Posner and Class Actions

Introduction

The hallmark of Judge Posner’s class action decisions is rigorous review to ensure that aggregate litigation serves the best interests of class members and does not unduly pressure defendants to settle. Although he championed class actions, especially as a way to provide efficient justice in cases involving numerous small claims, Posner also recognized that, because of the agency problems that pervade class action litigation, ordinary adversary procedures were not sufficient to protect class members.1 As a result, the judge had to act as a fiduciary for the class, especially when approving settlements and fee awards.2 In addition, the colossal liabilities potentially imposed by a class action meant that a defendant might settle even if the case had little merit, so judicial scrutiny—in particular interlocutory appellate review of certification decisions—was necessary to protect defendants.3

While Judge Posner’s rigorous review of class action litigation is sometimes misconstrued as hostility to class actions generally, this is not correct. It is true that, in his most famous and influential class action opinion, In the Matter of Rhone-Poulenc Rorer Inc,4 Judge Posner refused to certify a class and provided new and powerful tools for defendants to resist class certification.5 Nevertheless, in numerous other opinions, he extolled the benefits of class action litigation and provided arguments and innovative strategies to facilitate class certification, to provide for effective distribution of awards to class members, and to ensure full compensation to class lawyers.6

Economic analysis provided the analytic framework for most of Posner’s class action jurisprudence. Class actions were, in Posner’s view, particularly important when numerous persons suffered small but similar injuries. Without class actions, those cases would not be brought, and similar wrongs, potentially large in the aggregate, would not be deterred. As Posner famously quipped, “only a lunatic or a fanatic sues for $30.”7 In addition to allowing small-claims suits to be brought, class actions can “yield substantial economies in litigation” by eliminating duplicative legal expenses for both plaintiffs and defendants.8 These arguments in favor of class actions reflect views that Posner articulated decades earlier in his seminal economic analysis of procedure.9

The need for strict scrutiny of class actions also reflects economic considerations, principally agency costs. The lawyer is the client’s agent. Even in ordinary litigation, it is difficult for a client to effectively monitor and control the lawyer to ensure that the lawyer acts in the client’s best interest. That problem is aggravated in class actions, in which the lawyer effectively chooses the clients (the named representatives and the class) and in which individual class members, including the named representatives, usually have such small stakes that they have no incentive to monitor and control their lawyers. As a result, class lawyers may be tempted to enter into sweetheart deals, which provide ample fees to the lawyers, paltry compensation to class members, and minimal deterrence of future wrongdoing. This problem is exacerbated by the fact that there may be competing class actions or competing class action lawyers, so defendants may be able to play the lawyers off against each other in a “reverse auction” in order to negotiate an inexpensive settlement that bars concurrent and future litigation. Although the agency problems in class actions were most thoroughly documented by scholars such as Professors John Coffee, Jonathan Macey, and Geoffrey Miller,10 whose contributions Posner acknowledged,11 recognition of the conflict-of-interest problem inherent in class actions also goes back to Posner’s pre-judicial academic writings.12

In part because of his reliance on economic analysis, Posner had little use for traditional legal materials, including the Federal Rules of Civil Procedure (FRCP) and Supreme Court precedents. Posner’s opinions dug deeply into the facts to determine whether a class action was the best way to handle the claims and whether settlements reflected the merits of the case. While he did his best to explain how the rules and precedents could be interpreted to support his economic approach, he seldom engaged in careful analysis of the FRCP or case law but rather structured his opinions around his view of the crucial issues, occasionally tying his analysis back to authoritative legal sources. In his later, more candid years on the bench, which one lawyer called his “late decadent period,”13 Posner showed disdain for traditional legal reasoning and suggested that judges follow “common sense [ ], forgetting about the law” unless there was a statute or binding precedent blocking the sensible result.14 Posner advocated this approach explicitly in class actions, flatly admitting, “I don’t get a lot out of Rule 23. . . . Actually reading Rule 23, I just get lost in all the detail and the subsections.”15 Instead, he defended a “holistic” approach, which asks, “‘Is this an efficient way to deal with the dispute?’ ‘Is there a real class, a lot of people with a common interest?’ ‘Is their representation competent?’ ‘Are there clearly focused issues?’”16

The influence of Posner’s opinions on the development of class action case law and scholarship can be measured through citation analysis, but other methods better capture their true impact. A key vehicle of Posner’s influence has been, ironically, revisions of the Federal Rules of Civil Procedure, which Posner mocked for its confusing “detail and [ ] subsections.”17 When Posner came to the bench in 1981, the class action rules had not been revised since the 1966 amendments, which opened the way for mass tort litigation.18 Judge Posner’s 1995 decision in Rhone-Poulenc was the primary impetus for the 1998 amendments that allowed discretionary interlocutory appeals of orders granting or denying class certification.19 Posner’s decision was almost certainly also the source for the provision in the notes that justified that review, in part, by “the risk of potentially ruinous liability,”20 as that danger was a key argument in Rhone-Poulenc.21 The 2003 and 2018 amendments also reflect Posner’s influence, this time on the importance of rigorous judicial scrutiny of settlements, competent class counsel, and “[a]ctive judicial involvement in measuring fee awards.”22 While the Committee Notes are written in an arid style that conceals what influenced committee deliberations, Posner’s impact is clear both in the hearings23 and in the unofficial comments of committee and subcommittee members. Judge Anthony J. Scirica, for example, in discussing the 1998 amendment that allowed discretionary interlocutory appeals, noted:

Judge Posner’s jurisprudence, and the Rhone-Poulenc case in particular, was of great interest to the rules makers and was a defining moment in the way we looked at mass claims . . . [N]o case had articulated the problems to the same degree of detail and persuasiveness as Judge Posner.24

Similarly, Judge Robert Dow stated that Judge Posner was a “looming presence” in deliberations regarding the 2018 amendments because “many of the issues . . . came out of his opinions.”25

Judge Posner influenced not only the rules but also class action lawyers. Elizabeth Cabraser, a leading class action lawyer whose offices are in San Francisco, far from the direct supervision of the Seventh Circuit, confessed:

I did not read all of these [Posner class action] decisions in preparation for the [symposium] panel today, but I can tell you that I read every single one of them in real time when they came out, beginning with actually the second one, Rhone-Poulenc, because throughout my class actions practice, Judge Posner, more than any single jurist, has essentially been the boss of me. . . . We ask, “What would Richard Posner do?” Or, more to the point, “How would Richard Posner rule?” And that has been salutary.26

It is hard to imagine any other appellate judge whom a lawyer based in another circuit would have paid such close attention to, reading his class action opinions as they were handed down and asking, in internal deliberations, how he would rule. In addition, the fact that she, as a plaintiff’s lawyer, considers his influence overall to have been “salutary” confirms that Posner was concerned at least as much with ensuring that injured class members were competently represented as with protecting defendants from coercive settlements, if not more.

The rest of this Essay will amplify some of the themes touched on by this Introduction. Part I will discuss Posner’s certification decisions, with special emphasis on Rhone-Poulenc. Part II will analyze decisions on settlements and related issues (such as lawyer fees). Posner’s influence will be analyzed through citation analysis in Part III. Other themes in the Introduction—such as Posner’s use of economics, his disdain for traditional legal reasoning, and his influence on the FRCP and lawyers—will be touched on only tangentially, as there is little to add to what is already written in this Introduction.

I. Certification Decisions

Rhone-Poulenc merits separate analysis. It is Posner’s most cited class action decision.27 It was the primary influence on the 1998 FRCP Amendment allowing discretionary interlocutory appeals of certification decisions. It was cited extensively in the hearings on the Class Action Fairness Act,28 and it provoked spirited academic debate.

The plaintiffs in Rhone-Poulenc were hemophiliacs infected by HIV as a consequence of using blood solids manufactured by the defendants.29 The district judge had certified a class with respect to certain issues, and the defendants filed a petition for mandamus requesting appellate review.30 An ordinary appeal was not possible because certification of a class is not an appealable final order, and the Supreme Court had already decided, in Coopers & Lybrand v Livesay,31 that the collateral order doctrine, which carves out exceptions to the final order rule, does not apply.32 Mandamus is an extraordinary remedy, which, as Posner recognized, is supposed to be granted only when the district court “so far exceed[ed] the proper bounds of judicial discretion as to be legitimately considered usurpative in character, or in violation of a clear and indisputable legal right, or, at the very least, patently erroneous.”33 The district court judge’s decision may well have been erroneous and ill-advised, but it is hard to characterize as “usurpative” or “patently erroneous.” It was not like the Eastern District of Texas’s patent venue decisions, which tried to attract and keep as many infringement cases as possible in that district and which were thus properly the subject of repeated mandamus decisions by the Federal Circuit.34 Nor was the district court judge in Rhone-Poulenc so incompetent that his decision was patently erroneous. Posner had to advance several novel theories in order to demonstrate the trial judge’s error, and, as discussed in detail below, all have been subject to substantial academic critique. Rather, Posner clearly thought it would be better for certification decisions to be subject to more frequent and careful appellate review, and he was willing to bend mandamus doctrine to make that possible.

History has largely vindicated Posner’s view. Several other appellate courts soon followed Posner’s lead, and, as noted above, three years later, the Rules Committee amended the FRCP to allow discretionary appellate review without recourse to a writ of mandamus. Allowing appellate review of certification decisions has had a profound influence on class action jurisprudence, as certification decisions had previously been practically unreviewable. Because most class actions settle, the final judgment rule meant that very few certification decisions would ever be subject to appellate review. While 28 USC § 1292(b) allows appellate review of interlocutory decisions when both the trial judge and appellate court desire it, most trial judges are reluctant to voluntarily subject themselves to appellate review that could be otherwise avoided. By allowing appellate review of certification decisions, Rhone-Poulenc and the 1998 FRCP Amendment made possible the sort of rigorous review of class action decisions that Posner himself advocated. Without interlocutory review, Posner and other appellate judges would simply have been unable to examine the merits of the certification decision in the vast majority of cases.

Posner based his decision on three main arguments. The first, which is clearly where Posner’s heart lay, involved the danger that certification would coerce the defendants into settling a case that Posner thought was very weak.35 The danger of such a coercive settlement both provided a reason for interlocutory review—because if the case settled, no appeal could later be taken on the certification issue—and showed why injustice would result if certification were not reversed.

Posner’s second argument was that Rhone-Poulenc was a class action based on state law, but state laws vary on relevant legal issues.36 A key issue, for example, was whether it was negligence for the defendants not to heat their blood or screen donors to avoid contamination with Hepatitis B, and whether this negligence with respect to Hepatitis B also made the defendant negligent with respect to plaintiffs in this case.37 That was an issue that different state courts might decide differently because the plaintiffs in this case were not infected with Hepatitis B, but the same precautions that would have been effective against Hepatitis B would have “serendipitously” protected them against HIV.38 So Posner argued that, to resolve the case, the trial judge would have to give “a kind of Esperanto instruction, merging the negligence standards of the 50 states and the District of Columbia,”39 thus violating the command of Erie Railroad Co v Tompkins40 for federal courts to apply state law rather than a specially created federal common law.41

Posner’s third argument was that the division of issues between the class action and later individual cases would violate the Seventh Amendment because subsequent juries, who would have to decide issues such as comparative negligence and proximate causation, would have to implicitly reexamine the class action jury’s determination of the defendant’s negligence.42

Each of these arguments has been subject to critique. The argument that certification would coerce settlements has received the most attention from law-and-economics scholars. Some scholars disputed whether liability in the Rhone-Poulenc case itself was likely to be so ruinous as to lead the defendant to settle on unfavorable terms.43 Similarly, one can argue that ruinous liability is appropriate because the case against the defendants was much stronger than Posner assumed. Posner inferred that the class action was weak, in part, from the fact that the defendants had won twelve of thirteen individual cases (92.3 percent) that had gone to judgment. The importance Posner attributed to this statistic can be gleaned from the fact that he repeated it four times.44 Yet as Professor Charles Silver pointed out, Posner’s inference from the litigated cases is itself undermined by the economic analysis of litigation on which he otherwise relied heavily.45 As Professors George Priest and Benjamin Klein famously argued, litigated cases are not a random sample of all cases.46 While Priest and Klein argued that selection effects would ordinarily mean that only close cases would be litigated (resulting in a 50 percent plaintiff trial win rate),47 they also pointed out that, when defendants have more at stake—as in Rhone-Poulenc, in which an adverse judgment could have collateral estoppel effects against the defendants (but not the plaintiffs)—defendants would selectively settle cases they were likely to lose and take their chances litigating only the cases they were confident they would win.48 Modern asymmetric-information theories of litigation likewise predict that informed defendants (like those in Rhone-Poulenc) will disproportionately settle the cases they are likely to lose and litigate only cases they think they can win.49 As a result, the defendant’s 92.3 percent trial win rate may reflect the defendant’s strategic settlement decisions rather than the underlying merit of the class action.50 This speculation is confirmed by conversations with lawyers for hemophiliac patients, who asserted that a number of cases were, in fact, settled before Posner’s decision in Rhone-Poulenc and that defendants had been strategic in settling the cases plaintiffs were more likely to win, such as those involving favorable state law, young children (who are more sympathetic), persons able to identify the transfusion lot they received (obviating issues relating to market-share liability), and those with later seroconversions (for whom the statute of limitations would present fewer difficulties).51

While these arguments may suggest that Rhone-Poulenc itself was wrongly decided, they do not undermine the correctness of the precedent set by the case. Rhone-Poulenc may provide the correct framework for other cases, even if its holding was applied improperly in Rhone-Poulenc itself. A broader critique is that the problem of coercive settlements exists only if the defendant is risk averse.52 If the defendant were risk neutral, the expected value of a class settlement and numerous individual suits would be roughly the same (differing only on account of litigation costs). In fact, the defendant would likely underpay if potential liability exceeded the defendant’s assets, as the settlement would reflect the expected value of likely payments, not the expected value of judgments it could not possibly pay. On the other hand, while diversified shareholders are plausibly risk neutral, the managers who actually make settlement decisions are likely to be risk averse, so pointing out that Posner’s argument presumes risk aversion does not really undermine its validity.

Professor Warren Schwartz mounted a broader critique, arguing that there is nothing wrong with large settlements induced by defendant risk aversion because settlements are just transfers from the defendant to the plaintiff, and there is no reason to prefer a larger or smaller transfer.53 Schwartz concedes that “the incentive effect on the behavior of the parties” could provide a basis for discouraging larger payments reflecting risk aversion, but he then asserts that “[t]here is [ ] no systematic relationship between the amount of the transfer and any of the choices made by the parties in the events out of which the dispute arises.”54 While Schwartz is correct that the magnitude of settlement cannot affect the incentives of the defendant in the particular case which is settled, he ignores the effect of settlements on the future actions of those similarly situated. If class actions mean that settlements are generally larger than expected liability, they could induce defendants to be inefficiently cautious, to charge higher prices for products, to remove some products from the market, or to reduce other activities that are, on net, beneficial to society. That is, settlements that are too high, like damages that are too high, could result in inefficiency by inducing excessive precautions or by depressing activity levels.

Luke McCloud and Professor David Rosenberg take on Judge Posner’s second argument, that class certification is improper because it would require resolution of the case under “a law that is merely an amalgam, an averaging, of the nonidentical negligence laws of 51 jurisdictions.”55 McCloud and Rosenberg argue that averaging is actually desirable because defendants who sell the same product in multiple states (as defendants in Rhone-Poulenc presumably did) must adjust their behavior and safety precautions to the law of all affected jurisdictions anyway.56 In doing so, the rational defendant will not try to satisfy the law of the most stringent jurisdiction but will take into account expected liability as averaged across all states where it might be sued. Thus, defendant actions implicitly reflect “average law,” so it is appropriate that such amalgamated, synthetic law be applied in class actions and other multistate litigation.57

The third argument, about the Seventh Amendment, has been thoroughly examined by Professor Patrick Woolley, who concluded that, contrary to Posner’s opinion:

The [Reexamination] Clause requires only that later juries respect the formal findings of the first jury. Within these broad parameters, the Clause does not prohibit later juries from independently evaluating evidence on a previously decided issue in order to decide a related issue. For that reason, the Clause allows a jury charged with deciding the issue of comparative negligence to rehear evidence presented to an earlier jury on the defendant’s negligence, provided the later jury understands that the formal findings of the earlier jury are binding.58

Most other commentators and several courts have also rejected Posner’s Seventh Amendment arguments.59

Perhaps the strongest argument in favor of Posner’s approach, albeit one that he did not explicitly make because it does not fit within the doctrinal framework, is that appellate review of certification decisions is necessary to ensure the development of a body of case law on the subject. As noted above, because of the final order doctrine, certification decisions would hardly ever reach appellate courts without expansion of mandamus or some other mechanism overriding the final order doctrine. Appellate review is necessary not only to correct errors and prevent injustice (as Posner argued) but also to generate a body of precedents that could guide district courts in the future.

Those who are familiar only with Rhone-Poulenc could get the mistaken impression that Posner was generally hostile to class actions and was interested primarily in protecting defendants. That would ignore the many cases in which Posner pushed the law to facilitate class actions. He did so primarily in controversies involving a multitude of small claims, which otherwise would never result in suit. For Posner, Rhone-Poulenc was a different and unusual case because it involved plaintiffs whose claims were large enough that each could sue individually.60 The class action device was therefore unnecessary, although it might still economize on costs.

Carnegie v Household International, Inc61 is illustrative of the many cases in which Posner upheld class certification.62 After Rhone-Poulenc, Carnegie is Judge Posner’s second most cited certification decision.63 The plaintiffs were recipients of tax refund anticipation loans who claimed that they had been fraudulently induced into taking on these high-interest debts.64 A global settlement had been negotiated and then rejected by Posner and his Seventh Circuit colleagues in the Reynolds v Beneficial National Bank65 decision, discussed below.66 The court was “concerned that the settlement might have been the product of collusion between the defendants, eager to minimize their liability, and the class lawyers, eager to maximize their fees.”67 On remand, the judge to whom the case had been assigned then certified a very similar class action, albeit with different class action lawyers, and the defendants appealed. Posner used the case as a vehicle for explaining the benefits of small-claims class actions. In response to defendants’ argument that the class would be unmanageable because it involved millions of class members, Posner asserted:

That is no argument at all. The more claimants there are, the more likely a class action is to yield substantial economies in litigation. It would hardly be an improvement to have in lieu of this single class action 17 million suits each seeking damages of $15 to $30. . . . The realistic alternative to a class action is not 17 million individual suits, but zero individual suits, as only a lunatic or a fanatic sues for $30. But a class action has to be unwieldy indeed before it can be pronounced an inferior alternative—no matter how massive the fraud or other wrongdoing that will go unpunished if class treatment is denied—to no litigation at all.68

That is, Posner was willing to tolerate an “unwieldy” and difficult-to-manage class action if the alternative was that “massive [ ] fraud or other wrongdoing” would go “unpunished.” Posner then went on to list the many “imaginative solutions” that a district court judge could employ to render the case more manageable:

(1) bifurcating liability and damage trials with the same or different juries; (2) appointing a magistrate judge or special master to preside over individual damages proceedings; (3) decertifying the class after the liability trial and providing notice to class members concerning how they may proceed to prove damages; (4) creating subclasses; or (5) altering or amending the class.69

If the class action would economize on litigation costs and allow possibly meritorious claims to be resolved, Posner encouraged the district court to be creative in administering the litigation rather than using management difficulties as an excuse not to certify.

Posner also swatted away defendants’ numerous procedural objections—that the district court judge improperly put the burden of proof regarding certification on the defendants, that certification was barred by collateral estoppel, and that the district court had made no findings on three of the four certification criteria set out in FRCP 23(a): numerosity, commonality, and typicality.70 When Posner thought justice would be served by class action treatment, he would not allow procedural technicalities to stand in the way.

The most plausible explanation for Judge Posner’s anti–class action ruling in Rhone-Poulenc and his pro–class action decision in Carnegie is the differences between the underlying facts. The high stakes in Rhone-Poulenc meant both that plaintiffs could sue individually and that liability might bankrupt the defendants, whereas the small stakes in Carnegie meant that only the class action device could punish and deter fraud. Nevertheless, it is also possible to see the cases as reflecting a change in Posner’s thinking about class actions. There is some quantitative support for that hypothesis. In his first two decades on the bench, the 1980s and 1990s, Judge Posner’s decisions overwhelmingly resulted in noncertification. Less than 20 percent (2/11) affirmed a district court’s certification decision or reversed a decision not to certify. In contrast, in the 2000s, that percentage rose to more than 40 percent (6/14), and in the 2010s, the percentage rose to 50 percent (7/14).71 Thus, over his career, the percentage of procertification decisions more than doubled. Nevertheless, such numbers need to be interpreted with caution. They would mean that Posner’s views “evolved” only if the quality of cases reaching him remained the same. An alternative hypothesis is that, because of decisions like Rhone-Poulenc, class counsel became more careful about the class actions they brought, so those that reached appeal were more likely to merit certification.

II. Settlement Approvals and Related Issues

Judge Posner was exceptional in the degree of scrutiny with which he examined class settlements and related issues, such as fees to counsel and the distribution of settlement proceeds to the class. Reynolds v Beneficial National Bank72 is illustrative of such cases and is, deservedly, Posner’s most highly cited opinion relating to the approval of class action settlements.73

As already mentioned, Reynolds involved tax refund anticipation loans that allegedly reflected self-dealing contrary to statements that might have led customers to believe that the defendants were fiduciaries.74 The lawyers for the class and the defendants entered into a $25 million settlement, which, even after modest revisions required by the district court judge, contained a number of problematic features.75 Relief was limited to $15 per loan and $30 per plaintiff, which meant that plaintiffs who took out more than two loans would receive no additional compensation for the additional loans.76 There was scant evidence to support the idea that $25 million was a reasonable amount, especially in light of the fact that one of the defendants faced up to $2 billion in exposure in a Texas class action, which the settlement would have enjoined.77 Furthermore, the lawyers in the Texas class action were forbidden from notifying members of the class about the status of the Texas litigation, thus denying the class members information that might have led them to opt out of the settlement.78

Posner began his opinion by referring to “lawyers for the class who may, in derogation of their professional and fiduciary obligations, place their pecuniary self-interest ahead of that of the class.”79 He also reminded judges to “exercise the highest degree of vigilance in scrutinizing proposed settlements of class actions,” calling the district court judge “a fiduciary of the class, who is subject therefore to the high duty of care that the law requires of fiduciaries.”80 In the middle of the opinion, Posner referred to Professor Coffee’s “reverse auction” theory, “the practice whereby the defendant in a series of class actions picks the most ineffectual class lawyers to negotiate a settlement with in the hope that the district court will approve a weak settlement that will preclude other claims against the defendant.”81 While acknowledging that there was “no proof that the settlement [in the Reynolds case] was actually collusive in the reverse-auction sense,” Posner insisted that “the circumstances demanded closer scrutiny.”82

Judges, Posner argued, should make an effort to “quantify the net expected value” of the case so that they can evaluate whether the settlement is reasonable.83 Posner proceeded to do that and suggested that the case was worth at least double the amount of the proposed settlement.84 Posner also excoriated the settlement and district court judge for including in the scope of the release two other class actions making different, unrelated claims.85 Posner concluded the opinion by considering whether lawyers who made objections at the fairness hearing are entitled to a fee for having conferred a benefit on the class. Although the issue was moot because “the claim for attorneys’ fees falls with the settlement,” Posner devoted several paragraphs to the issue “for the sake of guidance for the future.”86 Such objectors are desirable “because of the risk of collusion over attorneys’ fees and the terms of settlement generally,” so they should be incentivized through the award of attorneys’ fees.87

Altogether, the opinion stands as a stunning rebuke to a district court judge and a stern admonition to all district court judges to scrutinize settlements rigorously, with attention to conflicts of interest. District court judges are urged to expend significant effort to estimate the expected value of litigation so as to ascertain whether a settlement is, in fact, reasonable rather than a sweetheart deal that benefits primarily class lawyers (through fees) and defendants (by barring related concurrent and future litigation). The opinion also evinces distrust, if not contempt, for class action lawyers, who are presumed to act in their own interest rather than in the interest of the class.

Although Posner was often suspicious of class action lawyers, he also believed that good ones were entitled to appropriate compensation. In fact, a decision he wrote on that subject is even more heavily cited than the Reynolds decision just discussed. In the Matter of Continental Illinois Securities Litigation,88 Posner reversed and remanded a case in which the district judge had cut the lawyer’s fee request in half.89 As Posner noted, “Having employed their professional skills to create a cornucopia for the class, the lawyers for the class were entitled . . . to suitable compensation for their efforts.”90 The district court judge, Posner ruled, had arbitrarily placed a $175 ceiling on the lawyers’ hourly rates, in spite of the fact that defense counsel were paid much more and that market rates, more generally, were much higher.91 Class lawyers were also entitled to market rates for paralegals and computerized research, to prejudgment interest, and to a risk multiplier that reflected the chance that they might be paid nothing at all.92 The district court judge had “cut legal research time by 40 percent on the ground that experienced securities counsel don’t need to do much research,” an argument Posner dismissed as “clearly incorrect,” as “[n]o one carries the whole of federal securities law . . . around in his head, and a lawyer who tries to respond to a motion or brief without conducting fresh research is courting sanctions or a malpractice suit.”93 Recognizing that some judges may be unwilling or unable to conduct the calculations necessary to determine an appropriate fee, Posner suggested “appointment of a special master to advise the court.”94

In sum, Posner’s class action jurisprudence required district court judges to conduct economically informed scrutiny of the terms of settlements, including fee awards. While such scrutiny was often conducted based on suspicion that class counsel had acted in self-interested fashion, it could also be used to ensure that class counsel received market rates for work on behalf of the class.

III. Citation Analysis

The table below provides some quantitative evidence of Posner’s influence by comparing citation of Posner’s class action decisions to those of other Seventh Circuit judges during the period when Posner was a judge.

The table looks only at cases involving class action issues, such as certification, settlement approval, and attorneys’ fees. The figures for the Seventh Circuit (excluding Posner) are based on a one-third systematic sample.95 The first column shows citations in the Seventh Circuit, including cases from district courts in Illinois, Indiana, and Wisconsin. Surprisingly, Posner’s opinions are cited, on average, less often than those of other Seventh Circuit appellate judges (forty-six versus seventy times). In contrast, if one looks at citations outside the Seventh Circuit, where precedent is persuasive rather than binding, Posner’s average citations are 50 percent higher than his colleagues’ (forty versus twenty-seven). Similarly, when looking at citations in law reviews, Posner’s average is more than twice that of his colleagues. Posner’s lead in all citations—which includes citations in non–law review secondary sources and in trial briefs—is also significant, although not as dramatic. In considering the figures in the table, it is important to note that the Seventh Circuit, for most of the time Posner was on the bench, also included several other distinguished jurists, including Frank Easterbrook.

As discussed in the introduction, citation analysis provides only one measure of Posner’s influence. Other, less quantitative measures, such as his influence on the FRCP and on practitioners, may be more persuasive.

Conclusion

Posner’s class action jurisprudence had a profound effect, especially through its influence on the Federal Rules of Civil Procedure. Posner pushed the federal courts to examine certification decisions and the approval of settlements more rigorously so as to ensure that class actions served both justice and class members and to prevent both coercive and collusive settlements. He encouraged district court judges to act as fiduciaries for the class because he thought agency problems inherent in class action litigation made normal adversary procedures an inadequate safeguard for plaintiffs’ interest. Although his opinions are sometimes misinterpreted as evincing hostility toward class action lawyers and class actions generally, he recognized that class actions and the lawyers who bring them serve an important function in improving access to justice, especially when many persons are injured in ways that result in claims that are individually small, but large in aggregate.

• 1See Creative Montessori Learning Centers v Ashford Gear LLC, 662 F3d 913, 917–18 (7th Cir 2011); In the Matter of Continental Illinois Securities Litigation, 962 F2d 566, 573 (7th Cir 1992).
• 2See, for example, Reynolds v Beneficial National Bank, 288 F3d 277, 279–80 (7th Cir 2002).
• 3See, for example, CE Design Ltd v King Architectural Metals, Inc, 637 F3d 721, 723 (7th Cir 2011); In the Matter of Rhone-Poulenc Rorer Inc, 51 F3d 1293, 1297–99 (7th Cir 1995).
• 451 F3d 1293 (7th Cir 1995).
• 5Id at 1299–1303.
• 6See, for example, Butler v Sears, Roebuck & Co, 727 F3d 796, 799 (7th Cir 2013); Carnegie v Household International, Inc, 376 F3d 656, 660–62 (7th Cir 2004); In the Matter of Continental Illinois Securities Litigation, 962 F2d 566, 570–71 (7th Cir 1992).
• 7Carnegie, 376 F3d at 661.
• 8Id.
• 9See Richard A. Posner, Economic Analysis of Law 349–50 (Little, Brown 1973); Richard A. Posner, An Economic Approach to Legal Procedure and Judicial Administration, 2 J Legal Stud 399, 438–41 (1973).
• 10See generally John C. Coffee Jr, Class Wars: The Dilemma of the Mass Tort Class Action, 95 Colum L Rev 1343 (1995); Jonathan R. Macey and Geoffrey P. Miller, The Plaintiffs’ Attorney’s Role in Class Action and Derivative Litigation: Economic Analysis and Recommendations for Reform, 58 U Chi L Rev 1 (1991).
• 11See, for example, Reynolds, 288 F3d at 279, 282; Eubank v Pella Corp, 753 F3d 718, 719, 725, 728 (7th Cir 2014).
• 12See, for example, Posner, Economic Analysis of Law at 350 (cited in note 9); Posner, 2 J Legal Stud at 440–41 (cited in note 9).
• 13Perry Cooper, Posner: Class Action Rules, Constitution Overrated (Bloomberg Law, Oct 28, 2016), archived at http://perma.cc/6A8Q-ZZEN (quoting “[p]laintiffs’ attorney Benjamin Gould of Keller Rohrback LLP [on] Twitter . . . [as] saying he is enjoying ‘Posner’s late decadent period’”).
• 14Id.
• 15Id.
• 16Id.
• 17Cooper, Posner: Class Action Rules (cited in note 13).
• 18Amendments in 1987 were “technical. No substantive change [was] intended.” FRCP 23, Advisory Committee Notes to the 1987 Amendments.
• 19Robert H. Klonoff, The Decline of Class Actions, 90 Wash U L Rev 729, 733 (2013) (“A critical event leading to Rule 23(f) and CAFA occurred in 1995 when the Seventh Circuit decided In re Rhone-Poulenc Rorer Inc.”).
• 20FRCP 23(f), Advisory Committee Notes to the 1998 Amendments.
• 21See Rhone-Poulenc, 51 F3d at 1299.
• 22FRCP 23(h), Advisory Committee Notes to the 2003 Amendments. See also FRCP 23(e), (g); FRCP 23(e), (g), Advisory Committee Notes to the 2003 and 2018 Amendments.
• 23See, for example, Hearing before the Judicial Conference Advisory Committee on Civil Rules, 114th Cong, 2d Sess (Tab 1) 3, 11 (Nov 2016) (statement of DRI: The Voice of the Defense Bar) (noting that its defense bar members protect clients against “potentially ruinous, and many times frivolous, lawsuit[s],” and advocating that “[d]ecisions on class certification motions should be subject to immediate and mandatory appellate review”); id at 7 (statement of John Sweeney) (quoting Judge Posner in his criticism of cy pres awards); Hearing before the Judicial Conference Advisory Committee on Civil Rules, 115th Cong, 1st Sess 3 (Jan 2017) (statement of Public Justice) (quoting Judge Posner to begin its discussion of class actions’ deterrent value); Hearing before the Judicial Conference Advisory Committee on Civil Rules, 115th Cong, 1st Sess, (Tab 7) 4, 12, 13, 16 (Feb 2017) (statement of Competitive Enterprise Institute) (explicitly referencing Posner in its discussion on how to improve settlement processes).
• 24Columbia Law School, Posner on Class Actions, First Panel: Class Certification, at 0:07, 0:18, online at http://cdnapisec.kaltura.com/html5/html5lib/v2.53/mwEmbedFrame.php/p/1758691/uiconf_id/38497331/entry_id/0_p09uk61t?wid=_1758691&iframeembed=true&playerId=kaltura_player&entry_id=0_p09uk61t&flashvars[streamerType]=auto (visited May 11, 2019) (Perma archive unavailable).
• 25Columbia Law School, Posner on Class Actions, Second Panel: Class Settlement, Fees, Claims Administration, at 1:00, online at http://cdnapisec.kaltura.com/html5/html5lib/v2.53/mwEmbedFrame.php/p/1758691/uiconf_id/38497331/entry_id/0_qtydlwlt?wid=_1758691&iframeembed=true&playerId=kaltura_player&entry_id=0_qtydlwlt&flashvars[streamerType]=auto (visited May 11, 2019) (Perma archive unavailable).
• 26Id at 1:06–1:08.
• 27Westlaw search, citing references of Rhone-Polenc, Feb 2019 (listing 3,940 total citing references); Westlaw search, citing references of Carnegie v Household International, Inc, 376 F3d 656 (7th Cir 2004), Feb 2019 (listing 2,355 total citing references for Posner’s second most cited case).
• 28See, The Class Action Fairness Act of 2000, S Rep No 106-420,106th Cong, 2d Sess 18 (2000) (2000 CAFA Report) (statement of Sen Hatch); id at 57 (statement of Sen Leahy, et al).
• 29In the Matter of Rhone-Poulenc Inc, 51 F3d 1293, 1295–96 (7th Cir 1995).
• 30Id at 1294.
• 31437 US 463 (1978).
• 32Id at 469.
• 33Rhone-Poulenc, 51 F3d at 1295.
• 34See Daniel Klerman and Greg Reilly, Forum Selling, 89 S Cal L Rev 241, 260–61 (2016).
• 35Rhone-Poulenc, 51 F3d at 1299.
• 36Id at 1300.
• 37Id at 1296.
• 38Id.
• 39Rhone-Poulenc, 51 F3d at 1300.
• 40304 US 64 (1938).
• 41Rhone-Poulenc, 51 F3d at 1300, citing Erie, 304 US at 78–80.
• 42Rhone-Poulenc, 51 F3d at 1302–03.
• 43See, for example, Charles Silver, “We’re Scared to Death”: Class Certification and Blackmail, 78 NYU L Rev 1357, 1376–78 (2003).
• 44Rhone-Poulenc, 51 F3d at 1296, 1298–1300.
• 45See Silver 78 NYU L Rev at 1378–79 (cited in note 43).
• 46See George L. Priest and Benjamin Klein, The Selection of Disputes for Litigation, 13 J Legal Stud 1, 14–15 (1984).
• 47Id at 4–5, 17–20.
• 48Id at 26. See also generally Yoon-Ho Alex Lee and Daniel Klerman, The Priest-Klein Hypotheses: Proofs and Generality, 48 Intl Rev L & Econ 59 (2016).
• 49See Abraham Wickelgren, Law and Economics of Settlement, in Jennifer Arlen, ed, Research Handbook on the Economics of Torts 344 (Elgar 2013).
• 50The 92.3 percent win rate may, however, also reflect the underlying merit of the class action because selection effects are partial. See Daniel Klerman and Yoon-Ho Alex Lee, Inferences from Litigated Cases, 43 J Legal Stud 209, 210–11, 230–34 (2014).
• 51Conversations between the author and Roy Spece, Jan 22, 1997, and July 27, 2018. See also Jay Tidmarsh, Mass Tort Settlement Class Actions: Five Case Studies 92 (Federal Judicial Center 1998).
• 52See Silver, 78 NYU L Rev at 1374 (cited in note 43). See also Warren F. Schwartz, Long-Shot Class Actions: Toward a Normative Theory of Legal Uncertainty, 8 Legal Theory 297, 308–09 (2002).
• 53See Schwartz, 8 Legal Theory at 308–09 (cited in note 52).
• 54Id at 309.
• 55Rhone-Poulenc, 51 F3d at 1302.
• 56See Luke McCloud and David Rosenberg, A Solution to the Choice of Law Problem of Differing State Laws in Class Actions: Average Law, 79 Geo Wash L Rev 374, 386 (2011).
• 57Id.
• 58Patrick Woolley, Mass Tort Litigation and the Seventh Amendment Reexamination Clause, 83 Iowa L Rev 499, 542 (1998).
• 59See generally, for example, Simon v Philip Morris Inc, 200 FRD 21 (EDNY 2001); Douglas McNamara, Blake Boghossian, and Leila Aminpour, Reexamining the Seventh Amendment Argument against Issue Certification, 34 Pace L Rev 1041 (2014). See also, for example, Coffee, 95 Colum L Rev at 1440 (cited in note 10) (“This Seventh Amendment objection seems a weak argument.”); Elizabeth F. Cabraser, Your Products Liability Hit Parade: A Class Torts “Top 20”, 37 Tort and Ins L J 169, 197 n 110 (2001) (citing cases and articles disagreeing with Posner’s analysis).
• 60Rhone-Poulenc, 51 F3d at 1299.
• 61376 F3d 656 (7th Cir 2004).
• 62Id at 659.
• 63See note 27.
• 64Carnegie, 376 F3d at 658–59.
• 65288 F3d 277 (7th Cir 2002).
• 66See Carnegie, 376 F3d at 659 (discussing the history of the litigation and the Reynolds reversal).
• 67Id.
• 68Id at 660–61.
• 69Id at 661, quoting In re Visa Check/MasterMoney Antitrust Litigation, 280 F3d 124, 141 (2001).
• 70Carnegie, 376 F3d at 661–64.
• 71The methodology for arriving at these figures involved first identifying a sample of cases by searching for all opinions written by Posner that contained the phrase “class action.” My research assistants and I then read each of these cases to identify those decisions bearing on class certification. Finally, we calculated the total number of those cases as well as the number and percentage of those cases that ruled favorably on class certification, separating the results by decade.
• 72288 F3d 277 (2002).
• 73Westlaw search, citing references of Reynolds, Feb 2019 (listing 1,551 total citing references).
• 74Reynolds, 288 F3d at 280.
• 75Id at 281.
• 76Id at 282.
• 77Id at 283.
• 78Reynolds, 288 F3d at 284.
• 79Id at 279.
• 80Id at 279–80.
• 81Id at 282.
• 82Reynolds, 288 F3d at 283.
• 83Id at 284.
• 84Id at 285.
• 85Id at 285–86.
• 86Reynolds, 288 F3d at 288.
• 87Id.
• 88962 F2d 566 (7th Cir 1992). Westlaw search, citing references of Reynolds, Feb 2019 (listing 1,551 total citing references); Westlaw search, citing references of Continental, Feb 2019 (listing 1,926 total citing references).
• 89Continental, 962 F2d at 568, 574.
• 90Id at 568.
• 91Id.
• 92Id at 569–70.
• 93Continental, 962 F2d at 570.
• 94Id at 573.
• 95All Seventh Circuit appellate decisions using the phrase “class action” were sorted by Westlaw’s “Most Cited” function. The second, fifth, eighth, and every third opinion thereafter were examined to see if it contained resolution of a class action issue, and, if it did, it was analyzed. To calculate total citations, the citation in the one-third sample was multiplied by three.