On August 16, 2011, Judge Jack Weinstein of the U.S. District Court for the Eastern District of New York rendered a decision in Haynes v. Planet Automall, Inc., No. 09-CV-3880 (E.D.N.Y. Aug. 16, 2011), rejecting class certification of claims by purchasers of used cars alleging violations of the federal Truth in Lending Act (“TILA”). The ruling is noteworthy for two reasons: (i) Judge Weinstein has a track record of certifying class actions via “push the envelope” interpretations of Rule 23, and (ii) the rejection of the plaintiff’s class certification theory in Haynes shows the power on the new commonality requirements established in Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541 (2011), and its application to statistical evidence presented by a plaintiff in support of a Rule 23 certification motion.
In Haynes, the named plaintiff had purchased a used car using credit provided through the dealer. She alleged that the dealer had failed to disclose that various fees charged to customers who received dealer-assisted financing were part of the finance charge for the automobile as required by TILA. She sought to certify a class of individuals who were similarly misled. The defense opposed the motion, contending that plaintiff’s claims were based on different individual oral representations, rendering commonality unlikely and a class action inappropriate.
In denying plaintiff’s motion for class certification the Court held that plaintiff failed to show that the critical disputed questions – whether the fees are part of the finance charge pursuant to TILA – could be answered uniformly on a class-wide basis, and had therefore failed to establish Rule 23(a)(2)’s commonality requirement. Id. at 24. Plaintiff had proposed the following set of state law questions, which she argued were subject to uniform adjudication: “whether the alleged violations of TILA constitute deceptive practices under New York General Business Law § 349; caused defendants to be unjustly enriched under New York law; constitute a constructive trust under New York law; or warrant other damages.” Id. Relying on the Supreme Court’s recent Dukes decision, Judge Weinstein held that such questions were peripheral to the certification decision. He reasoned that “[r]equired is a determination that plaintiff’s claims ‘depend up on a common contention’ that is ‘of such a nature that it is capable of class-wide resolution.’ . . . A claim is capable of class-wide resolution when the ‘determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke.’” Id. at 13 (quoting Dukes, 131 S. Ct. at 2551). Because plaintiff’s proposed common questions went to remedies and would not resolve the central disputed issues, the Court concluded that those questions could not satisfy the commonality requirement. Id. at 24.
As to the disputed questions themselves, the Judge Weinstein rejected plaintiff’s attempt to use statistics as a means to resolve those issues on a uniform, class-wide basis. The critical question was whether the fees at issue were part of the finance charge. If so, then TILA requires that they be disclosed to creditors to ensure that consumers are not misled about the cost of credit. TILA defines the finance charge as “the sum of all charges, payable directly or indirectly by the person to whom the credit is extended, and imposed directly or indirectly by the creditor as an incident to the extension of credit.” Id. at 15 (quoting 15 U.S.C. § 1605(a)). In the Second Circuit, a charge must be incident to, or a condition of, the extension of credit in order to be considered a finance charge. Id. at 15 (citing Pechinski v. Astoria Fed. Sav. & Loan Ass’n, 345 F.3d 78, 80 (2d Cir. 2003)). The determination of whether a charge is incident to the extension of credit is “extremely fact-intensive,” and the key is to show that the extension of credit caused plaintiff to pay a particular fee or charge. Id. at 16 (quoting McAnaney v. Astoria Fin. Corp., 357 F. Supp. 2d 578, 584 (E.D.N.Y. 2005)).
Judge Weinstein applied the approach of the Seventh and Sixth Circuits to determining causation for a class. Relying on Gibson v. Bob Watson Chevrolet-Geo, Inc., 112 F.3d 283 (7th Cir. 1997), and Cornist v. B.J.T. Auto Sales, Inc., 272 F.3d 322 (6th Cir. 2001), the Court held that “[i]f plaintiffs can prove that defendants systematically charged a fee to credit customers that they did not charge to cash customers, then a factfinder could reasonably infer that he fee was ‘incident to the extension of credit.’” Haynes, at 18. Moreover, proving that a defendant systematically charged a particular fee to credit customers but not to cash customers would require plaintiff to demonstrate that the defendants applied a uniform policy to its customers that distinguished between the two. Id. Although the Court determined that such a demonstration could be made on other forms of proof, the plaintiff in Haynes had decided to try to prove her case relying on statistics alone. Id. at 19-22.
The Court concluded that plaintiff’s statistical evidence was insufficient to carry this burden. That evidence showed that 63.9% of credit customers and 4.1% of cash customers were charged certain fees during the relevant time period. Id. at 24. The Court noted that this evidence showed that the fees were not charged exclusively to credit customers, and that plaintiff had provided no explanation for why cash customers would ever be required to pay a fee that plaintiff alleged was incident to the extension of credit. Id. at 25. Second, the percentage of credit customers who were charged the fees was simply not high enough. Id. The Court found that plaintiff needed to demonstrate that the fees were systematically charged to all credit customers; while she had convincingly shown a correlation between purchasing a car on credit and being charged the challenged fees, the statistical evidence was not sufficient to show that a systematic illegal policy was in place. Id. The Court analyzed plaintiff’s statistics concerning extended warranty fees and came to the same conclusion. Id. at 25-26.
Although the opinion in Haynes focused primarily on commonality, Judge Weinstein briefly addressed the typicality and adequacy prongs of the class certification analysis. The Court held that the oral nature of negotiations for price and services, which would vary in each individual case, along with the lack of any showing of a uniform approach by salespersons, suggested that typicality could not be established. Id. at 26. Moreover, because plaintiff had demonstrated that she may have an animus towards defendants concerning events that were unrelated to her action concerning the disclosure of finance charges, she had not demonstrated that she would be able to litigate procedural and tactical questions on a wholly objective basis, independent of her own special relationship with defendants. Id. at 27.
Haynes illustrates some important lessons for litigants involved in workplace class actions. Judge Weinstein is known as a friend of the class action device, and therefore his denial of plaintiff’s motion for class certification is noteworthy in and of itself. As his opinion cited his view that a “class action serves an important function in protecting consumers from abusive business practices,” and that consideration given to plaintiffs’ motion therefore “must be respectful and generous,” Id. at 3-4, the decision is Haynes is well worth a read by corporate counsel.
First, plaintiffs could not meet the commonality requirement simply by pointing out the fact that there exist questions of law or fact that are subject to universal class-wide determination. Rather, those questions must lie at the core of the case, in the sense that they are central to the validity of all class members’ claims. Plaintiff’s attempt to describe such questions in Haynes was insufficient because her common questions each required the resolution of a more fundamental question. Any common questions concerning remedies, for example, presuppose the underlying violation that confers the right to a remedy. A common question must go to that underlying violation, rather than the remedies themselves. In Haynes, the fundamental common question was whether the fees at issue were part of the finance charge and thus had to be disclosed. Plaintiff had not met her burden of showing that that question was subject to universal, class-wide determination. Similarly, a workplace discrimination class action might involve many common questions concerning the company’s policies and practices relative to compensation, evaluations, or promotions. Unless those common questions are central to the validity of plaintiffs’ claims of discrimination, they are peripheral to the class certification analysis and not apt to be successfully certified.
Second, statistical evidence is a common form of proof in workplace class actions, especially for those alleging discrimination. Haynes shows just how strictly courts will construe statistical evidence in determining class causation issues. Although a systematic policy can be proven through the use of statistics alone, Judge Weinstein determined that those calculations must be of “an overwhelmingly convincing nature.” Id. at 19 (quoting United States v. Rioux, 97 F.3d 648, 658 (2d Cir. 1996)). In Haynes, this meant that plaintiffs had to have an explanation for why cash customers were sometimes (though rarely) charged fees that they alleged were charged only to credit customers, as well as an explanation for why those fees were not charged to credit customers closer to 100% of the time. The defense in workplace class actions should look for similar soft spots in the statistical analyses presented by plaintiffs. Haynes shows that courts will be open to the argument that plaintiffs’ statistical evidence is insufficient where plaintiffs cannot explain instances where their evidence appears to contradict their claims.