AN EMPIRICAL STUDY OF CLASS-ACTION APPEALS.

• Author: Lammon, Bryan

Federal Rule of Civil Procedure 23(f) allows parties in a class action to seek immediate appellate review of class-certification decisions. Criticisms of the rule abound. Some see Rule 23(f) as a tool defendants use to drag out litigation and avoid class certification. Others contend that the rule is inconsistently applied among the circuits and should be replaced with a right to appeal.

Yet there is little reliable data that might back up these criticisms. This article brings some hard data to this discussion. I created an original dataset of all Rule 23(f) petitions to appeal filed from 2013 through 2017. The data alone is revealing--it has been hard to come by reliable data on the number of petitions, the rate at which different courts grant them, and what those courts do (affirm or reverse) after granting a petition. The data also does not support common criticisms of 23(f). And--perhaps surprisingly--the data unveils one corner of the class-action universe in which plaintiffs are not predominantly losing: in the Rule 23(f) context, the courts of appeals reached a plaintiff-favorable outcome over 50% of the time.

1. INTRODUCTION

   A district court's class-certification decision is a key--if not the key--decision in any case brought as a class action. (1) If the district court certifies a class, the defendant can face thousands or even millions of individual claims--worth millions or even billions in potential damages--rather than just those of the named plaintiffs. Given the sharp increase in the case's stakes, most defendants will settle instead of risking a monumental damages award. A decision denying class certification, in contrast, means the case involves only the claims of the named plaintiffs. And these claims are often so small that pursuing the case makes no economic sense; the plaintiffs will either abandon their claims or settle for a relatively small amount. So the class-certification decision--not a decision on the merits of the plaintiffs' claims--can dictate who prevails in any case brought as a class action. A certified class is essentially a victory for the plaintiffs. The denial of class certification is essentially a plaintiff loss.

   To ensure some appellate review of class-certification decisions, the Supreme Court added Rule 23(f) to the Federal Rules of Civil Procedure in 1998. (2) That rule gives appellate courts discretion to hear an immediate appeal from a district court's class-certification decision. Rule 23(f) is unique among the statutes, rules, and judicial doctrines that govern appellate jurisdiction. It is one of the few instances in which the courts of appeals have discretion over whether to hear an appeal. (3) And it is the only instance in which the Supreme Court has used its relatively new power under the Rules Enabling Act to craft a rule of appellate jurisdiction. (4)

   The rule has also been a persistent target of criticism. A few observers have suggested that the rule is working well, providing sufficient opportunities for appellate review and developing the law of class actions. (5) But most observers are unhappy with it. (6) Academics are almost unanimous in viewing Rule 23(f) as a tool that benefits defendants and limits the availability of class actions. (7) Defense-side interests contend that the courts are inconsistent in applying Rule 23(f) and that the rule insufficiently protects defendants from the pressure to settle. (8)

   But like so much of the universe of class actions, we know little about how Rule 23(f) actually operates in the courts. (9) Most denials of Rule 23(f) petitions are available only on the courts' dockets and thus fly under the radar. When Rule 23(f) petitions are granted, they normally receive only passing mention in the opinion that reviews the merits of class certification. And little reliable data on Rule 23(f) exists. (10) Without that data, it's difficult to show that Rule 23(f) is good, bad, or a bit of both. (11) But criticisms of the rule persist.

   To bring some hard data to this discussion, I created an original dataset of petitions to appeal under Rule 23(f) that were filed from 2013 through 2017. The numbers alone are illuminating. The data also provides little support for the common criticisms of Rule 23(f), at least during the period studied. (12) There is little evidence of "party effects"--i.e., that a relationship exists between the petitioning party (plaintiff or defendant) and the court's decision. I also found little evidence that the circuits are inconsistent in granting Rule 23(f) petitions or in reversing class-certification decisions in the Rule 23(f) context--what I call "circuit effects." And what evidence there is comes with some serious caveats.

   On top of this empirical analysis, my data also sheds some light on class actions generally. Although Rule 23(f) petitions offer only a glimpse into one corner of the class-action universe--and are almost certainly not representative of that entire universe--these petitions offer at least some evidence that class actions are still a viable means for plaintiffs to obtain relief. The class-action literature regularly diagnoses the class action as dead (or at least dying) insofar as it could be used to redress collective injuries. Five years of Rule 23(f) petitions show that plaintiffs are still filing class actions. These plaintiffs sometimes win at class certification--the dataset contains hundreds of plaintiff victories on this issue. And if class certification is really the whole game in cases brought as a class action, that means hundreds of successful class actions. Granted, these victories might be a small part of the larger class-action universe, as we do not know how representative Rule 23(f) petitions are of that universe. (13) But the data shows at least one corner of the class-action universe in which plaintiffs are not predominantly losing.

   I proceed as follows. Part II explains the law governing interlocutory class-action appeals, both before and after Rule 23(f), as well as common criticisms of that rule and prior empirical or semi-empirical studies. Part III presents my findings. I start with the basic numbers and then explore whether those numbers support the party-effects and circuit-effects criticisms. I end Part III with suggestions for future empirical research. And in Part IV, I discuss what light my study might shed on the larger class-action universe. Part V briefly concludes.

2. CLASS-CERTIFICATION APPEALS

   District court judges often decide a number of issues in the course of federal litigation. (14) Nearly all of these decisions are "interlocutory"--they are made at some point before the end of district court proceedings and leave other issues for later resolution. As a general rule, federal litigants must wait until the end of proceedings--when all issues have been decided and all that remains is enforcing the judgment--before appealing any of those interlocutory decisions. (15) This limit on appellate jurisdiction comes from 28 U.S.C. [section] 1291, which gives the courts of appeals jurisdiction over only "final decisions" of the district courts, and it is commonly called the "final-judgment rule." (16)

   But the final-judgment rule is only the general rule. It has lots of exceptions. Some are found in statutes. (17) Other exceptions come from judicial decisions. (18) And still others are found in rules of procedure. (19) My focus here is one of those procedural rules: Federal Rule of Civil Procedure 23(f), which permits discretionary appeals from class-certification decisions.

   1. Appeals from Class-Certification Decisions

      Federal cases don't begin as class actions. Plaintiffs purporting to represent a class file putative class actions, and class-wide representation comes only after the district court certifies the case as a class action. (20)

      When the plaintiffs seek damages, that certification decision can be a major moment in the purported class action. (21) If the district court certifies the class, the named plaintiffs now represent not only themselves but also every member of the class. (22) And if the class seeks damages, the defendant's potential liability has accordingly multiplied, sometimes immensely. Class certification can transform potential liability from a few thousand dollars into hundreds of millions or even billions. Defendants' incentive to settle thus increases upon class certification; whatever their chance of prevailing on the merits, the threat of an immense damages award can warrant settling. (23)

      If the district court denies class certification, the named plaintiffs have only their own claims. Denial can thus make the stakes of the case quite low and mark the end--or, more dramatically, the "death-knell"--of the action. (24) One of the key reasons to pursue claims on behalf of a class is because individual claims are too small to be worth the costs of litigating. Aggregating lots of small claims can turn a negative-value case into one that makes economic sense. But if class certification is denied, all that remains are the named plaintiffs' small claims. These plaintiffs face an incentive to abandon their negative-value claims or settle them for relatively small amounts, even if they might prevail on the merits. (25)

      For these reasons, the class-certification decision is often regarded as the main event in a purported class action, if not the entire game. (26) This creates a problem for appellate review of class-certification decisions. If class certification is denied, the named plaintiffs' decision to abandon or settle their individual claims is not appealable. And appellate review of a class-wide settlement requires an objector to the settlement. (27) Absent an exception to the final-judgment rule, the only way to obtain appellate review of a class-certification decision is to litigate a case to a decision on the merits.

      Before Rule 23(f)'s addition to the civil rules in 1998, the avenues for interlocutory...