In Cox v. Sherman Capital LLC, 2016 WL 274877, at *4-6 (S.D.Ind., 2016), Judge Pratt properly found that a proposed class action was an impermissible fail-safe class.  The allegations were as follows:

As a result of securitization, the Plaintiffs contend that LVNV did not own their debts based on three different, but related, arguments.
First, as explained above, Plaintiffs argue that, after their debts were “written-off” by the originating bank, the Defendants could only legally purchase “data” of the debts or receivables. Thus, the Defendants falsely represented that LVNV “owned” the Plaintiffs’ debts when it engaged in collection activities against them because it only owned “data” of the Plaintiffs’ debts. (See Filing No. 303 at 58, 70; Filing No. 493 at 3.) Second, Plaintiffs contend that even if LVNV eventually acquired title to their debts, LVNV prematurely began collection activities when it was still only in possession of the “data” and before LVNV had obtained title to the debts. Therefore, Defendants falsely represented that LVNV “owned” the Plaintiffs’ debts when it engaged in collection activities against them. (See Filing No. 493 at 2-3.)  Third, Plaintiffs argue that regardless of whether the Defendants acquired “data” or title to their debts, the Defendants securitized whatever they owned (either data or debt) into a new asset-backed security after obtaining it. Plaintiffs assert that as a result of this second securitization, LVNV became a “servicer” to the new asset-backed security rather than an “owner” of the Plaintiffs’ debts or receivables. As a result, the Defendants falsely represented that LVNV “owned” the Plaintiffs’ debts when LVNV was actually a “servicer” of the new asset-based security. (See Filing No. 303 at 70; Filing No. 440 at 4-5.)

The Court found that class certification was improper, based on the fail-safe class definition.

The Plaintiffs seek certification of three sub-classes. The first, a FDCPA subclass, is proposed as follows, “All Indiana citizens who were the subject of collection activity or activities which violate the FDCPA by the Defendants or Defendants’ agents in an attempt to collect a debt incurred for personal, family or household purposes which were served with process or contacted in any matter by Defendants or Defendants’ agents during the period beginning November 9, 2011 (one year prior to the filing of the original complaint in this action) through trial of this caseæ.. . A “fail-safe” class is one that is defined so that “whether a person qualifies as a member depends on whether a person has a valid claim.” Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 825 (7th Cir. 2012); see also Dafforn v. Rousseau Assocs., Inc., Civil No. F 75-74, 1976 WL 1358, *1 (N.D. Ind. July 27, 1976) (defining a “fail-safe” class as a class “which would be bound only by a judgment favorable to plaintiffs but not by an adverse judgment.”). Such a definition is improper because a class member either wins or, by virtue of losing, is defined out of the class and is therefore not bound by the judgment. Messner, 669 F.3d at 825. “Rule 23 was never meant to be an exception to the rules of res judicata or to provide a risk-free method of litigation. The class definition must be such that all (except those who opt out) are as much bound by an adverse judgment as by a favorable one.” Dafforn, 1976 WL 1358 at *1.   Each of the proposed sub-classes includes the language of a valid claim in its definition. For instance, the FDCPA sub-class contains the criterion that the Defendants’ collection activities “violate the FDCPA”. Similarly, the RICO sub-class includes the criterion that the Defendants acted pursuant to a “scheme to defraud”. Finally, the Restitution sub-class includes the criterion that the class members paid an “alleged debt”. It is reasonably foreseeable that, should the class members fail to prove their claims based on one of these claim-specific criteria, they would not be bound by the judgment because they would no longer be part of the class. See, e.g., Dafforn, 1976 WL 1358 at *1 (concluding that a class defined by whether a homeowner was charged “an artificially fixed and illegal brokerage fee” was a “fail-safe” class because a jury determination that the defendants did not charge an illegal fee would determine there was no class and allow absent class members to relitigate the legality of the defendant’s fee structures). Accordingly, as defined, the Plaintiffs’ proposed sub-classes are improperly “fail-safe”. In this regard, the Plaintiffs urge the Court to exercise its discretion and redefine the class instead of denying class certification. See Messner, 669 F.3d at 825 (noting that the “fail-safe problem” can be often be avoided by refining the class definition rather than denying class certification on that basis). However, the Plaintiffs do not offer any alternatives in their motion, and the Court is unable to conceive of a class definition that would encompass the Plaintiffs’ claims and still satisfy the requirements of Fed. R. Civ. P. 23.. . Because Plaintiffs’ three proposed sub-classes are “fail-safe” classes and the Court does not conceive how the class can be amended to avoid individual factual inquiries, class certification is not appropriate for this reason alone.