosity, requirement also promotes judicial economy by sparing courts the burden of having to decide numerous, sufficiently similar individual actions seriatim. Id. at § 3:11. As for its second objective, Rule 23(a)(1) creates greater access to judicial relief, particularly for those persons with claims that would be uneconomical to litigate individually. See Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 809 (1985). Finally, the rule prevents putative class representatives and their counsel, when joinder can be easily choice-of-law arguments as they relate to predominance. We note, however, that predominance is not defeated merely because different states laws apply to different class members claims. See Sullivan v. DB Invs., Inc., 667 F.3d 273, 301-02 (3d Cir. 2011) (en banc). In this regard, Rule 23(b)(3) requires merely that common issues predominate, not that all issues be common to the class. Id. (quoting Smilow v. Sw. Bell Mobile Sys., Inc., 323 F.3d 32, 39 (1st Cir. 2003) (emphasis added)). 4 BMW and Bridgestone do not challenge the District Court s finding that Marcus is an adequate representative of the class. 18 accomplished, from unnecessarily depriving members of a small class of their right to a day in court to adjudicate their own claims. 7A Charles Alan Wright et al., Federal Practice and Procedure § 1762 (3d ed. 2005). There is no minimum number of members needed for a suit to proceed as a class action. See Stewart v. Abraham, 275 F.3d 220, 226-27 (3d Cir. 2001). We have observed, however, that generally if the named plaintiff demonstrates that the potential number of plaintiffs exceeds 40, the first prong of Rule 23(a) has been met. Id. Nonetheless, Rule 23(a)(1) requires examination of the specific facts of each case. Gen. Tel. Co. of the N.W. v. EEOC, 446 U.S. 318, 330 (1980). Critically, numerosity like all Rule 23 requirements must be proven by a preponderance of the evidence. See Hydrogen Peroxide, 552 F.3d at 307. When a plaintiff attempts to certify both a nationwide class and a state-specific subclass, as Marcus did here, evidence that is sufficient to establish numerosity with respect to the nationwide class is not necessarily sufficient to establish numerosity with respect to the state-specific subclass. The Court of Appeals for the Eleventh Circuit s decision in Vega v. T-Mobile USA, Inc. is particularly instructive on this point. 564 F.3d 1256, 1266-68 (11th Cir. 2009). There, the Court reversed a finding of numerosity concerning a proposed class of all T-Mobile employees in Florida who received commissions for the sale of cellphone plans. It held that the plaintiff could not simply rely on the nationwide presence of T-Mobile to satisfy the numerosity requirement without Florida-specific evidence: Vega has not cited, and we cannot locate in the record, any evidence whatsoever (or even an allegation) of the number of retail sales associates T-Mobile employed during the class 19 period in Florida who would comprise the membership of the class, as certified by the district court. Yes, T-Mobile is a large company, with many retail outlets, and, as such, it might be tempting to assume that the number of retail sales associates the company employed in Florida during the relevant period can overcome the generally low hurdle presented by Rule 23(a)(1). However, a plaintiff still bears the burden of establishing every element of Rule 23, and a district court s factual findings must find support in the evidence before it. In this case, the district court s inference of numerosity for a Florida-only class without the aid of a shred of Florida-only evidence was an exercise in sheer speculation. Id. at 1267 (emphasis added) (citations and footnotes omitted). Like Vega, Marcus offered sufficient company-wide evidence to the District Court to support a finding of numerosity for a nationwide class. According to BMW, it sold or leased 740,102 vehicles equipped with RFTs nationwide during the class period. But not all those vehicles came with Bridgestone RFTs. The group includes cars with RFTs from seven different manufacturers Goodyear, Pirelli, Michelin, Dunlop, Continental, Uniroyal, and Bridgestone. With respect to these 740,102 vehicles, BMW recorded and produced 582 unique customer contacts from consumers across the nation who called BMW to complain about their RFTs. Of those 582 customers, 196 specifically identified their RFTs as Bridgestone RFTs. Marcus submitted 29 of the 582 complaints as evidence of the 20 proposed nationwide class s numerosity. In addition, he submitted records, obtained from various companies offering road hazard products, showing thousands of claims made for road hazard damage on 2006-2009 BMW vehicles. He even provided loss-ratio data (the ratio of dollars received in premiums to dollars paid out in claims) for one warranty company s contracts covering BMWs. Piled on top of that were internal BMW emails about customer complaints regarding Bridgestone RFTs. But after digging through this supposed mountain of evidence, we can only speculate as to how many 2006-2009 BMWs were purchased or leased in New Jersey with Bridgestone RFTs that have gone flat and been replaced. To begin, we can only guess as to how many 2006-2009 BMWs were purchased or leased in New Jersey regardless of tire brand. That information is not in the record. There is also no evidence of how many of the 740,102 vehicles bought and leased nationwide had Bridgestone RFTs. No evidence shows that BMW purchased tires from its seven RFTsuppliers in roughly equal proportions or even if Bridgestone was among its larger or smaller suppliers. As noted, Marcus submitted 29 of the 582 nationwide complaints as evidence of numerosity. The complaints, however, do not indicate in any readily apparent way whether the complaining customers purchased or leased their vehicles in New Jersey. Four complaining customers did list New Jersey addresses. See J.A. 1612, 1632, 1634, 1636. But only two of these four customers referred to Bridgestone RFTs specifically, and one of them apparently accepted $500 in Lifestyle Gifts as a settlement from BMW. See J.A. 1634, 1636-43. And, like Marcus, they could have crossed state lines to purchase or lease their cars. Not to pile on, but Marcus has not pointed us to any evidence in the record not in the customer complaints, the road hazard warranty claims, the loss-ratio data, or the internal BMW emails that identifies another 21 purchaser or lessee of a 2006-2009 BMW that was sold or leased in New Jersey and equipped with Bridgestone RFTs that have gone flat and been replaced. In short, he has offered proof of only one potential class member: himself. Nonetheless, the District Court found that the New Jersey class met the numerosity requirement because it is common sense that there will be more members of the class than the number of consumers who complained probably significantly more, and common sense indicates that there will be at least 40. Marcus, 2010 WL 4853308 at *3. That may be a bet worth making, but it cannot support a finding of numerosity sufficient for Rule 23(a)(1). As we explained in Hydrogen Peroxide, a district court must make a factual determination, based on the preponderance of the evidence, that Rule 23 s requirements have been met. See 552 F.3d at 307. Mere speculation is insufficient. See 7A Wright, supra, § 1762 (collecting cases); Roe v. Town of Highland, 909 F.2d 1097, 1100 n.4 (7th Cir. 1990) ( The party supporting the class cannot rely on conclusory allegations that joinder is impractical or on speculation as to the size of the class in order to prove numerosity. ) (quotation marks omitted). Of course, Rule 23(a)(1) does not require a plaintiff to offer direct evidence of the exact number and identities of the class members. But in the absence of direct evidence, a plaintiff must show sufficient circumstantial evidence specific to the products, problems, parties, and geographic areas actually covered by the class definition to allow a district court to make a factual finding. Only then may the court rely on common sense to forgo precise calculations and exact numbers. See In re Prudential Ins. Co. of Am. Sales Practices Litig., 962 F. Supp. 450, 468, 510 (D.N.J. 1997), aff d 148 F.3d 283 (3d Cir. 1998) (finding numerosity requirement satisfied without pinpointing an exact number because the evidence suggested 22 a class of over 8,000,000 policyholders and [c]ommon sense suggest[ed] that it would be at best extremely inconvenient to join all class members ); cf. Lloyd v. City of Phila., 121 F.R.D. 246, 249 (E.D. Pa. Aug. 22, 1988) (refusing to certify a class when, despite plaintiff s mere speculation that the class could exceed 10,000 employees, the evidence demonstrated that only the four plaintiffs met the class definition). Given the complete lack of evidence specific to BMWs purchased or leased in New Jersey with Bridgestone RFTs that have gone flat and been replaced, the District Court s numerosity ruling crossed the line separating inference and speculation. BMW is a large company that sells and leases many cars throughout the country, many with RFTs. But the Court did not certify a nationwide class of BMW owners and lessees with any brand of RFTs. It certified a New Jersey class of owners and lessees with Bridgestone RFTs that have gone flat and been replaced. It is tempting to assume that the New Jersey class meets the numerosity requirement based on the defendant companies nationwide presence. But the only fact with respect to numerosity proven by a preponderance of the evidence is that Marcus himself is a member of the proposed class. [I]f there are no members of the class other than the named representatives, then Rule 23(a)(1) obviously has not been satisfied. 7A Wright, supra, § 1762. And we are not prepared to read the numerosity requirement out of the class action rule, as we have no authority to do so. Gurmankin v. Costanzo, 626 F.2d 1132, 1135 (3d Cir. 1980). Accordingly, we hold that the District Court abused its discretion by finding the numerosity requirement to be satisfied with respect to the New Jersey class. 23 B. Commonality Rule 23(a)(2) s commonality requirement does not require identical claims or facts among class member[s]. Chiang v. Veneman, 385 F.3d 256, 265 (3d Cir. 2004), abrog. on other grounds by Hydrogen Peroxide, 552 F.3d at 318 n.18. For purposes of Rule 23(a)(2), even a single common question will do. Dukes, 131 S.Ct. at 2556 (quotation marks and alterations omitted); see also In re Schering Plough Corp. ERISA Litig., 589 F.3d 585, 597 (3d Cir. 2009). Marcus seeks to offer evidence about, among other things, whether Bridgestone RFTs are defective, whether the defendants had a duty to disclose those defects, and whether the defendants did in fact fail to disclose those defects.5 These issues of fact and law (or some subset of them) apply to each of Marcus s causes of actions against each of the defendants, and are issues common to all class members. Accordingly, the District Court did not abuse its discretion in finding the commonality requirement to be satisfied. 5 Whether Marcus has alleged actionable defects or simply trade-offs that a consumer should consider in the purchase or lease calculus was the not the question before the District Court at the certification stage nor is it the one before us now. So while BMW argues that it had no duty to provide purchasers and lessees with a part-by-part price list for each of its vehicles and warn them about Bridgestone RFTs being exorbitantly priced, we cannot consider that argument at this stage of the litigation. 24 C. Typicality The concepts of typicality and commonality are closely related and often tend to merge. See Baby Neal v. Casey, 43 F.3d 48, 56 (3d Cir. 1994). Both serve as guideposts for determining whether under the particular circumstances maintenance of a class action is economical and whether the named plaintiff s claim and the class claims are so interrelated that the interests of the class members will be fairly and adequately protected in their absence. Falcon, 457 U.S. at 158 n.13. Typicality, however, derives its independent legal significance from its ability to screen out class actions in which the legal or factual position of the representatives is markedly different from that of other members of the class even though common issues of law or fact are present. 7A Wright, supra, § 1764. To determine whether a plaintiff is markedly different from the class as a whole, we consider the attributes of the plaintiff, the class as a whole, and the similarity between the plaintiff and the class. See Schering Plough, 589 F.3d at 597. This comparative analysis addresses three distinct, though related, concerns: (1) the claims of the class representative must be generally the same as those of the class in terms of both (a) the legal theory advanced and (b) the factual circumstances underlying that theory; (2) the class representative must not be subject to a defense that is both inapplicable to many members of the class and likely to become a major focus of the litigation; and (3) the interests and incentives of the representative must be sufficiently aligned with those of the class. 25 Id. at 599. If a plaintiff s claim arises from the same event, practice or course of conduct that gives rises to the claims of the class members, factual differences will not render that claim atypical if it is based on the same legal theory as the claims of the class. Hoxworth v. Blinder, Robinson & Co., 980 F.2d 912, 923 (3d Cir. 1992). The District Court found that Marcus is typical of the class, because [w]hat is crucial to the typicality analysis here is that Marcus and the class members purportedly suffered harm from the same alleged course of conduct: the defendants failed to disclose defects in their products, misrepresented or omitted important information about the products, and made promises about the products that were not true. Marcus, 2010 WL 4853308 at *6. BMW and Bridgestone argue that the Court erred in its finding for three reasons. First, Marcus performed very little research before leasing his car, whereas the average BMW purchaser or lessee performs significant research prior to purchase or lease. Second, Marcus leased only one model BMW with one kind of Bridgestone RFT, yet he seeks to represent a class composed of people with other model-year BMWs and Bridgestone tire types. According to Bridgestone, Marcus s claims potentially cover 49 different tire designs of varying specifications, including size, load rating, and model type. Third, New York law not New Jersey law applies to Marcus s claims and, as a result, Marcus is subject to unique defenses that do not apply to all members of the proposed class. We see no abuse of discretion behind any of these arguments. First, even if there are marked differences in the amounts of research class members performed, these differences by themselves do not render Marcus s claims atypical. As discussed below, if a class member knew about the alleged defects prior to purchase or lease, then that 26 knowledge could break the proximate cause link between the alleged defect and any damages suffered. Determining whether a class member had such knowledge requires an individualized inquiry. That creates a predominance problem. If Marcus knew of the defects prior to his lease, then he would risk having the causal link between the defendants conduct and his damages broken. If that were so, he would be unable to represent fairly the interests of class members who did not have such knowledge. That would create a typicality problem. But Marcus may be in a better position than other potential class members because he did not do significant research before leasing his car. That fact may distinguish him from other class members, but it does not prejudice his ability to protect absent class members interests fairly and adequately. The fact that Marcus leased only one model BMW with one kind of Bridgestone RFT also does not pose a typicality problem. When a class includes purchasers of a variety of different products, a named plaintiff that purchases only one type of product satisfies the typicality requirement if the alleged misrepresentations or omissions apply uniformly across the different product types. See, e.g., Wiener v. Dannon Co., 255 F.R.D. 658, 666-67 (C.D. Cal. Jan. 30, 2009); Gonzalez v. Proctor & Gamble Co., 247 F.R.D. 616, 621-22 (S.D. Cal. Sept. 12, 2007); Lewis Tree Serv. Inc. v. Lucent Techs. Inc., 211 F.R.D. 228, 232-34 (S.D.N.Y. Nov. 20, 2002); Kaczmarek v. Int Bus. Machs. Corp., 186 F.R.D. 307, 313 (S.D.N.Y. May 12, 1999). The parties agree that different tire models and sizes, and different vehicles, have different performance characteristics, compositions, designs, purposes, and uses. But Marcus alleges that the problems with Bridgestone RFTs that they are highly susceptible to road hazard damage, unrepairable, expensive, difficult to purchase, and that a vehicle cannot be converted for use of 27 conventional tires are uniform and apply to all 2006-2009 BMW vehicles equipped with Bridgestone RFTs. As we discuss below with respect to predominance, the District Court did not abuse its discretion when finding that a defect making Bridgestone RFTs highly susceptible to road hazard damage will show itself in a substantially similar way across the various tire models and sizes at issue here. Nor is there any indication that BMW s and Bridgestone s representations differed significantly depending on the model-year BMW or specification of Bridgestone RFT. Therefore, the fact that Marcus leased only one model BMW with one kind of Bridgestone RFT does not pose a typicality problem. Finally, [i]t is well-established that a proposed class representative is not typical under Rule 23(a)(3) if the representative is subject to a unique defense that is likely to become a major focus of the litigation. Schering Plough, 589 F.3d at 598 (quoting Beck v. Maximus, Inc., 457 F.3d 291, 301 (3d Cir. 2006)). Other courts of appeals emphasize, as do we, the challenge presented by a defense unique to a class representative the representative s interests might not be aligned with those of the class, and the representative might devote time and effort to the defense at the expense of issues that are common and controlling for the class. Beck, 457 F.3d at 297. Here, BMW and Bridgestone argue that New York law, not New Jersey law, applies to Marcus s claims and, along with it, certain unique defenses. For example, they claim that New York law, unlike New Jersey law, requires privity of contract in order to pursue a claim for breach of implied warranty. Compare Arthur Jaffe Assocs. v. Bilsco Auto Serv. Inc., 89 A.D.2d 785, 785 (App. Div. 1982), aff d 448 N.E.2d 792, 792 (N.Y. 1983), with Paramount Aviation Corp. v. Augusta, 288 F.3d 67, 73-76 (3d Cir. 2002) (applying New Jersey law). 28 The defendants presented this argument to the District Court, but it did not respond to it. The Court did conduct a choice-of-law analysis with respect to the class members MMWA claims when considering issues of predominance. It did not, however, conduct a choice-of-law analysis with respect to Marcus s consumer fraud or common law claims when considering issues of typicality, specifically whether Marcus s claims are subject to unique defenses that are likely to become a major focus of the litigation. But even if New York, rather than New Jersey, law applies to Marcus s claims, BMW and Bridgestone have failed to demonstrate how any defenses unique to Marcus s claims will become a major focus of the litigation. On remand, the parties and the District Court may find that the contours of the litigation have changed significantly and what will or will not likely be a major focus of the litigation will have to be assessed. We leave this to the District Court to address when and if the issue should arise. V. RULE 23(B)(3): PREDOMINANCE 6 Under Rule 23(b)(3), questions of law or fact common to class members [must] predominate over any questions affecting only individual members. This predominance requirement tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation. Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623 (1997). To assess predominance, a court at the certification stage must examine each element of a legal claim through the prism of Rule 23(b)(3). In re DVI, Inc. Sec. 6 Neither BMW nor Bridgestone challenge on appeal the District Court s findings that a class action is the superior method for adjudication. 29 Litig., 639 F.3d 623, 630 (3d Cir. 2011). A plaintiff must demonstrate that the element of [the legal claim] is capable of proof at trial through evidence that is common to the class rather than individual to its members. Hydrogen Peroxide, 552 F.3d at 311. Because the nature of the evidence that will suffice to resolve a question determines whether the question is common or individual, a district court must formulate some prediction as to how specific issues will play out in order to determine whether common or individual issues predominate in a given case. Id. (quotation marks omitted). Marcus asserts four claims on behalf of the New Jersey class against BMW and Bridgestone: (1) violations of the NJCFA; (2) breach of the implied warranty of merchantability; (3) breach of contract; and (4) breach of the implied covenant of good faith and fair dealing. He also asserts a claim for breach of express warranty against BMW. We consider the elements of these claims through the prism of the predominance requirement to determine whether they are capable of proof with common, class-wide evidence. A. The Common Law Claims 7 The essence of each of Marcus s common law claims (at least for purposes of our predominance analysis) is that he 7 Like the District Court, we refer to Marcus s breach of warranty, breach of contract, and breach of the implied covenant of good faith and fair dealing claims as his common law claims, even though the breach of implied warranty claims now derive from New Jersey statutes. See N.J. Stat. Ann. § 12A:2-314. 30 purchased a defective product that caused him damage. 8 In his complaint, Marcus alleges that Bridgestone RFTs (and, in 8 To state a claim for breach of the implied warranty of merchantability . . . , a plaintiff must allege (1) that a merchant sold goods, (2) which were not merchantable at the time of sale, (3) injury and damages to the plaintiff or its property, (4) which were was caused proximately and in fact by the defective nature of the goods, and (5) notice to the seller of injury. In re Ford Motor Co. E-350 Van Prods. Liab. Litig. (No. II), No. 03-4558 (HAA), 2008 WL 4126264 at *19 (D.N.J. Sept. 2, 2008). A claim for breach of express warranty similarly requires proof of proximate cause. See Collins v. Uniroyal, Inc., 315 A.2d 16, 20 n.4 (N.J. 1974) ( [P]roximate cause and damages. . . . are, of course, necessary for a breach of express warranty verdict. ); Prashker v. Beech Aircraft Corp., 258 F.2d 602, 607 (3d Cir. 1958) ( The measure of damages for a breach of warranty is, under New Jersey law, the loss directly and naturally resulting in the ordinary course of events from such breach ) (emphases added). To state a claim for breach of contract, a plaintiff must [1] show that the parties entered into a valid contract, [2] that the defendant failed to perform his obligations under the contract and [3] that the plaintiff sustained damages as a result. Murphy v. Implicito, 920 A.2d 678, 689 (N.J. Super. Ct. App. Div. 2007) (quotation marks omitted). Marcus s implied covenant of good faith and fair dealing claims also hinge on his purchasing a defective product that caused him damage. See Amended Class Action Complaint (J.A. 109-110) ( BMW breached those implied covenants by selling or leasing to Plaintiff and members of the Sub-Class Vehicles that were not fit for ordinary use, and that were equipped with Tires that were inherently defective 31 turn, the BMW vehicles that they equip) are defective for several reasons: (1) Bridgestone RFTs pop or sustain bubbles from use under normal driving conditions, making them more susceptible to road hazard damage than conventional tires and other brands of RFTs; (2) they cannot be repaired in the event of even a small puncture; and (3) they are extremely expensive to replace. In addition, he alleges that his BMW vehicle is defective because it cannot be reconfigured to operate with conventional tires. The District Court found that Marcus could prove these alleged defects at trial with common, class-wide evidence. BMW and Bridgestone contest the Court s finding with respect to the first alleged defect. 9 They also argue that issues of proximate causation i.e., determining why each class member s tires have gone and been replaced will require individualized inquiries that will predominate over any common ones. 1. Common Proof of Susceptibility to Road Hazard Damage According to BMW and Bridgestone, Marcus has failed to identify any particular defect that supposedly makes Bridgestone RFTs more susceptible to road hazard damage when BMW knew, or should have known, that the Tires were defective. . . . As a direct and proximate result of BMW s breach of its implied covenants, Plaintiffs and Class members have been damaged. ); see id. at 111-12 (making similar allegations against Bridgestone). 9 Because BMW and Bridgestone do not challenge the District Court s findings that, with respect to Marcus s common claims, the other alleged defects are capable of common proof, we need not consider them. 32 than other tires. In addition, they claim that any defect should one exist at all will not be evident uniformly across all tires, regardless of size or other specifications, included in the class definition. They argue that the District Court erred by accepting without question Marcus s expert testimony on these points without considering their own. In In re Hydrogen Peroxide Antitrust Litigation, we clarified a district court s duty when confronted with competing expert testimony about a plaintiff s ability to prove a claim through evidence that is common to the class. 552 F.3d at 307, 322-24. We held that the court s obligation to consider all relevant evidence and arguments [on a motion for class certification] extends to expert testimony, whether offered by a party seeking class certification or by a party opposing it. Id. at 307. We explained that [e]xpert opinion with respect to class certification, like any matter relevant to a Rule 23 requirement, calls for rigorous analysis. Id. at 323. Therefore, [w]eighing conflicting expert testimony at the certification stage is not only permissible[, but] it may be integral to the rigorous analysis Rule 23 demands, especially when a party opposing certification offers its own competing expert opinion. Id. We further assured district courts that [r]igorous analysis need not be hampered by a concern for avoiding credibility issues. Id. at 324. In that case, we ruled that the district court abused its discretion because it appeared to have assumed that it was barred from weighing one expert opinion against another for the purpose of determining whether the requirements of Rule 23 had been met, specifically whether the plaintiff s claims were susceptible to common proof. Id. at 322. Like the District Court in Hydrogen Peroxide, the District Court here was confronted with conflicting expert testimony about whether the plaintiff could prove its claim with common proof. On the one hand, Marcus s tire expert, 33 Charles Gold, opined on the similarity of Bridgestone RFTs. After a detailed analysis of the thousands of pages of specifications produced, he found that all Bridgestone runflat tires relevant to this action, despite variations due to size, are substantially similar in construction. J.A. 1978-79. He concluded that a proven defect arising from construction would manifest itself in all relevant tires. J.A. 1979. In addition, Gold suggests in his expert report that not only would all Bridgestone RFTs have a similar defect, but in fact they all do have a particular defect. He explains that the major difference between RFTs and conventional tires is the inclusion of extra components added to the sidewall and assemblies of RFTs, allowing RFTs to be operated at zero, or near zero, inflation pressure. J.A. 1978. Unfortunately, he adds, these same components stiffen the tire during regular inflated use and [t]he extra stiffness [in RFTs] can make the tire more susceptible to road hazard damages during normal use. J.A. 1978-1979. BMW and Bridgestone counter by highlighting that Gold recanted his extra stiffness opinion during his deposition. He admitted he had no published testing, studies or scientific data to support his opinion. J.A. 2031-32. In fact, Gold admitted that he cannot offer an opinion, to a reasonable degree of engineering certainty, that Bridgestone RFTs are more susceptible to road-hazard damage during normal use. J.A. 2074-75. When Bridgestone and BMW moved to exclude Gold s opinion, Marcus too seems to have changed course with respect to the stiffness theory. Rather than rely on that theory, he argued (and still argues now) that Bridgestone RFTs are more susceptible to road hazard damage because they are low aspect ratio tires and that proof of this defect is found in the defendants own 34 documents, not in any expert report. Marcus Br. 27-28; J.A. 2138. 10 Despite this apparent retreat from Gold s stiffness theory, the District Court found that Marcus has offered evidence that because run-flat tires are, universally, substantially stiffer than conventional tires, they are therefore more susceptible to road hazard damage. Such evidence makes it likely that common issues of proof will establish the class members claims. Marcus, 2010 WL 4853308, at *13. Ultimately, however, whether Bridgestone RFTs are more susceptible to road hazard damage than other tires due to their extra stiffness, their low aspect ratio, or anything else is not the issue before us. Our inquiry is limited to whether the District Court abused its discretion when finding that, should a defect exist at all in Bridgestone RFTs that makes them more susceptible to road hazard damage, Marcus will be capable of proving that defect at trial through evidence that is common to the class rather than individual to its members. See Hydrogen Peroxide, 552 F.3d at 311. On this point, the Court discussed and apparently credited Gold s similarity opinion. Marcus, 2010 WL 4853308, at *4, *5 (noting that Marcus will also offer Gold s 10 An aspect ratio is, roughly speaking, the relationship between the height and width of a tire. A tire with a low aspect ratio, sometimes referred to as a low profile tire, is short and squatty and generally provides for better handling. J.A. 421. Bridgestone argues that a low aspect ratio is not a defect. It is simply a design trade-off that allows for better handling at the cost of a softer ride, among other things. Bridgestone Reply Br. 8-10. To repeat, see supra note 5, whether Marcus has alleged actionable defects or nonactionable design trade-offs is not before us. 35 expert testimony that all Bridgestone run-flat tires, regardless of model, are substantially similar and that Gold s expert testimony opines that all of Bridgestone s run-flat tires are substantially similar, irrespective of model ). Bridgestone offered its own expert evidence (reports and deposition testimony from its experts, Brian Queiser and James Gardner) that the different tires and sizes in the class are not substantially similar given the differences in design, components and materials in different tires specified as standard and optional equipment for different BMW vehicles. J.A. 399; 401-06; 413-15; 424-26; 452-53. The District Court did not explicitly discuss these expert opinions, which challenge the similarity opinion of Marcus s expert. Although we would prefer a more explicit discussion and comparison of Bridgestone s competing expert testimony from the District Court to aid our appellate review, we cannot conclude that the Court abused its discretion in violation of Hydrogen Peroxide. Unlike the District Court s opinion in Hydrogen Peroxide, nothing in the District Court s opinion in our case suggests that it assumed it was barred from weighing the credibility of the expert opinions. Instead, it appears that the Court consistent with Hydrogen Peroxide simply found Gold s opinion about the similarity of Bridgestone RFTs to be more persuasive than the opinions put forth by Bridgestone. This was not an abuse of discretion. 2. Common Proof of Proximate Causation Having found that Marcus could show a common, class-wide defect, the District Court then found he could show, without resort to individual proofs, that this defect caused the class members damages. Considering the damages that Marcus alleges, we believe the District Court s causation finding was an abuse of discretion. 36 Recall that Marcus defines the class in terms of certain owners and lessees of BMW vehicles with Bridgestone RFTs that have gone flat and been replaced. He claims that [a]ll class members were damaged when their tires suffered a flat and they were forced to pay for a new Tire or when they purchased road hazard coverage to insulate them from financial hardship due to cost of the Tires. Pl s Am. Br. in Support of Class Cert. (J.A. 1255). Accordingly, he asserts that [e]ach Class member s damages can be measured by the cost of a replacement Tire. For Class members who purchased road hazard coverage, the damages will be the greater of either the cost of replacement Tires or the cost of road hazard coverage. Id. at 1253-54. But these damages allegations beg the question of what caused class members tires to go flat and need replacement. Causation is pivotal to each of Marcus s claims. See, e.g., N.J. Stat. Ann. § 12A:2-314 cmt. 13 (discussing how, in an action based on breach of warranty, it is of course necessary to show . . . that the breach of warranty was the proximate cause of the loss sustained, and that an affirmative showing by the seller that the loss resulted from some action or event following his own delivery of the goods can operate as a defense ). Here the District Court should have addressed an undisputed, fundamental point: any tire can go flat for myriad reasons. See J.A. 307-308, 448. Even defective tires can go flat for reasons completely unrelated to their defects. Critically, to determine why a particular class member s Bridgestone RFT has gone flat and been replaced requires an individual examination of that class member s tire. See J.A. 305, 399, 1476-77. These individual inquiries are incompatible with Rule 23(b)(3) s predominance requirement. In another RFT case brought against BMW and Goodyear involving nearly identical allegations as those 37 Marcus makes here, Judge Holwell of the United States District Court for Southern District of New York denied class certification and aptly explained why individual issues of causation create irremediable predominance problems: Even if the plaintiffs were to show that the Goodyear RFTs suffered from a common defect, they would still need to demonstrate that this defect caused each class member s RFT to puncture. But tires can puncture for any number of reasons, and not all of these reasons will relate to the defect. As defendants properly note, RFTs can go flat for reasons that would also cause a standard radial tire to go flat, for example, if the driver ran over a nail, tire shredding device, or large pothole, or if a vandal slashed the tire. . . . [P]laintiff would have to demonstrate in each individual case that the tire punctured for reasons related to the defect, rather than for a reason that would cause any tire to fail. Oscar v. BMW of N. Am., LLC, 274 F.R.D. 498, 511 (S.D.N.Y. June 7, 2011) ( Oscar I ). 11 Other federal courts have also recognized that suits alleging defects involving motor vehicles often involve complicated issues of individual causation that predominate over common questions regarding the existence of a defect. Id. at 510 (collecting cases); see 11 After Judge Holwell denied Oscar s motion for class certification, Judge Engelmayer denied Oscar s second motion for class certification based on causation and predominance problems. See Oscar v. BMW of N. Am., LLC, No. 09 Civ.11 (PAE), 2012 WL 2359964 (S.D.N.Y. June 19, 2012) ( Oscar II ). 38 also Chin v. Chrysler Corp., 182 F.R.D. 448, 455 (D.N.J. Sept. 11, 1998) (refusing to certify a class of purchasers and lessee of vehicles with allegedly defective anti-lock brake systems because, among other things, [e]ven where the alleged defect has manifested itself, individual issues of actual cause must be adjudicated ). Marcus s own experience illustrates the problem. Of the two tires he presented for inspection in this lawsuit, one went flat and was replaced because he ran over a jagged chunk of metal and the other because he ran over a sharp object that tore and gouged the tire and damaged the sidewall. See J.A. 300, 400, 409-10. The experts agree that the two tires could not have been repaired and that any tire (run-flat or conventional) would also have been damaged under the circumstances. See J.A. 309-10, 400, 412, 414, 426. In other words, it is undisputed that even if Marcus could prove that Bridgestone RFTs suffer from common, class-wide defects, those defects did not cause the damage he suffered for these two tires: the need to replace them. In this sense, Marcus is no different than a class member who, seconds after buying his car, pulls off the dealership lot and runs over a bed of nails, as neither can claim a defect caused his tires to go flat and need replacement. Because Marcus s common law claims require an individualized inquiry into why any particular consumer s Bridgestone RFTs went flat and had to be replaced, the District Court abused its discretion in finding that the claims satisfy the predominance requirement. B. The New Jersey Consumer Fraud Act Claims Next, we turn to the District Court s certification of Marcus s NJCFA claims. At the outset, it is important to disentangle two fundamental questions that are at the core of the District Court s analysis but that can be conflated: (1) Under New Jersey law, what must a plaintiff prove to succeed 39 on an NJCFA claim and what evidence can a defendant put forth to rebut and defeat that claim?; and (2) When a plaintiff seeks to certify an NJCFA claim for class treatment under Rule 23(b)(3), when might common questions of fact fail to predominate over individual ones? We begin with examining the elements of a NJCFA claim. The NJCFA prohibits certain deceptive commercial behavior that it calls unlawful practice[s]. The act, use or employment by any person of any unconscionable commercial practice, deception, fraud, false pretense, false promise, misrepresentation, or the knowing, concealment, suppression, or omission of any material fact with intent that others rely upon such concealment, suppression or omission, in connection with the sale or advertisement of any merchandise or real estate, or with the subsequent performance of such person as aforesaid, whether or not any person has in fact been misled, deceived or damaged thereby, is declared to be an unlawful practice . . . . N.J. Stat. Ann. § 56:8-2. The Attorney General of New Jersey may take action under the NJCFA to address an unlawful practice and, in doing so, does not have to prove that the victim of the fraudulent conduct had in fact been misled, deceived or damaged thereby . . . . Meshinsky v. Nichols Yacht Sales, Inc., 541 A.2d 1063, 1067 (N.J. 1988). 40 A private plaintiff, however, must do more than prove an unlawful practice. Another section of the NJCFA provides that [a]ny person who suffers any ascertainable loss of moneys or property, real or personal, as a result of the use or employment by another person of any method, act, or practice declared unlawful under this act or the act hereby amended and supplemented may bring an action or assert a counterclaim therefor in any court of competent jurisdiction. N.J. Stat. Ann. § 56:8-19 (emphases added). So, [w]hile the Attorney General does not have to prove that the victim was damaged by the unlawful conduct, a private plaintiff must show that he or she suffered an ascertainable loss . . . as a result of the unlawful conduct. Meshinsky, 541 A.2d at 1067. 12 An ascertainable loss is either an out-of-pocket loss or a demonstration of loss in value that is quantifiable or measurable. Thiedemann v. Mercedes-Benz U.S.A., LLC, 872 A.2d 783, 792-93 (N.J. 2005). Put differently, a plaintiff is not required to show monetary loss, but only that he 12 At times, Marcus erroneously suggests that the lower standard of proof applicable to actions by the Attorney General applies to his claims as well. See Marcus Br. at 2223 ( Even if . . . class members had differing levels of knowledge, such an inquiry is irrelevant under the NJCFA because Defendants may be liable under the NJCFA whether or not any person has in fact been misled, deceived or damaged by their deception. ) (quoting N.J. Stat. Ann. § 56:8-2). 41 purchased something and received less than what was promised. Union Ink Co., Inc. v. AT&T Corp., 801 A.2d 361, 379 (N.J. Super. Ct. App. Div. 2002); see also In re Mercedes-Benz Tele Aid Contract Litig., 257 F.R.D. 46, 73 (D.N.J. Apr. 24, 2009) ( [T]he sum of each class member s loss is the amount necessary to fulfill his or her expectation of a functioning [product].) . Importantly, unlike common law fraud, the NJCFA does not require proof of reliance. See Gennari v. Weichert Co. Realtors, 691 A.2d 350, 366 (N.J. 1997). That is, it does not require proof that a consumer would not have purchased a product absent the alleged unlawful practice or even proof that the unlawful practice played a substantial part in his or her decisionmaking. Cf. Restatement (Second) of Torts § 546. Nonetheless, the CFA requires a consumer to prove that [his or her] loss is attributable to the conduct that the CFA seeks to punish by including a limitation expressed as a causal link. Bosland v. Warnock Dodge, Inc., 964 A.2d 741, 748 (N.J. 2009); see also Meshinsky, 541 A.2d at 1067 ( [A] plaintiff must establish the extent of any ascertainable loss, particularly proximate to a misrepresentation or unlawful act of the defendant condemned by the [Act]. ) (quoting Ramanadham v. N.J. Mfrs. Co., 455 A.2d 1134, 1136 (N.J. Super. Ct. App. Div. 1982)). In other words, the alleged unlawful practice must be a proximate cause of the plaintiff s ascertainable loss. Here, Marcus argues that BMW and Bridgestone violated the NJCFA by failing to disclose salient facts about the Tires that would be material to the average consumer who is not knowledgeable about tires. Marcus Br. at 21. Specifically, he asserts that they failed to disclose the alleged defects previously discussed (i.e., susceptibility to damage, repairability, exorbitant price, etc.). He claims that, as a result of BMW s and Bridgestone s wrongdoing, he suffered 42 two forms of ascertainable loss: (a) he spent money to replace his tires that went flat, and (b) he leased a product that was worth less than what he expected. See J.A. 101, 103 (alleging ascertainable loss as monies spent to replace the Tires and/or in diminution in value ). To the extent Marcus relies on the cost of replacing his tire as his ascertainable loss, his claims cannot meet the predominance requirement for the reasons related to causation discussed above. See Oscar I, 274 F.R.D. at 513 ( [D]etermining whether each tire failed as a result of the allegedly concealed defect or as a result of unrelated issues, e.g., potholes or reckless driving habits, will devolve into numerous mini-trials. ). The District Court s NJCFA analysis, however, seems to have proceeded on the assumption that class members ascertainable losses would be the value of the product they expected to purchase minus the value of the product they actually purchased. The Court noted that, with respect to ascertainable loss, [t]he relevant issue . . . is whether the class members got less than what they expected. Marcus, 2010 WL 4853308 at *11. Hence unlike his common law claims, Marcus s NJCFA claims proceed on the theory that class members suffered a loss the instant they purchased or leased their cars, not when their tires went flat. Compare J.A. 101, 103 (alleging with respect to NJCFA claims ascertainable loss as monies spent to replace the Tires and/or in diminution in value ), with Pl s Am. Br. in Support of Class Cert. (J.A. 1255) (arguing with respect to the common law claims that [a]ll class members were damaged when their tires suffered a flat and they were forced to pay for a new Tire or when they purchased road hazard coverage to 43 insulate them from financial hardship due to cost of the Tires ). 13 In response, BMW and Bridgestone point out that a person cannot succeed on an NJCFA claim against them if that person knew about the alleged defects, yet decided to purchase or lease a BMW anyway. In such a case, under the NJCFA any alleged unlawful practice by the defendants would not have been the proximate cause of the knowledgeable consumer s ascertainable loss. Turning to the requirements of Rule 23, they then point out that what a particular class member knew about Bridgestone RFTs before purchase or lease would require an individualized inquiry, making Marcus s NJCFA claims not susceptible to proof with common, class-wide evidence. The District Court concluded that a presumption of causation should apply to Marcus s NJCFA claims, and therefore that common issues of fact would predominate. See Marcus, 2010 WL 4953308 at *12. According to the Court, a causal relationship between an alleged unlawful practice and a consumer s ascertainable loss may be presumed under the NJCFA when a defendant is alleged to have omitted (rather 13 We do not consider whether Marcus s damages allegations raise another set of predominance problems on which we may soon receive relevant guidance from the Supreme Court. See Comcast Corp. v. Behrend, No. 11-864, --S. Ct.-- (June 25, 2012) (granting certiorari on the question of [w]hether a district court may certify a class action without resolving whether the plaintiff class has introduced admissible evidence, including expert testimony, to show that the case is susceptible to awarding damages on a class-wide basis ). 44 than affirmatively misrepresented) material information in written representations and when a defendant s marketing statements do not differ from one consumer to another. Id. If the Court is correct, then Marcus would be able to prove his NJCFA claims with nothing more than evidence of BMW s and Bridgestone s unlawful practices (thus common issues of fact would predominate). To justify this conclusion, both New Jersey law and Rule 23 require factual findings that the District Court did not make. What a consumer knew about Bridgestone RFTs prior to purchasing or leasing his or her car is highly relevant to whether that consumer can succeed on an NJCFA claim. The District Court correctly noted that [t]he relevant issue . . . is whether the class members got less than what they expected. Marcus, 2010 WL 4853308 at *11. But what a class member expected of Bridgestone RFTs and BMWs depends on what information, if any, about the alleged defects was available during the class period and whether that class member knew about it. If a consumer did know about the defects but, despite that knowledge, still decided to purchase or lease a BMW at the same price anyway because, for example, he or she decided that the other safety and convenience benefits RFTs and BMWs offer outweigh the costs of their defects then the consumer would not have received something less than expected. If the evidence indicates that this could be true for a significant number of class members, then common questions of fact will not predominate. Instead, individual issues about what each class member expected when purchasing or leasing his or her car would swamp the inquiry, making the NJCFA claims inappropriate for class treatment. Because, as we explain below, the District Court made its presumption of causation ruling without making key factual findings that the NJCFA and Rule 23 require, it is an errant conclusion of law and an abuse of discretion. Hydrogen Peroxide, 552 F.3d at 312. 45 The Supreme Court of New Jersey has addressed when a presumption of causation may apply to NJCFA claims in two cases: International Union of Operating Engineers Local No. 68 Welfare Fund v. Merck & Co., 929 A.2d 1076 (N.J. 2007), and Lee v. Carter-Reed Co., LLC, 4 A.3d 561 (N.J. 2010). Both cases arose in the context of class actions brought under the New Jersey rule that, like Rule 23(b)(3), requires a finding of predominance. See N.J. Ct. R. 4:321(b)(3). Although we must follow a state s highest court when interpreting state law, see Ill. Nat l Ins. Co. v. Wyndham Worldwide Operations, Inc., 653 F.3d 225, 231 (3d Cir. 2011), we decide the legal standards that Rule 23 requires as a matter of federal law. See Hydrogen Peroxide, 552 F.3d at 312. Given this confluence of state and federal law, we take pains to isolate what the NJCFA requires and what Rule 23 requires in this case. International Union involved a putative NJCFA class action against Merck, the manufacturer and marketer of the pain medication Vioxx. The class consisted of third-party payors, such as corporate health insurers and union benefit organizations, who administer health benefit plans and make payments to pharmaceutical companies for medications prescribed to their members. Int l Union, 929 A.2d at 1080. The plaintiff alleged that Merck fraudulently marketed Vioxx as safer and more effective than similar medications, and thereby caused class members to afford Vioxx preferred status in their formularies, the approved listings of drugs third-party payors have authorized for purchase. Id. at 108081. Merck argued that class certification was inappropriate because class members differed greatly in how their formularies dealt with Vioxx, especially as new information about the drug became available. Id. at 1081. Initially, some third-party payors gave Vioxx preferred status and assigned it 46 to a low co-payment tier. Others put it in a non-preferred tier with high co-payments. Critically, Merck offered evidence that third-party payors, when they eventually learned of the supposedly fraudulent misrepresentations and omissions, did not react uniformly. Id. Merck s expert testified that decisionmakers for third-party payors responded to ongoing releases of information about Vioxx in different ways, with some altering its placement in their formularies. Id. Merck asserted that the essence of the claims must be that each class member was injured individually when it decided, based on defendant s allegedly fraudulent conduct, to include Vioxx in its formulary and, more to the point, when it paid for the drug purchased by plan members in its home state. . . . [B]ecause each class member made that evaluation at different times and based on different criteria and because each reacted differently with regard to formulary placement even after the facts that allegedly prove the fraud became known, there are overwhelmingly individual, rather than common, questions of law and fact. Id. at 1087 (emphasis added). The Supreme Court of New Jersey agreed with Merck that certification of a NJCFA class is not proper when class members do not react to misrepresentations or omissions in a sufficiently similar manner. Plaintiff does not suggest that each of these proposed class members, receiving the same information from defendant, reacted in a uniform or even similar manner. Rather, the 47 record speaks loudly in its demonstration that each third-party payor . . . made individualized decisions concerning the benefits that would be available to its members for whom Vioxx was prescribed. The evidence about separately created formularies, different types of tier systems, and individualized requirements for approval or reimbursement imposed on various plans members and, to some extent, their prescribing physicians, are significant. That evidence convinces us that the commonality of defendant s behavior is but a small piece of the required proofs. Standing alone, that evidence suggests that the common fact questions surrounding what defendant knew and what it did would not predominate. Id. (emphasis added). As a matter of substantive New Jersey law, International Union instructs that when a plaintiff knows the truth behind a fraudulent misrepresentation or omission, that plaintiff may not succeed under the NJCFA. Merck presented evidence that some third-party payors knew about the facts that allegedly prove the fraud but did not alter the placement of Vioxx in their formularies. Id. at 1087. Even if other class members did not know that Vioxx was not a safer and more effective pain medication than other available products, the evidence spoke loudly that third-party payors reacted very differently to the available information. So from the class action perspective, the case also makes clear that the commonality of [the] defendant s behavior is but a small piece of the required proofs if class members do not react to defendants conduct in a sufficiently uniform or even similar manner. Id. 48 Three years later, New Jersey s Supreme Court decided Lee v. Carter-Reed Company, L.L.C., which the Court described as fully in line with International Union. Lee, 4 A.3d at 577. In Lee, a plaintiff sought to certify a class of purchasers of Relacore, a dietary supplement the defendants marketed as a weight-reduction product with the additional benefits of lessening anxiety and elevating mood. Id. at 566. According to the plaintiff, the defendants misrepresented the benefits of Relacore in their advertising and promotions. Id. The Court held that the plaintiff s NJCFA claims satisfied the predominance requirement. Id. at 579-80. It noted that [i]f the entire marketing scheme was based on fictional benefits that Relacore offered, then whether the class member enjoyed good or impaired health or took other medications would hardly matter. Id. at 580. The Court continued that [w]hen all the representations about the product are baseless, a trier of fact may infer the causal relationship between the unlawful practice the multiple deceptions and the ascertainable losses, the purchases of the worthless product. Id. Not surprisingly, the Lee Court gave no indication that there was evidence of any class members who could have known that Relacore was a worthless product but decided to purchase it anyway. Such purchasers would not have been able to succeed on their NJCFA claims under New Jersey law. Turning to the predominance requirement, the Lee Court noted that, [i]n contrast to this case, International Union involved a proposed class of diverse entities, which the plaintiff did not suggest reacted in a uniform or even similar manner to the same information broadcast by Merck. Id. at 582 (quoting Int l Union, 929 A.2d at 1076). Read together, International Union and Lee convince us that, before applying a presumption of causation to an NJCFA claim, a court must consider not only the defendants 49 course of conduct, but also that of the plaintiffs. Specifically, it must consider whether plaintiffs could have known the truth underlying the defendant s fraud. See McNair v. Synapse Grp., Inc., No. 06-5072 (JLL), 2009 WL 1872582, at *10 (D.N.J. June 29, 2009) ( Thus, to establish causation under the NJCFA the New Jersey Supreme Court [in International Union] looked not only to the defendant s conduct but also to the class members conduct . . . . ). Although not binding on us, we also find the Supreme Court of New Jersey s predominance rulings to be instructive. In the class action context, if the truth behind alleged defects is knowable and if evidence suggests that class members did not react to information about the product they purchased or leased in a sufficiently similar manner such that common issues of fact would predominate, then certification is improper. See Demmick v. Cellco P ship, No. 06-2163 (JLL), 2010 WL 3636216, at*16 (D.N.J. Sept. 8, 2010) ( [A]lleging a uniform course of conduct by itself is not sufficient to satisfy the predominance element for the causation element of a NJCFA claim. ). Relying heavily on the Appellate Division of the New Jersey Superior Court s decision in Varacallo v. Massachusetts Mutual Life Insurance, 752 A.2d 807 (N.J. Super. Ct. App. Div. 2000), Marcus argues nonetheless that we need not consider class members conduct here. In Varacallo, however, the record did not indicate that the facts underlying the alleged fraud were knowable to class members or, even if they were, that a sufficient number of class members would have purchased the product anyway. The case involved a putative class of New Jersey residents who purchased so-called vanishing premium whole life insurance policies. Id. at 808. A vanishing premium policy allows insureds to pay premiums out-of- 50 pocket for a limited number of years only, until such time as dividends can completely fund the purchased plan. Id. Plaintiffs claimed that Mass Mutual violated the NJCFA by intentionally inflating its advertised dividend rates and then concealing a plan to reduce dividends to insupportable levels. They contended that this information was not revealed to agents who used Mass Mutual s prepared illustrations in selling the policies. Id. Mass Mutual countered that class certification would be improper because individual issues of causation would predominate over common issues of law and fact. The Court held that if the plaintiffs established the core issue of liability, then they would be entitled to a presumption of reliance and/or causation. Id. at 818. It explained: It is inconceivable to us, considering the assumption that Mass Mutual s liability is premised on a knowing omission of material information, that discovery will reveal more than a very small number of policyholders who would have purchased an N-Pay Policy, rather than a competitor s policy, if the Mass Mutual literature stated that the illustrated dividends probably will, rather than may, decrease and that the payout period probably will, rather than may, be longer than projected. Plaintiffs are required to prove only that defendant s conduct was a cause of damages. They need not prove that Mass Mutual s conduct was the sole cause of loss. . . . . It may be that some extraordinary sales agent could overcome such negative information[,] thereby becoming the superceding cause of a 51 policyholder s loss, but we think it a small possibility. Id. at 816-17. In short, unlike in International Union, there was no evidence that class members could have known the truth behind the defendant s representations and it was inconceivable that more than a very small number would have purchased their policies despite knowing the risks that defendants allegedly concealed. In this case, the District Court should not have focused exclusively on BMW s and Bridgestone s conduct, and should have made factual findings critical to the predominance analysis. Before certifying a class, the Court needed to have found (among other things) either (1) that the alleged defects were not knowable to a significant number of potential class members before they purchased or leased their BMWs, or (2) that, even if the defects were knowable, that class members were nonetheless relatively uniform in their decisionmaking, which would indicate that, at most, only an insignificant number of class members actually knew of the alleged defects and purchased or leased their cars at the price they did anyway. These findings cannot be side-stepped. They are necessary to determine whether the predominance requirement is met in this case. See Hydrogen Peroxide, 552 F.3d at 316-18. If class members could have known of the alleged defects and the evidence shows that they do not react to information about the cars and tires they purchased or leased in a sufficiently uniform manner, then individual questions related to causation will predominate. See In re Ford Motor Co. E-350 Van Prods. Liab. Litig. (No. II), No. 03-4558, 2012 WL 379944 at *14-*15 (D.N.J. Feb. 6, 2012) (denying certification because evidence showed that significant information about the alleged defects was in the public domain and, as a result, class members who knew the E-350 van to have significant handling problems will have a 52 difficult time proving causation, and in doing so, they would not rely on common proof ). Marcus disputes that the drawbacks of the tires were knowable to class members before purchase. He claims that [d]efendants have not produced any evidence showing that that information [about the drawbacks] was or could have been communicated to some class members prior to purchase. Marcus Br. at 36. The defendants have, however, shown significant evidence that the allegedly omitted information was available to class members from various sources. See, e.g., J.A. 462-554; 940-948, 961-1108; 2308-2310, 2313. The District Court did not explicitly address this issue, however, and it should be the first Court to do so. Furthermore, the District Court may find that class members did not react uniformly when presented with information about the cars they purchased or leased. The evidence might suggest that a significant number of class members could have known of the alleged defects, but decided to purchase or lease their cars at the same price anyway. This may be so because other class members may consider the features of Bridgestone RFTs and BMWs that Marcus styles as defects to be simply trade-offs. Some purchasers or lessees may have found that the significant safety and convenience benefits RFTs offer in the event of a flat tire outweighed their downsides. See, e.g., J.A. 961 ( Despite the disadvantages and inconveniences of run-flat tires for many, Consumer Reports believes that the safety benefits can outweigh the downsides. ). Others may simply have had the means to purchase or lease the luxury car of their choice and brush off how exorbitantly expensive it is to replace their tires and how frequently they must do that. For some consumers, the RFTs may have been an important factor; for others, not at all; for others, somewhere in between 53 . . . . Oscar II, 2012 WL 2359964 at *4. If, on remand, the evidence shows that no more than very few class members could have known of the alleged defects and for reasons either related or unrelated to RFTs would have still purchased or leased their cars at the same price despite that knowledge, then perhaps Marcus s NJCFA claims may be suitable for class treatment. If not, then individual proof would be needed to determine whether a particular class member broke the causal link between defect and ascertainable loss by purchasing or leasing a car despite knowing of the alleged defects. That would distinguish this case from Lee and Varacallo. Lee involved a worthless product for which all representations . . . [were] baseless. 4 A.3d at 580. There was no reason for the Lee Court to believe that a significant number of class members would, despite knowing that the product was worthless, purchase Relacore anyway. Similarly, in Varacallo it was inconceivable that more than a very small number of plaintiffs would purchase a vanishing premium insurance policy despite knowing that the promised dividend rates were inflated, making that knowledge a superceding cause of a policyholder s loss. 752 A.2d at 816-17. Here, if the evidence shows on remand that the defects were knowable and that more than a very small number of class members would have purchased or leased their cars despite knowing of those defects, then Jeffrey Marcus is destined to meet the same fate on his RFT claims as Gerald Oscar met on his. See Oscar I, 274 F.R.D. 498; Oscar II, 2012 WL 2359964. 54 VI. CONCLUSION With this context, we vacate the District Court s certification order and remand for proceedings consistent with this opinion. 55