"Sweetheart" and "blackmail" settlements in class actions: Reality and remedy

Notre Dame Law Review

Volumn 75, Issue 4, 2000, Pages 1377-1403

  Hay, Bruce L ^a   Rosenberg, David ^a  

^a HARVARD LAW SCHOOL   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0034405886     PISSN: 07453515     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (60)

1. As Judge Henry Friendly observed long before class settlements reached today's astronomical proportions, the attorney has "every incentive to accept a settlement that runs into high six figures or more regardless of how strong the claims for much larger amounts may be. . . . [A] juicy bird in the hand is worth more than the vision of a much larger one in the bush." Alleghany Corp. v. Kirby, 333 F.2d 327, 347 (2d Cir. 1964). More recently, the Supreme Court has expressed a like concern. See Ortiz v. Fibreboard Corp., 119 S. Ct. 2295, 2318 & n.30 (1999).
2. See, e.g., John C. Coffee Jr., Class Wars: The Dilemma of the Mass Tort Class Action, 95 COLUM. L. REV. 1343 (1995); Susan P. Koniak, Feasting While the Widows Weep: Georgine v. Amchem Products, Inc., 80 CORNELL L. REV. 1045 (1995); Susan P. Koniak & George M. Cohen, Under Cloak of Settlement, 82 VA. L. REV. 1051 (1996).
3. See, e.g., John C. Coffee Jr., Class Wars: The Dilemma of the Mass Tort Class Action, 95 COLUM. L. REV. 1343 (1995); Susan P. Koniak, Feasting While the Widows Weep: Georgine v. Amchem Products, Inc., 80 CORNELL L. REV. 1045 (1995); Susan P. Koniak & George M. Cohen, Under Cloak of Settlement, 82 VA. L. REV. 1051 (1996).
4. See, e.g., John C. Coffee Jr., Class Wars: The Dilemma of the Mass Tort Class Action, 95 COLUM. L. REV. 1343 (1995); Susan P. Koniak, Feasting While the Widows Weep: Georgine v. Amchem Products, Inc., 80 CORNELL L. REV. 1045 (1995); Susan P. Koniak & George M. Cohen, Under Cloak of Settlement, 82 VA. L. REV. 1051 (1996).
5. This concern has also been around for a long time. See HENRY J. FRIENDLY, FEDERAL JURISDICTION: A GENERAL VIEW 120 (1973) (referring to the possible use of class actions to extract "blackmail settlements"); Milton Handler, The Shift from Substantive to Procedural Innovations in Antitrust Law - The Twenty Third Annual Anti-Trust Review, 71 COLUM. L. REV. 1, 9 (1971) (calling certain class actions "legalized blackmail").
6. This concern has also been around for a long time. See HENRY J. FRIENDLY, FEDERAL JURISDICTION: A GENERAL VIEW 120 (1973) (referring to the possible use of class actions to extract "blackmail settlements"); Milton Handler, The Shift from Substantive to Procedural Innovations in Antitrust Law - The Twenty Third Annual Anti-Trust Review, 71 COLUM. L. REV. 1, 9 (1971) (calling certain class actions "legalized blackmail").
7. The blackmail effect of class actions involves "forcing . . . defendants," as Judge Richard Posner explained, "to stake their companies on the outcome of a single jury trial, or be forced by fear of the risk of bankruptcy to settle even if they have no legal liability." In re Rhone-Poulenc Rorer Inc., 51 F.3d 1293, 1299 (7th Cir. 1995).
8. See, e.g., Sheila Birnbaum, Class Certification - The Exception, Not the Rule, 41 N.Y.L. SCH. L. REV. 347, 350 (1997); Mary J. Davis, Toward the Proper Role for Mass Tort Class Actions, 77 OR. L. REV. 157, 187 (1998); Peter A. Drucker, Class Certification and Mass Torts: Are "Immature" Tort Claims Appropriate for Class Action Treatment?, 29 SETON HALL L. REV. 213, 228 (1998); Barry F. McNeil & Beth L. Fancsali, Mass Torts and Class Actions: Facing Increasing Scrutiny, 167 F.R.D. 483, 489-90 (1996).
9. See, e.g., Sheila Birnbaum, Class Certification - The Exception, Not the Rule, 41 N.Y.L. SCH. L. REV. 347, 350 (1997); Mary J. Davis, Toward the Proper Role for Mass Tort Class Actions, 77 OR. L. REV. 157, 187 (1998); Peter A. Drucker, Class Certification and Mass Torts: Are "Immature" Tort Claims Appropriate for Class Action Treatment?, 29 SETON HALL L. REV. 213, 228 (1998); Barry F. McNeil & Beth L. Fancsali, Mass Torts and Class Actions: Facing Increasing Scrutiny, 167 F.R.D. 483, 489-90 (1996).
10. See, e.g., Sheila Birnbaum, Class Certification - The Exception, Not the Rule, 41 N.Y.L. SCH. L. REV. 347, 350 (1997); Mary J. Davis, Toward the Proper Role for Mass Tort Class Actions, 77 OR. L. REV. 157, 187 (1998); Peter A. Drucker, Class Certification and Mass Torts: Are "Immature" Tort Claims Appropriate for Class Action Treatment?, 29 SETON HALL L. REV. 213, 228 (1998); Barry F. McNeil & Beth L. Fancsali, Mass Torts and Class Actions: Facing Increasing Scrutiny, 167 F.R.D. 483, 489-90 (1996).
11. See, e.g., Sheila Birnbaum, Class Certification - The Exception, Not the Rule, 41 N.Y.L. SCH. L. REV. 347, 350 (1997); Mary J. Davis, Toward the Proper Role for Mass Tort Class Actions, 77 OR. L. REV. 157, 187 (1998); Peter A. Drucker, Class Certification and Mass Torts: Are "Immature" Tort Claims Appropriate for Class Action Treatment?, 29 SETON HALL L. REV. 213, 228 (1998); Barry F. McNeil & Beth L. Fancsali, Mass Torts and Class Actions: Facing Increasing Scrutiny, 167 F.R.D. 483, 489-90 (1996).
12. Thus, most of the critics cited in supra notes 2 and 5 have made various arguments for greater adherence to, or tightening of, the prerequisites to class certification. We do not review the details of their proposals here. For our purposes it is enough to observe that the concern over collusive and blackmail settlements has led many to urge greater restraint in the use of class actions. For a prominent example in the mass tort area, see Coffee, supra note 2 (urging less use of class actions in mass tort cases). Also, we note that these critics generally ground their concerns on anecdotes, not systematic empirical studies, and they give little or no attention to the risk of sweetheart and blackmail settlements in the conventional process of separate actions.
13. By economies of scale, we refer to the idea that the cost of litigating a claim goes down if it is bundled with other similar claims. This occurs because the investment on a common issue - that is, an issued shared by the claims - can be used to litigate all of the claims. So for example, in a product liability case, once the defendant has invested in a scientific study on the nature of a product, that study can be used for all claims concerning the product. Suppose that the study costs five million dollars. If there are 1000 claims, the per-claim cost of the study is claims is $5000; if there are 100,000 claims, the per-claim cost of the study is $50. All else being equal, the more claims there are, the more worthwhile the study becomes. Accordingly, all else being equal, the more claims there are, the more the defendant will invest in the case.
14. This is an instance of the familiar problem of collective action. The plaintiffs as a group would be benefited by a substantial investment in the litigation, but no individual plaintiff (or his lawyer) finds it in his interest to make the investment. See infra text accompanying note 18.
15. That is, it frustrates the law's general goals of deterrence and compensation.
16. This point is implicit in the widely held view that class treatment should be available in the case of "small claims" that are not economically viable on their own. See Amchem Products, Inc. v. Windsor, 521 U.S. 591 (1997). The policy at the very core of the class action mechanism is to overcome the problem that small recoveries do not provide the incentive for any individual to bring a solo action prosecuting his or her rights. A class action solves this problem by aggregating the relatively paltry potential recoveries into something worth someone's (usually an attorney's) labor. Id. at 617 (quoting Mace v. Van Ru Credit Corp., 109 F.3d 338, 344 (7th Cir. 1997)). However, the point also applies to cases involving "large claims" that can be brought separately. Even in such large claims, the defendant will outspend the plaintiffs on common issues, unless the claims are aggregated in a class action.
17. As an additional protection, we also propose that courts relax the preclusive effect of settlements that turns out to be insufficient to cover future claims.
18. We accept for purposes of this argument that defendant firms are risk averse in their litigation decisions, which underlies the concern for blackmail settlements. The validity of the assumption has not, to our knowledge, been demonstrated empirically. 13 As we describe below, the court might take the "average" of the verdicts and compute damages accordingly. (Thus, if the court held 10 trials on liability, and these resulted in two verdicts for the plaintiff and eight for the defendant, the defendant would pay 20% of the plaintiffs' damages.) Alternatively, it might base its judgment on the majority of verdicts: for example, if more than half of the verdicts favor the plaintiffs, they collect their full damages; otherwise, they collect nothing.
19. The class device may also eliminate wasteful redundancy of expenditures, but we consider this of secondary importance.
20. Sometimes a single lawyer will be hired by multiple plaintiffs in the case. Unless he is hired by all of them (a practical impossibility), however, he will not come close to matching the defendant's investment.
21. Indeed, it will not even be one million dollars, assuming that the lawyer works on a contingent fee that gives him only a fraction of the recovery.
22. Here we assume the lawyer is given a fractional share of the recovery, usually in the 20 to 30% range. Of course, the court may calculate counsel's fee on some other basis, such as an hourly rate or a more complex percentage-of-the-recovery basis (e.g., 10% of the first $20 million, 15% of the next $20 million, etc.), which might improve counsel's incentives to match the defendant's investment. Frequently, however, a simple percentage-of-the-recovery formula is used.
23. In addition, once the opt-out period has passed, class action plaintiffs cannot threaten to exit the collective effort. In contrast, under voluntary joinder, the constant risk of exit may hamper coordination.
24. Even if the concerns were valid, there would remain the empirical question whether the class action procedure created benefits in excess of its costs. We do not take that issue up here.
25. See, e.g., Coffee, supra note 2 at 1347-48; Koniak, supra note 2, at 1055-56; Koniak & Cohen, supra note 2, at 1053-57.
26. To some extent, the court depends for this information on class counsel, who according to this argument would have an incentive to understate the value of the claims in order to make the settlement look generous. For example, class counsel and the defendant might implicitly agree to understate the number of future claim-ants in a mass exposure tort case.
27. See, e.g., Todd W. Latz, Note, Who Can Tell the Futures? Protecting Settlement Class Action Members Without Notice, 85 VA. L. REV. 531 (1999); Alex Raskolnikov, Note, Is There a Future for Future Claimants After Amchem Products, Inc. v. Windsor ?, 107 YALE L.J. 2545 (1998).
28. See, e.g., Todd W. Latz, Note, Who Can Tell the Futures? Protecting Settlement Class Action Members Without Notice, 85 VA. L. REV. 531 (1999); Alex Raskolnikov, Note, Is There a Future for Future Claimants After Amchem Products, Inc. v. Windsor ?, 107 YALE L.J. 2545 (1998).
29. See, e.g., Coffee, supra note 2, at 1373; Richard A. Nagareda, Turning from Tort to Administration, 94 MICH. L. REV. 899, 950-60 (1996); Brian Wolfman & Alan B. Morrison, Representing the Unrepresented in Class Actions Seeking Monetary Relief, 71 N.Y.U. L. REV. 439, 452 (1996).
30. See, e.g., Coffee, supra note 2, at 1373; Richard A. Nagareda, Turning from Tort to Administration, 94 MICH. L. REV. 899, 950-60 (1996); Brian Wolfman & Alan B. Morrison, Representing the Unrepresented in Class Actions Seeking Monetary Relief, 71 N.Y.U. L. REV. 439, 452 (1996).
31. See Coffee, supra note 2, at 1370-73; see also Nagareda, supra note 23, at 960.
32. That is, on issues common to all of the claims, such as whether the product in question was defective or whether the defendant was negligent.
33. Here the argument assumes that there is no collateral estoppel or other preclusive effect that would bind the defendant in other plaintiffs' cases.
34. In re Rhone-Poulenc Rorer, Inc., 51 F.3d 1293, 1299 (7th Cir. 1995).
35. See id. For reasons discussed in supra Part I, this is a non-sequitur; the plaintiffs' losses may reflect a weak case or may reflect the fact that they were outspent by the defendant. We express no view on the merits of the case.
36. Similar considerations were invoked by the court of appeals to decertify a class in the Castano cigarette litigation. See Castano v. American Tobacco Co., 84 F.3d 737, 746 (5th Cir. 1996) (arguing that class certification "dramatically affects the stakes for defendants," "magnifies and strengthens the number of unmeritorious claims," "makes it more likely [that a jury will find a defendant liable]," and "creates insurmountable pressure on defendants to settle," leading to the risk of "judicial blackmail").
37. Further analysis of the appropriate fee structure is developed in Brace Hay, Asymmetric Rewards: Why Class Actions (May) Settle for Too Little, 48 HASTINGS LJ. 479 (1997).
38. We drop these assumptions below, and they are not essential to our analysis.
39. For example, if his allowed share of a settlement is 20%, then he will perhaps be willing to accept a settlement offer of $50 million awarded to the class, because such a settlement offer would give him $10 million and thus make him as well off as he would be by going to trial.
40. In other words, the appropriate regulation assures that the share of recovery from settlement does not substantially exceed the forgone share of recovery from trial (essentially the "opportunity cost" of ettling).
41. To be sure, the court may compensate counsel for these costs by calculating his fee on a time-plus-expenses basis. Frequently, however, courts use a percentage-of-the-recovery approach, either explicitly or implicitly. Our assumption in this analysis is that the court is using a percentage-of-the-recovery approach.
42. In these cases, the opportunity cost of settling represents the counsel's share of recovery from trial less the costs of going to trial.
43. Five percent represents counsel's real (or net) return from trial after deducting costs and thus the share of trial recovery counsel forgoes to settle.
44. We assume here that the plaintiffs face a single, common defendant.
45. More precisely, he represents plaintiffs whose claims add up to one-tenth of the aggregate value of the plaintiffs' claims as a whole.
46. This example assumes that the attorney will collect one-third of the recovery on each of his clients' claims if another lawyer negotiates a class settlement. That is, it assumes that his share of his clients' recoveries is the same as another lawyer who negotiates a class-wide settlement, as it would be if there were no class settlement at all. The analysis must be modified if this assumption is not true. In particular, suppose that if another lawyer negotiates a class-wide settlement, the original lawyer's share of his clients' recoveries is cut from one-third to some smaller amount. Then that smaller amount must be used as the baseline for calculating the lawyer's fee in the event he were to negotiate a class-wide settlement. Thus, for example, suppose that if another lawyer were to negotiate a class-wide settlement, the original lawyer's share of his clients' recoveries would be cut from one-third to one-sixth. Then his share of the class's claims drops from one-thirtieth to one-sixtieth. To ensure that the lawyer does not undercut his rival's offer and agree to a settlement of less than $100 million, the court would have to cap the lawyer's fee at one-sixtieth of the class settlement. When courts restrict fees of non-class counsel, as they often do, don't they reduce incentives of all counsel to invest in developing the recovery-value of claims in the separate action process? Shouldn't we note that this seeming noblesse oblige is really antithetical to plaintiffs' interests?
47. See generally Peter H. Schuck, Mass Torts: An Institutional Evolutionist Perspective, 80 CORNELL L. REV. 941, 959 (1995).
48. Indeed, these attorneys have a more pronounced interest, because they will claim a share of the "future" claims and therefore a related share of the recovery from class settlement.
49. The defendant could be given the option of replenishing the fund to pay these remaining claims, which would then be barred from further litigation.
50. Note, also, that risk-averse defendants could purchase insurance to cover the possibility that future claims will arise in numbers far above the predicted average.
51. The court may use this together with contingent res judicata, if the case involves future claims.
52. More precisely: a risk-neutral person is indifferent between a definite loss of one dollar and a one in one hundred chance of losing $100; he therefore would pay no more than one dollar to avoid taking that chance. In contrast, a risk-averse person would pay more than one dollar - perhaps much more - to avoid a one in one hundred chance of losing $100.
53. As noted previously, settlement class actions do not pose the risk of "blackmail" against defendants, because the alternative to class settlement of the class claim is not one but a series of separate action trials of the underlying classable claims.
54. There is some reason to doubt that defendants in class actions view the prospect of trial with substantial risk aversion. Generally, class actions do not involve aggregate damages of high magnitude relative to the wealth of defendant firms. Some defendants, of course, are designed to become judgment-proof and bankrupt in the event of class action. But most are organized and operated to respond in a relatively risk-neutral fashion to all sorts of major business uncertainties, including those posed by civil litigation that often arises from the firm's generalized risk-taking and thus involves multiple claims of a classable nature. Many firms are owned by stockholders with widely diversified investment portfolios and limited liability for managerial wrongs; the firm's structure - parent-subsidiary and otherwise - also may confine the impact of liability; and compensation and indemnity arrangements address the risk aversion of executives and other employees. Liability insurance further spreads the risk of a litigation catastrophe.
55. There is little reason to suppose that class counsel and class members are less risk-averse than defendant firms.
56. In the separate action process, trial is an all-or-nothing event for the plaintiff; a loss may be catastrophic. But it is not an all-or-nothing event for the defendant firm; a loss means paying only one plaintiff.
57. We conjecture that this variance-reduction is desirable on grounds of individual justice, quite apart from its effect of reducing blackmail settlements. We do not pursue that point here. For discussion, see Bruce Hay and David Rosenberg, The Individual Justice of Averaging (March 1997) (unpublished manuscript on file with the authors).
58. The discounting method could be used to tailor the aggregate award according to a series of special verdicts on causation, differences in governing law, and other elements and variables affecting the true value of the class claim.
59. See U.S. CONST. amend. VII ("No fact tried by a jury shall be otherwise reexamined in any court of the United States."). The question whether the Seventh Amendment limits the power of federal courts to hold a series of trials in this fashion is beyond the scope of this Article.
60. In response to the formal restrictions of the Seventh Amendment, courts could subdivide the class or certify a number of smaller class actions, and appoint the same team of class counsel to represent all subclasses or class actions. This process would achieve the effect of providing multiple trials while at the same time preserving the scale economies of a single-trial class action. It should also be noted that even if holding a series of trials in this fashion were not permitted, it might nonetheless be permissible to hold a single trial with several juries simultaneously impaneled to render judgment, with the court taking the average verdict to decide the case.