Class certification typically turns on an analysis of the familiar elements of Federal Rule 23(a) and (b). While much has been written about these standards, surprisingly little has been written about the most critical element of all: the class definition. Like a canary in a coal mine, improper class definitions are often the first sign of deeper and sometimes fatal problems that get to the heart of the class itself.

Randleman v. Fidelity Nat’l Title Ins. Co., 646 F.3d 347 (6th Cir. 2011), is the first case from the Sixth Circuit to expressly address the fail-safe class, which is defined as a class that requires a decision on the merits of a claim in order to determine who is within the class.

The case is significant to all class action practitioners because it highlights a pleading challenge for plaintiffs, and a potent defense for defendants. The case also illustrates how a flawed class definition can be symptomatic of claims that are inappropriate for class treatment.

The Class That Paid Too Much

Randleman was a consolidated appeal that included a companion case captioned: Hickman v. First American Title Ins. Co., 2010 WL 3075157 (N.D. Ohio Aug. 5, 2010). Both cases raised virtually identical substantive claims against Ohio-based title insurance companies, but with slightly different formulations of the class definition. Here’s what happened.

Both the Randelmans and the Hickmans refinanced their homes. In connection with these transactions, both were required by their mortgagees to purchase new title insurance policies. Randelman, 646 F.3d at 349, 351. Under Ohio law, the Ohio Title Insurance Rating Bureau, (the “Bureau”), requires that consumers receive a discounted premium rate if they refinance within 10 years of having received title insurance from another company on the same property. Id. at 349.

Although both the Randelmans and the Hickmans were eligible for the “refinance rate,” neither knew that the discount existed and neither asked for it. Id. at 350, 351. As a result, they paid $213.57 and $134.40 more for their policies, respectively, than they should have. Id.

According to the Rate Manual produced by the Bureau, title insurance companies are required to offer the refinance rate:

. . . provided the Insurer is given a copy of the prior policy, or other information sufficient to enable the Insurer to identify such prior policy upon which reissue is requested, and the amount of the unpaid principal balance secured by the original loan.

Both appellants brought suit against their insurance companies seeking to certify a class. Here is where their stories diverge, ever so slightly.

The Randleman’s class was defined as consisting of all who had purchased title insurance from Fidelity who “were entitled to receive the ‘reissue’ or ‘refinance’ rate for title insurance” and who “paid more than the ‘reissue’ or ‘refinance’ rate for such insurance.” Id. at 350 (emphasis added).

The Hickmans defined their class as including all those who purchased title insurance from First American, and who “paid more than the discounted ‘reissue’ or ‘refinance’ rate for such title insurance.” Id. at 351 (emphasis added).

Thus, both definitions required a finding that a borrower paid too much for their title insurance premium.

According to the Sixth Circuit, the difference between the definitions was the fact that “membership in the Hickman’s class does not turn on an ‘entitlement’ to receive the discount, but includes all who refinanced a mortgage within the 10-year look-back period.” Id.

The Class That Asked Too Much

Affirming denial of certification in both cases, the Sixth Circuit started its analysis with the Randlemans. According to the court, the Randleman’s class definition was fail-safe because “it only included those who are ‘entitled to relief.’” This “shields the class members from receiving an adverse judgment. Either the class members win or, by virtue of losing, they are not in the class and, therefore, not bound by the judgment.” Id. at 352.

The court rejected this heads-I-win-tails-you-lose class definition, concluding that it was “an independent ground for denying class certification.” Id. And so it is. “The existence of an ascertainable class of persons to be represented by the proposed class representative is an implied prerequisite of Federal Rule of Civil Procedure 23[.]” Romberio v. UnumProvident Corp., 385 F.App’x. 423, 431 (6th Cir. 2003) (quoting John v. Nat’l Sec. Fire and Cas. Co., 501 F.3d 443, 445 (5th Cir. 2007)).

A “class definition is inadequate if a court must make a determination of the merits of the individual claims to determine whether a particular person is a member of the class.” MOORE’S FEDERAL PRACTICE §23.21[3][c] (3d ed. 2007). By this rationale, the Hickman’s class definition was surely fail-safe as well. After all, in order to be a member of their class, a plaintiff had to show they had been charged too much for the policy premium. That was the ultimate liability issue in the case.

Either a plaintiff paid too much and is in the class or, by virtue of failing to make that showing, is excluded from the class and not bound by the judgment. Surely omitting the word “entitled” wouldn’t change the fail-safe character of the class since the substantive inquiry — and the risks of certification — are the same. If it could, the result would exalt form over substance and reduce the ascertainability requirement to an exercise in semantics.

Yet the Sixth Circuit did not directly address the fail-safe issue for the Hickman class. Instead, it said that the Hickman class failed “[f]or the same reasons the Randlemans failed to establish that common issues predominate . . . with respect to the class they propose.” Randleman, 646 F.3d at 355.

Indeed, the Hickmans, like the Randlemans, had proposed classes that required individualized, factual inquiries into each loan file. According to the Sixth Circuit, “determining liability would require an examination of each individual policy . . . to determine if the [insurer] had received a copy of the prior policy or ‘other information’ suggesting that a policy had previously been issued.” Id. at 353. Since both plaintiffs proposed to certify their class pursuant to Rule 23(b)(3), which requires that common issues “predominate” over individual ones, their classes could not be sustained. Id.

Randleman is instructive for several reasons. First, the court was on firm ground when it rejected the Randleman’s class definition as fail-safe. Definitions that require a merits determination just to figure out who is in the class are flawed and should be stricken. In Randleman, the Sixth Circuit joined other courts in so holding.

And Randleman is no anomaly. Fail-safe class definitions appear with surprising frequency, and other courts within the Sixth Circuit have not hesitated to strike them.1 Randleman teaches that a careful examination of the class definition can yield significant results in defending against class claims.

The dilemma for defendants is whether to challenge the class definition early in the proceedings, or to wait until the class certification process. The Randleman class definition was addressed on appeal, following extensive pre-certification discovery. Yet where there is a facial flaw in the class definition, there is no reason to wait until the certification phase to move to strike the definition. Indeed, the flaw in the Randleman’s definition appeared in the original complaint. There was arguably no need for the district court require extensive certification discovery to decide that discreet issue.

The Sixth Circuit’s election not to analyze the Hickman definition — which was as flawed as the Randleman’s — should not be understood as encouraging an easy out for class plaintiffs. Simply removing the word “entitled” does not really solve the definitional flaw where the remaining elements of the class definition are also dispositive of all or part of the main claims.

The Sixth Circuit’s resolution of the Hickman claim reveals a deeper truth about fail-safe definitions: they often invite the same individualized inquiries that run afoul of the “predominance” standard of Rule 23(b)(3). Take the Hickmans. In order to determine whether a class member “paid more than the discounted ‘reissue’ or ‘refinance’ rate.’” requires the court to decide, plaintiff by plaintiff, file by file, whether each meets the requirements of the class.

The individualized claims swamp the common ones, and will inevitably confuse class and merits issues. And that’s the second lesson in Randleman. Merit-driven class definitions foretell individualized inquiries that undermine the whole point of class certification: “Promotion of efficiency and economy of litigation.” Crown, Cork & Seal Co. v. Parker, 462 U.S. 345, 349 (1983). Linking the fail-safe definition to an individualized and fact-intensive class inquiry presents a compelling argument to any court to strike a fail-safe definition at the pleading stage of the case.

Thus, in Randleman the Sixth Circuit established that fail-safe classes will not be permitted. The legacy of the failed Hickman class — and the judicial resources expended to deny certification — is that courts should not exalt form over substance. Instead, class definitions that require a merits determination of any of the elements of the class should also be stricken as fail-safe.