Litigation governance: Taking accountability seriously

Columbia Law Review

Volumn 110, Issue 2, 2010, Pages 288-351

  Coffee Jr , John C ^a,b  

^a The Earth Institute at Columbia University   (United States)

^b Center on Corporate Governance   (Australia)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 77950490352     PISSN: 00101958     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (274)

1. My colleague and friend, Henry Monaghan, has long been a master at this rulebased genre. See Henry Paul Monaghan, Antisuit Injunctions and Preclusion Against Absent Nonresident Class Members, 98 Colum. L. Rev. 1148, 1153 (1998) (arguing nonresident class members should not be precluded from litigating their substantive claims despite prior class judgment purporting to bind them).
2. Inherently, a governance perspective focuses on structural relationships, particularly those by which (1) the members of the organization or entity define their rights among themselves, and (2) authorize their agents to act for them. Most of corporate governance addresses the principal/agent relationships between shareholders and managers, and corporate law typically confers on shareholders the right to elect the board of directors and to approve certain fundamental transactions (such as a merger, sale of substantially all the corporation's assets, or amendment of the certificate of incorporation). In contrast, the Federal Rules of Civil Procedure are largely silent on these structural issues, requiring only that the court approving a class action settlement make certain requisite findings.
3. See Fed. R. Civ. P. 23(e) (specifying that court must approve settlement of class action, hold hearing, and find settlement "fair, reasonable, and adequate"). The Private Securities Litigation Reform Act of 1995, discussed infra notes 17, 32-33, 99-115 and accompanying text, presumptively makes the largest stakeholder in the action the vicarious representative for all class members in the case of securities class actions. 15 U.S.C. § 78u-4(a) (3) (B)(iii) (2006). But there is no corresponding rule governing nonsecurities class actions, which constitute the majority of class actions. Instead, all is left to judicial discretion. Although few would recommend that important issues facing the class be resolved by a majority vote (both because it is difficult to identify class members prior to the case's resolution and because small claimants tend to be rationally apathetic), civil procedure law has a void at its center to the extent that it only obliquely addresses the rights and remedies of class members. This Essay focuses on a key governance issue: whether class litigation should be organized around an entrepreneurial model, as it certainly is in the United States, or a nonentrepreneurial model, as Europe seems to favor. In a nonentrepreneurial model, neither the attorneys nor the representative plaintiff will receive disproportionate benefits from the action. Assessing these alternative models in any depth requires a broader governance perspective.
4. In my judgment, the two best-known leaders of the plaintiffs' securities class action bar were William S. Lerach and Melvyn I. Weiss. Both have served or are now serving criminal sentences in federal prison for convictions in connection with concealing illegal payments to class action plaintiffs who in effect serviced their law firms as in-house clients. See Jonathan D. Glater, Class-Action Lawyer Given a 30-Month Prison Term for Hiding Kickbacks, N.Y Times, June 3, 2008, at C3 (discussing both convictions). In addition, Richard F. Scruggs, who may have been an even more famous plaintiffs' attorney for his success in winning a multibillion dollar settlement from the tobacco industry in the 1990s, was also sentenced in 2008 for attempting to bribe a Mississippi state court judge.
5. See Abha Bhattarai, Class-Action Lawyer Given 5 Years In a Bribery Case, N.Y. Times, June 28, 2008, at C3 (discussing conviction).
6. Some academics simply deny that there is an accountability problem in class actions. See Myriam Gilles &: Gary B. Friedman, Exploding the Class Action Agency Costs Myth: The Social Utility of Entrepreneurial Lawyers, 155 U. Pa. L. Rev. 103, 104-05 (2006) (describing accountability problems as mostly a "mirage"). More academics agree that there are problems, but still defend the securities class action as a necessary mechanism for enhancing shareholder monitoring or generating general deterrence.
7. See generally Symposium, The Continuing Evolution of Securities Class Actions, 2009 Wis. L. Rev. 151. In that Symposium, Professors Merritt Fox and Jill Fisch argue that the securities class action improves corporate governance, and Professor Lawrence Mitchell argues that it generates deterrence. See Jill E. Fisch, Confronting the Circularity Problem in Private Securities Litigation, 2009 Wis. L. Rev. 333, 345-48 (seeking to justify litigation remedy as compensating informed traders for losses stemming from fraudulent corporate representations);
8. Merritt B. Fox, Why Civil Liability for Disclosure Violations When Issuers Do Not Trade?, 2009 Wis. L. Rev. 297, 310-18 (arguing mandatory disclosure regimes enhance corporate efficiency);
9. Lawrence E. Mitchell, The "Innocent Shareholder": An Essay on Compensation and Deterrence in Securities Class-Action Lawsuits, 2009 Wis. L. Rev. 243, 243 (arguing securities liability provides incentives for shareholders to take responsibility for market integrity). These academics are largely responding to new critiques, made by a number of other academics, that the securities class action today yields little compensation or deterrence.
10. See, e.g., John C. Coffee, Jr., Reforming the Securities Class Action: An Essay on Deterrence and Its Implementation, 106 Colum. L. Rev. 1534, 1534 (2006) [hereinafter Coffee, Reforming the Securities Class Action] (arguing contemporary securities class actions provide minimal deterrence or compensatory benefits). This Essay certainly does not seek to curtail class actions, but to survey the range of possibilities for their reform and to consider alternative starting points for other countries seeking for the first time to address aggregate litigation.
11. The uniqueness of the American procedures for aggregate litigation is widely recognized. See Richard L. Marcus, Review Essay: Putting American Procedural Exceptionalism into a Globalized Context, 53 Am. J. Comp. L. 709, 735 (2005) (noting importance of class action in American civil procedure). The term "entrepreneurial litigation" was probably first coined by this author (or at least legal databases do not show earlier uses of the term in law reviews).
12. See John C. Coffee, Jr., The Regulation of Entrepreneurial Litigation: Balancing Fairness and Efficiency in the Large Class Action, 54 U. Chi. L. Rev. 877 (1987). In an even earlier work, I modeled the plaintiff's attorney who handles an inventory of contingent fee actions as a risk-taking entrepreneur.
13. See John C. Coffee, Jr., Understanding the Plaintiff s Attorney: The Implications of Economic Theory for Private Enforcement of Law Through Class and Derivative Actions, 86 Colum. L. Rev. 669, 701-24 (1986) (developing entrepreneurial model of contingent fee lawyering). Most who use the term entrepreneurial litigation see three elements as defining it: (1) the availability of the class action; (2) the contingent fee; and (3) the American rule under which each side bears its own legal fees. But there are actually three other important elements that facilitate entrepreneurial litigation in the United States: (1) the right to a jury trial in common law cases brought in federal court (which is guaranteed by the Seventh Amendment to the U.S. Constitution); (2) the availability of punitive damages; and (3) the "common fund" doctrine that entitles a successful plaintiff who has created a fund that benefits others to recover reasonable attorneys' fees from the fund based on common law principles of restitution and unjust enrichment,
14. see Boeing Co. v. Van Gemert, 444 U.S. 472, 478-81 (1980) (discussing common fund doctrine);
15. Trustees v. Greenough, 105 U.S. 527, 527 (1881) (acknowledging principles underlying common fund doctrine). The first two of these additional elements cause defendants to fear the risk of large recoveries, which would be unlikely in other countries where judges impose damages and do not seek to punish the defendants for misconduct in civil cases; the third element, the common fund doctrine, eliminates the need for each class member to consent to the contingent fee. Effectively, it converts the contingent fee from something that is contractually permissible to a legal right.
16. This development required not only a procedural rule (Rule 23 of the Federal Rules of Civil Procedure), but also a constitutional theory as to why one litigant could bind others who had not consented to the representation. In Phillips Petroleum Co. v. Shutts, the Supreme Court set forth the "due process minima" on such unconsented representation, identifying three requisite elements: (1) individual notice; (2) an opportunity to exclude oneself by opting out; and (3) adequate and unconflicted representation in the proceeding. 472 U.S. 797, 811-812 (1985).
17. The contingent fee predates the class action and arose first in the United States, chiefly in personal injury litigation during the late nineteenth century. For a history of its acceptance in the United States, see James Moliterno, Broad Prohibition, Thin Rationale: The "Acquisition of an Interest and Financial Assistance in Litigation" Rules, 16 Geo. J. Legal Ethics 223, 229-30 (2003). By 1876, the Supreme Court had declared that the lawfulness of contingent fees was "beyond legitimate controversy."
18. See Stanton v. Embrey, 93 U.S. 548, 556-57 (1876) (discussing validity of contingent fees). For data on the typical plaintiff's attorney's fee in class actions, see infra notes 24, 56, 59.
19. See Alyeska Pipeline Serv. Co. v. Wilderness Soc'y, 421 U.S. 240, 257 (1975) (citing cases affirming attorney's right to collect fee). The prevailing rule in the United States, known as the "American Rule," requires litigants to pay their own attorneys' fees and costs of litigation, absent special statutory provisions for fee shifting.
20. See Thomas D. Rowe, Jr., The Legal Theory of Attorney Fee Shifting: A Critical Overview, 1982 Duke LJ. 651, 652-66 (analyzing various rationales for fee shifting rules). Limited exceptions to this rule exist where one side has acted in bad faith.
21. See, e.g., Christianburg Garment Co. v. EEOC, 434 U.S. 412, 419 (1978) (discussing ability to award attorneys' fees where plaintiff proceeds in bad faith);
22. Alyeska, 421 U.S. at 258-259 (discussing exceptions to American rule). The opposing rule, that the losing side should pay the winning side's reasonable fees, is known as the "English Rule."
23. Even courts of appeals critical of the class action have recognized that "the existence of . . . negative value" claims constitutes the "most compelling rationale for finding superiority in a class action." Castano v. Am. Tobacco Co., 84 F.3d 734, 748 (5th Cir. 1996). For a definition of the term "negative value claims" as "claims in which the costs of enforcement in an individual action would exceed the expected individual recovery,"
24. see In re Inter-Op Hip Prosthesis Liab. Litig., 204 F.R.D. 330, 348 (N.D. Ohio 2001).
25. For a good and recent overview, see Christopher Hodges, The Reform of Class and Representative Actions in European Legal Systems (2008);
26. Richard Nagareda, Aggregate Litigation Across the Atlantic and the Future of American Exceptionalism, 62 Vand. L. Rev. 1, 19-37 (2009) (providing comparative analysis of recent developments in Europe regarding aggregate litigation);
27. Michael Gousgounis, Association-Driven Aggregate Litigation: Peanuts Effect and Democratization of Litigation Governance, 2-4 (May 1, 2009) (unpublished manuscript, on file with the Columbia Law Review), available at http:// ssrn.com/abstract=1469046 (providing overview of recent European developments).
28. The terms "exit" and "voice" were first coined by the Harvard economist Albert O. Hirschman in 1970 to contrast rival strategies for influencing large organizations (including governments). Albert O. Hirschman, Exit, Voice, and Loyalty 4 (1970).
29. See Samuel Issacharoff &: Geoffrey P. Miller, Will Aggregate Litigation Come to Europe?, 62 Vand. L. Rev. 179, 180-81 (2009) (arguing that need to incentivize class counsel and representative plaintiff requires adoption in some form of American innovations, such as contingency fee). Professors Issacharoff and Miller argue that European reformers are so reluctant to adopt the central elements of the American model that there is a danger "of throwing out the baby with the bathwater."
30. Id. at 191. This line between the baby and the bathwater may often lie in the eye of the beholder.
31. No sooner did the class action develop in the late 1960s than some proclaimed that it would lead to "legalized blackmail." See, e.g., Milton Handler, The Shift from Substantive to Procedural Innovations in Antitrust Suits-The Twenty-Third Annual Antitrust Review, 71 Colum. L. Rev. 1, 9 (1971). Both before and after that, an unending number of defense counsel (and some academics) regularly proclaimed securities and corporate litigation to be particularly subject to "extortion." One can trace this critique back well into the last century.
32. See Note, Extortionate Corporate Litigation: The Strike Suit, 34 Colum. L. Rev. 1308 (1934) (discussing effects of "extortionate" securities litigation). Empirical support for this claim is less in evidence. In enacting the Private Securities Litigation Reform Act of 1995, however, Congress largely accepted the diagnosis that the scales of justice had been tilted too far in the direction of plaintiffs. The Conference Report accompanying that legislation emphasized the "abusive practices committed in private securities litigation," including "routine filing of lawsuits against issuers of securities and others whenever there is a significant change in an issuer's stock price, without regard to any underlying culpability of the issuer."
33. H.R. Rep. No. 104-369, at 31 (1995) (Conf. Rep.), reprinted in 1995 U.S.C.C.A.N. 730, 730.
34. Plaintiffs' law firms today make substantial political contributions to the political officials responsible for, or with influence over, the public pension funds that typically serve as "lead plaintiffs" in securities class actions. Their goal appears to be to induce these lead plaintiffs to select them as class counsel. This practice-known as "pay-to-play"-has been much discussed and reform legislation has been recently proposed. For an overview, see James D. Cox &: Randall Thomas, Does the Plaintiff Matter? An Empirical Analysis of Lead Plaintiffs in Securities Class Actions, 106 Colum. L. Rev. 1587 (2006) [hereinafter Cox &: Thomas, Does the Plaintiff Matter?]. Professors Cox and Thomas discuss one wellknown case in which the New York State Comptroller received $200,000 in political contributions from law firm partners (and the families of partners) at two law firms in the year following their selection as co-class counsel to the New York State Comptroller, which was serving as lead plaintiff in the action. Eventually, the two law firms received a $55 million fee award. Id. at 1611-1612 Many public pension funds have internal governance structures, such as independent boards, that are designed to protect them from pay-to-play influences, and no universal criticism is here intended.
35. For one of the first and most influential statements of this thesis that corporate legal rules work only within specific institutional and cultural contexts, see Bernard Black &: Reinier Kraakman, A Self-Enforcing Model of Corporate Law, 109 Harv. L. Rev. 1911 (1996). Their starting point is that "national contexts" shape corporate law. Id. at 1920-29. Cultural norms are very much part of this national context. For a similar critique of the idea that law can be standardized successfully,
36. see Katharina Pistor, The Standardization of Law and Its Effect on Developing Economies, 50 Am. J. Comp. L. 97 (2002) (arguing that attempts at legal standardization often impede economic development).
37. A rich literature in law and economics has explored the difficulties in transplanting legal rules (generally from developed countries to "emerging market" nations). See, e.g., Daniel Berkowitz, Katharina Pistor &: Jean-Francois Richard, The Transplant Effect, 51 Am. J. Comp. L. 163, 161-67 (2003) (analyzing historical legal transplants and proposing that success of transplant depends more on process than content of legal rules). Although some transplanted legal rules do survive, these authors find that transplants generally only take when they are adapted to local conditions or when they find a receptive audience because that audience is already familiar with a similar underlying structure to the law.
38. Id. at 167-68
39. This Essay posits that this same "transplant effect" will likely also limit the acceptance of exported litigation rules, just as it has in the case of corporate law rules.
40. See Securities Exchange Act of 1934 § 21D(a) (3)(B)(iii), 15 U.S.C. § 78u-4(a) (3) (B)(iii) (2006) (creating presumption in favor of class member seeking appointment as lead plaintiff who "in the determination of the court, has the largest financial stake in the relief sought by the class"). This provision is discussed infra notes 32-33, 99-115 and accompanying text.
41. In an earlier article, I surveyed this trend toward increased opting out in securities class actions. See John C. Coffee, Jr., Accountability and Competition in Securities Class Actions: Why "Exit" Works Better Than "Voice," 30 Cardozo L. Rev. 407, 425-429 (2008) (describing trend of increased opting out by institutional investors). This Essay seeks to place this development in a larger context and does not address (as the earlier article did) the options by which courts could encourage or discourage opting out.
42. For the standard definition of "agency costs," see Michael C.Jensen &: William H. Meckling, Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure, 3 J. Fin. Econ. 305, 308 (1976). Agency costs are defined as the sum of the preventive costs taken by firms to limit misappropriations by their agents, the bonding costs undertaken by agents themselves to demonstrate their loyalty, and the irreducible minimum of losses from agent misconduct that it is not efficient to seek to prevent. For an application of agency costs to class actions,
43. see Jonathan R. Macey &: 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, 19-27 (1991).
44. This is the key insight of the Jensen and Meckling model: The founders of a firm bear the agency costs of investor uncertainty. But as next discussed, this insight has less application to litigation governance.
45. Technically, there is only a right to opt out (that is, to exclude oneself from the class so that the action's resolution is not binding on such person) in the case of class actions certified under Rule 23(b)(3). Fed. R. Civ. P. 23(c)(2)(B). Although a class certified instead under Rule 23(b)(1) or (b)(2) does not confer a right to opt out, some decisions have suggested that the right to opt out may be constitutionally required when class members have substantial monetary claims.
46. See Molski v. Gleich, 318 F.3d 937, 950-53 (9th Cir. 2003) (reversing mandatory class certification under Rule 23(b)(2) because failure to provide notice or opportunity to opt out denied due process);
47. In re Telectronics Pacing Sys., Inc., 221 F.3d 870, 873-74, 881-882 (6th Cir. 2000) (finding due process requires right to opt out of class "[w]here defendants have sufficient funds to compensate class members through individual litigation").
48. Just how powerless even the class representatives can be is revealed by Lazy Oil Co. v. Witco Corp., 166 F.3d 581, 589-91 (3d Cir. 1999) (approving class action settlement and refusing to disqualify class counsel despite objections from lead plaintiff and other major class members). There, three of the four class representatives who had brought an antitrust class action became dissatisfied with the class counsel and its proposed settlement. Id. at 583. These class representatives were sophisticated oil companies who had originally planned the action. In addition, some 384 class members also opposed the settlement and filed objections. Id. at 584. Nonetheless, the district court and the Third Circuit permitted the case to be settled over their objections.
49. Id. at 589-91.
50. The lesson here is simple: Both courts and class counsel often have strong incentives to favor settlements, even when they are on terms that are unsatisfactory to sophisticated class members. For a similar view that the representative plaintiff has little role or impact, see Geoffrey P. Miller, Competing Bids in Class Action Settlements, 31 Hofstra L. Rev. 633, 634 n.2 (2003) (citing other sources that discuss minimal role or impact of representative plaintiff). Although class members may object to the attorneys' fee awarded by the court, the fee award is a matter for judicial decision and discretion. Fed. R. Civ. P. 23(h) (governing procedures for attorney fee awards and requiring judicial approval). The removal of class counsel is also at the discretion of the court, and even the lead plaintiff in securities litigation does not have the power to remove class counsel. Fed. R. Civ. P. 23(g).
51. A debate continues over the significance of pay-to-play. A recent investigation by the New York Daily News found that lawyers representing the State of New York received over $518 million in fee awards in class action cases over the last ten years and that the last three New York State Controllers had received over $1 million in political contributions from plaintiffs' law firms. See Kenneth Lovett, Pension Pay-To-Play: Law Firms Gave Controllers Big Bucks, Then Got $518M in Fees from State Fund, N.Y. Daily News, Oct. 8, 2009. Others believe that the impact of pay-to-play has been overstated.
52. See David H. Webber, Is Pay-to-Play Driving Public Pension Fund Activism in Securities Class Actions?; An Empirical Study 2 (N.Y.U. Ctr. for Law, Econ. & Org., Working Paper No. 09-28, 2009), available at http://ssrn.com/ abstract=1432497)(on file with the Columbia Law Review). Nonetheless, when political contributions are made to state and municipal comptrollers by lawyers in a remote jurisdiction (as they often are), it is hard to imagine any plausible explanation other than a desire to influence the future selection of class counsel. 24. For the fullest empirical study demonstrating this point,
53. see Theodore Eisenberg &: Geoffrey P. Miller, Attorney Fees in Class Action Settlements: An Empirical Study, 1 J. Empirical Legal Stud. 27, 28 (2004) (arguing size of fund is best predictor of fee, regardless of fee formula used).
54. I have surveyed the range of conflicts that can arise between class counsel and class members elsewhere. See John C. Coffee, Jr., Class Action Accountability: Reconciling Exit, Voice, and Loyalty in Representative Litigation, 100 Colum. L. Rev. 370, 385-393 (2000).
55. See Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 816-23 (1999) (holding that applying Kansas law to all claims was unconstitutional because Kansas lacked interest in claims unrelated to itself and conflict of law existed between Kansas and other states).
56. See Fed. R. Civ. P. 23(b)(3). In recent years, the federal courts have tightened class certification requirements substantially, principally by interpreting this predominance standard more rigorously. For a review of these decisions,
57. see John C. Coffee, Jr. &: Daniel Wolf, Class Certification: Developments over the Last Five Years 2004-2009, 10 Class Action Litig. Rep. (BNA) (Nov. 13, 2009).
58. See Fed. R. Civ. P. 23(a)(4). The adequacy-of-representation requirement looks to both "(1) whether any substantial conflicts of interest exist between the representatives and the class; and (2) whether the representatives will adequately prosecute the action." See Busby v. JRHBW Realty, Inc., 513 F.3d 1314, 1323 (11th Cir. 2008) (reversing denial of class certification and assessing standards for certification under Rule 23).
59. Fed. R. Civ. P. 23(a)(3). Typicality is satisfied "when each class member's claim arises from the same course of events, and each class member makes similar legal arguments to prove the defendant's liability." Robinson v. Metro-North Commuter R.R., 267 F.3d 147, 155 (2d Cir. 2001) (reversing denial of class certification and analyzing standards under Rule 23).
60. For example, in tort cases, proximate causation is typically found to be an individual issue that prevents a finding of predominance; in fraud cases, the victim's reliance on the alleged misrepresentation is similarly an individual issue (except as explained in the next footnote), and in contract cases when laws of multiple states apply, this variation in law is also an individual legal issue. See Coffee &: Wolf, supra note 27, at 195-218 (reviewing recent decisions rejecting class certification for failure to satisfy predominance standard).
61. In large part, this is because the Supreme Court in Basic Inc. v. Levinson adopted the "fraud on the market" doctrine, which eliminated any need for the class members to show individual reliance on the material misstatement or omission in cases where the security traded in an efficient market. 485 U.S. 224, 250 (1988). Because reliance would have been an "individual" issue, the need to prove reliance by each class member would have precluded a finding of predominance under Rule 23(b)(3).
62. Id. at 229-30;
63. see also Fed. R. Civ. P. 23(b)(3).
64. See Private Securities Litigation Reform Act of 1995, Pub. L. No.104-67, § 101 (b), 109 Stat. 737, 743-44 (codified as amended at 15 U.S.C. § 78u-4(a) (3) (B)(iii) (2006)) (amending Section 21D of Securities Exchange Act of 1934 to create presumption in favor of person seeking appointment who "in the determination of the court, has the largest financial interest in the relief sought by the class").
65. Id. (codified as amended at 15 U.S.C. § 78u-4(a)(3)(B)(v)). This selection is, however, made expressly "subject to the approval of the court," which must satisfy itself as to the adequacy of representation standard under Rule 23(a) (4), both with respect to the class representative and the class counsel. Id.
66. By their terms, Rules 23(b)(1) and 23(b)(2) do not require either notice or a right to opt out. Fed. R. Civ. P. 23(b) (1)-(2). Rule 23(b) (1) authorizes a mandatory nonopt-out class action, usually relating to a "limited fund." Fed. R. Civ. P. 23(b)(1)(B). A limited fund is essentially a fund that will be inadequate to satisfy the claims of all claimants. Such a limited fund may arise either when the litigation is over a specific asset (for example, a Rembrandt claimed by various heirs to a decedent), or when the defendant's funds are inadequate to pay all claimants. In such instances, any individual recovery in a separate action might frustrate the class action by depleting the assets available, and hence Rule 23(b)(1)(B) authorizes a mandatory class from which class members may not opt out. In Ortiz v. Fibreboard Corp., the Supreme Court insisted upon a narrow definition of "limited fund," restricting the reach of Rule 23(b)(1)(B) to cases where the "totals of the aggregate liquidated claims and the fund available for satisfying them, set definitely at their maximums, demonstrate the inadequacy of the fund to pay all the claims." 527 U.S. 815, 838 (1999). Much more frequently used than Rule 23(b)(1) is Rule 23(b)(2), which authorizes a class action seeking predominantly equitable or injunctive relief. Fed. R. Civ. P. 23(b)(2). Traditionally, only "incidental" money damages could be sought under this provision, but the circuit courts are now clearly split over the meaning of incidental money damages, as explained infra note 35.
67. Both the Second and the Ninth Circuits have allowed class actions to seek compensatory damages under an "ad hoc balancing" test where "the positive weight or value [to the plaintiffs] of the injunctive or declaratory relief sought is predominant even though compensatory or punitive damages are also claimed." Robinson v. Metro-North Commuter R.R., 267 F.3d 147, 164 (2d Cir. 2001)
68. (quoting Allison v. Citgo Petroleum Corp., 151 F.3d 402, 430 (5th Cir. 1998) (Dennis, J., dissenting)); see also Molski v. Gleich, 318 F.3d 937, 950 (9th Cir. 2003) ("In order to determine predominance, we have focused on the language of Rule 23(b)(2) and the intent of the plaintiffs in bringing the suit."). Robinson's broad test for when compensatory relief can be obtained under Rule 23(b)(2) looks to whether "(1) even in the absence of a possible monetary recovery, reasonable plaintiffs would bring the suit to obtain the injunctive or declaratory relief sought; and (2) the injunctive or declaratory relief sought would be both reasonably necessary and appropriate were the plaintiffs to succeed on the merits." Robinson, 267 F.3d at 164. A number of other circuits have rejected this broad reading of Rule 23(b)(2) and permitted under it only the recovery of damages easily computed pursuant to a formula or statutory penalty.
69. See Murray v. Auslander, 244 F.3d 807, 812 (11th Cir. 2001) (denying class certification under Rule 23(b)(2) because nonincidental damages were sought);
70. Jefferson v. Ingersoll Int'l Inc., 195 F.3d 894, 897-98 (7th Cir. 1999) (same);
71. Allison, 151 F.3d at 415-416 (same).
72. For example, the Ninth Circuit has upheld class certification in Dukes v. Wal-Mart, Inc., 509 F.3d 1168 (9th Cir. 2007), reh'g en banc granted, 556 F.3d 919 (9th Cir. 2009), even though the panel noted that the compensatory damages sought by the plaintiffs "may amount to billions of dollars."
73. Id. at 1186.
74. Even if Dukes is reversed on rehearing, a considerable conflict remains among the circuits over the scope of Rule 23(b)(2). 37. See Hodges, supra note 10, at 119 ("The model that has been adopted in most EU jurisdictions for court-based rules is that claimants must take a positive step to assert their rights and formally join a coordinated procedure (opt-in)."). This study lists England, Sweden, Germany, and Italy as normally following this position, but notes that Portugal has adopted an opt-out procedure. See also Issacharoff &: Miller, supra note 12, at 192 (also noting that Europe has resisted the opt-out class).
75. See Kern ex rel. Estate of Kern v. Siemens Corp., 393 F.3d 120, 124-26 (2d Cir. 2004), cert, denied, 544 U.S. 1034 (2005) (reversing class certification of opt-in class consisting of heirs, beneficiaries, and personal representatives of victims who died in ski train fire in Austria where district court had limited class to those who consented to be bound by any judgment).
76. Id. at 124 ("Not only is an 'opt-in' provision not required, but substantial legal authority supports the view that by adding the 'opt-out' requirement to Rule 23 in the 1966 amendments, Congress prohibited 'opt-in' provisions by implication.");
77. see also Clark v. Universal Builders, Inc., 501 F.2d 324, 340 (7th Cir. 1974) ("[T]he requirement of an affirmative request for inclusion in the class is contrary to the express language of Rule 23(c)(2)(B) ....").
78. See Commission White Paper on Damages Actions for Breach of the EC Antitrust Rules, at 4, COM (2008) 165 final (Apr. 2, 2008) [hereinafter White Paper]. Under the European Commission's practice, a White Paper is a follow-up document to an earlier Green Paper that outlines possible options. Thus, a Green Paper is a discussion document intended to stimulate debate and elicit specific proposals, whereas a white paper is an official set of proposals that is intended to lead to legislation. An earlier Green Paper on antitrust enforcement was published by the European Commission in 2005, which was then followed in 2008 by the above-noted white paper. See Commission Green Paper on Damages Actions for Breach of the EC Antitrust Rules, COM (2005) 672 final (Dec. 19, 2005). For a general discussion of the White Paper and the earlier Green Paper, see Hodges, supra note 10, at 170-81. For a critical review of the EC's decision to prefer the opt-in model to the American opt-out version,
79. see Robert Gaudet, Jr., Turning a Blind Eye: The Commission's Rejection of Opt-Out Class Actions Overlooks Swedish, Norwegian, Danish and Dutch Experience, 30 Eur. Competition L. Rev. 107, 107 (2009) ("Because the opt-out mechanism provides the greatest compensation to the most victims. . . [it] is vastly preferable to the opt-in mechanism . . . .").
80. Commission Green Paper on Consumer Collective Redress, COM (2008) 794 final (Nov. 27, 2008).
81. Hodges finds the consumer ombudsmen to be the "most developed public system to date" in Europe. Hodges, supra note 10, at 228.
82. See Issacharoff &: Miller, supra note 12, at 192 (noting European "tendency to allow only organizations to represent consumers in class action cases").
83. The White Paper appears to contemplate that damages would be awarded to the association for injury to its members, and that individuals would sue separately on an opt-in basis. White Paper, supra note 40, at 4.
84. In a staff report accompanying the White Paper, the staff argued that associations would sue when individuals would not, because they were better prepared to accept the administrative burdens and did not need to fear alienating a business partner. See Commission Staff Working Paper Accompanying the White Paper on Damages Actions for Breach of the EC Antitrust Rules, at 19-20, COM (2008) 165 final (Apr. 2, 2008).
85. Some European nations already have this procedure. Spain permits consumer organizations to sue on behalf of both their members and unidentified victims. See Laurel J. Harbour &: Marc E. Shelley, The Emerging European Class Action: Expanding MultiParty Litigation to a Shrinking World, Prac. Litigator, July 2007, at 23, 28.
86. Hunt v. Wash. State Apple Adver. Comm'n, 432 U.S. 333, 343 (1977);
87. see also UAW v. Brock, 477 U.S. 274, 290 (1986) (reaffirming Hunt principles and noting "[t]he very forces that cause individuals to band together in an association will thus provide some guarantee that the association will work to promote their interests").
88. For example, in the earlier discussed case of Dukes v. Wal-Mart, Inc., the action was litigated by a combination of public interest firms and a traditional entrepreneurial plaintiffs' law firm. 509 F.3d 1168, 1173-1174 (9th Cir. 2007).
89. These tensions can arise even in the standard context of a school desegregation case in which the NAACP may be representing the plaintiff schoolchildren and parents. For a candid discussion of these tensions, see Derrick A. Bell, Jr., Serving Two Masters: Integration Ideals and Client Interests in School Desegregation Litigation, 85 Yale LJ. 470, 505-11 (1976) (arguing for courts to be sensitive to disagreements in black communities over the type of school relief). For the fullest assessment of how these conflicts surrounding public interest litigation might be resolved,
90. see William B. Rubenstein, Divided We Litigate: Addressing Disputes Among Group Members and Lawyers in Civil Rights Campaigns, 106 Yale LJ. 1623 (1997) [hereinafter Rubenstein, Divided We Litigate].
91. For this figure of 0.2% (or two in a thousand), see Theodore Eisenberg &: Geoffrey Miller, The Role of Opt-Outs and Objectors in Class Action Litigation: Theoretical and Empirical Issues, 57 Vand. L. Rev. 1529, 1532 (2004) [hereinafter Eisenberg &: Miller, Opt-Outs] (basing figure on sample consisting of several thousand class actions between 1993 and 2003).
92. See infra notes 127-136 and accompanying text (discussing several cases where significant proportions of qualified individuals opted in).
93. See White Paper, supra note 40, at 4 (arguing for necessity of opt-in class actions and representative actions complementing each other).
94. For example, in Dukes v. Wal-Mart, Inc., the plaintiffs in this much-noted employment discrimination case were represented by several public interest firms (The Impact Fund, Equal Rights Advocates, Public Justice Center) and by a traditional class action plaintiffs' firm, Cohen, Milstein, Hausfeld & Toll. 509 F.3d at 1173-1174 In the United States, however, there is little practical need for the representative plaintiff to possess high reputational capital because in opt-out class actions no reason exists either to solicit the entire class or to engage in other marketing activity in order to obtain universal inclusion of class members. To the extent that any marketing is done, it is largely focused on assembling a coalition of would-be lead plaintiffs with an aggregate ownership greater than that of rival coalitions.
95. These techniques are used today in cases brought under the Fair Labor Standards Act, which provides a mandatory opt-in procedure in wage and hour cases. See 29 U.S.C. § 216(b) (2006) (specifying opt-in procedure); infra notes 138-145 and accompanying text.
96. This is a less-recognized reason why institutional investors may sometimes want the position of lead plaintiff. The longer the class period is defined to be, the more vulnerable it becomes to attack by defendants. Yet, a prominent institutional investor serving as lead plaintiff may insist that the class period's definition be stretched far enough to cover most or all of its losses. Its self-interest here may be in conflict with that of other class members.
97. For one of the best treatments of the complex issues in fee award decisions and the limitations on the use of this policy lever, see John Leubsdorf, The Contingency Factor in Attorney Fee Awards, 90 Yale LJ. 473, 481 (1981). For the fullest recent judicial examination of the standards for attorney fee awards, see In re Enron Corp. Sec., Derivative & "ERISA" Litig., 586 F. Supp. 2d 732 (S.D. Tex. 2008).
98. This is particularly true in mega-class actions, which may settle for amounts over $100 million. Because U.S. courts typically award fees on a declining percentage of the recovery, a point may often be reached after which increasing the settlement's size produces less and less benefit to the plaintiffs attorney. At this point, class counsel's selfinterest will increasingly favor settlement.
99. Coupon settlements are settlements in which class members are compensated with coupons giving them discounts off the price of products or services, rather than monetary awards. For an example of their misuse, see In re Gen. Motors Corp. Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 780, 819-22 (3d Cir. 1995) (rejecting proposed settlement consisting of nontransferrable coupons entitling recipient to $1,000 discount on purchase of truck). The Class Action Fairness Act of 2005, Pub. L. No.109-2, 119 Stat. 4 (codified in scattered sections of 28 U.S.C), restricted the use of coupons in settlements by a variety of means, but chiefly by instructing courts to award attorneys' fees to class counsel based only on the coupons redeemed by the class, not the coupons distributed to the class. See 28 U.S.C. § 1712(a) (2006).
100. For an identical assessment (including the estimate that the fee award will be between ten and thirty percent), see Issacharoff &: Miller, supra note 12, at 186. During the early-to-mid 1990s, the standard fee award in securities class actions was higher and averaged thirty-two percent of the settlement. See Frederick C. Dunbar, et al., Nat'l Econ. Research Assoes., Recent Trends III: What Explains Settlements in Shareholder Class Actions 7 (1995). For a later study updating these results and also finding the average fee award to be thirty-two percent of the settlement,
101. see Denise M. Martin et al., Recent Trends IV: What Explains Filings and Settlements in Shareholder Class Actions, 5 Stan. J.L. Bus. &: Fin. 121, 141 (1999). More recently, the percentage fee has declined as settlements have soared over the billion dollar mark, reflecting the tendency for courts to follow a "declining percentage of the recovery" fee formula.
102. See In re Enron Corp., 586 F. Supp. 2d at 774-775 (collecting recent studies).
103. Social scientists have coined the term "peanuts effect" to refer to the demonstrated tendency of small claimants to exhibit risk-seeking behavior. See Bethany J. Weber &: Gretchen B. Chapman, Playing for Peanuts: Why Is Risk Seeking More Common for Low-Stakes Gambles?, 97 Organizational Behav. &: Hum. Decision Processes 31, 31-32 (2005) ("[D]ecision-makers are more willing to take risks when playing for 'peanuts.'"). On this basis, the plaintiffs attorney and negative value claimants may have interests that are often poorly aligned. For a fuller discussion of the peanuts effect and a criticism of the PSLRA's preference for the largest stakeholder in light of it, see Gousgounis, supra note 10, at 16-19.
104. For the origin of these terms, see Hirschman, supra note 11, at 3-5 (defining terms and identifying them as two mechanisms through which managers are alerted to firm's poor performance).
105. Civil procedure scholars have recurrently suggested reforms of class action practices that seek to improve or enhance the "voice" of class members. See, e.g., Patrick Woolley, Rethinking the Adequacy of Adequate Representation, 75 Tex. L. Rev. 571, 604 (1997) (proposing greater rights for class members to participate in litigation decisions).
106. Some academics have discussed expanded "exit" options as a corrective for class action abuses. See Mark C Weber, A Consent-Based Approach to Class Action Settlement: Improving Amchem Products, Inc. v. Windsor, 59 Ohio St. LJ. 1155, 1193 (1998) (proposing requirement of individual consent in class action settlements).
107. Alexandra Lahav, Fundamental Principles for Class Action Governance, 37 Ind. L. Rev. 65, 82 (2003).
108. For the view that this is an effective remedy in corporate governance, see Alex Edmans, Blockholder Trading, Market Efficiency, and Managerial Myopia, 64 J. Fin. 2481, 2482-2484 (2009).
109. Eisenberg &: Miller, Opt-Outs, supra note 50, at 1532 (studying primarily consumer class actions).
110. For data showing that retail shareholders today own only about twenty-six percent of the stock of public corporations, see J. Choper, J. Coffee &: R. Gilson, Cases and Materials on Corporations 15 (7th ed. 2008).
111. See James D. Cox &: Randall S. Thomas, Letting Billions Slip Through Your Fingers: Empirical Evidence and Legal Implications of the Failure of Financial Institutions to Participate in Securities Class Action Settlements, 58 Stan. L. Rev. 411, 413 (2005) [hereinafter Cox &: Thomas, Slip Through Your Fingers] ("[L] ess than thirty percent of institutional investors with provable losses perfect their claims in [securities class action] settlements.");
112. see also James D. Cox &: Randall S. Thomas, Leaving Money on the Table: Do Institutional Investors Fail to File Claims in Securities Class Actions?, 80 Wash. U. L.Q. 855, 879 (2002) [hereinafter Cox &: Thomas, Leaving Money] (discussing plausible reasons for institutional investors failure to file claims).
113. According to the National Economic Research Associates, a consulting firm, the ratio of securities class action settlements to investor economic losses in 2002, 2003, and 2004 was 2.7%, 2.9%, and 2.6%, respectively. See Elaine Buckberg, et al., Nat'l Econ. Research Assoes., Recent Trends in Shareholder Class Action Litigation: Are WorldCom and Enron the New Standard? 6 (2005), available at http://www.nera.com/ Publication.asp?p-ID=2544&login= 6923124 (on file with the Columbia Law Review). The highest percentage they found was 7.2% in 1996. Id. This ratio appears to be decreasing as the size of the potential damages increases, possibly because defendants are insolvency constrained.
114. See Kevin LaCroix, Oakbridge Ins. Servs., Opt-Outs: A Worrisome Trend in Securities Class Action Litigation, Insights, Apr. 2007, at 1, available at http://www. oakbridgeins.com/newsletters/April-Opt- OutsAWorrisomeTrendinSecuritiesClassAction Litigation.pdf (on file with the Columbia Law Review) [hereinafter LaCroix, Worrisome Trend] (discussing Lerach's ability to persuade investors to opt out of class settlements).
115. Id.
116. Three large California pension funds-CalPERS, CalSTRS, and the Los Angeles County Employee Retirement System-announced in 2005 that they had settled their actions for $257.4 million. Id. at 3. Five New York City pension funds similarly opted out and settled their $130 million in claims for $78.9 million, and their counsel announced that this settlement amounted to "three times more than they would have recovered if they had joined the class." Id. For similar estimates, see Dave Lenckus, Individual Suits Likely Over Subprime Losses; Some Investors Expected to Opt-Out of Class Actions, Business Ins., Nov. 19, 2007 (describing payouts for investors who opted out of settlements with AOL and WorldCom).
117. See Josh Gerstein, Investors Opt Out of Time Warner Class Action Suit, N.Y. Sun, Feb. 8, 2006, at 5, available at LexisNexis, NYSUN file (providing reasons for investors' decision to opt out of settlements proceedings).
118. See Gilbert Chan, CalPERS' Time Strategy Pays Off: The State Pension Fund Gets $117.7 Million After Opting Out of Class Action Against Media Giant, Sacramento Bee, Mar. 15, 2007, available at LexisNexis, MCTBUS file (describing settlement received by opting out of class action); LaCroix, Worrisome Trend, supra note 70, at 2. Christopher Patti, an in-house counsel for the University, provided the "16 to 24 times" estimate. Chan, supra.
119. See Time Warner Settles Lawsuit for $144 Million, L.A. Times, Mar. 8, 2007, at C6 (quoting Ohio Attorney General Marc Dann that Ohio received $135 million more than it would have received under the settlement). Ohio received a total of $175 million, because its legal fees of $31 million were paid by Time Warner in addition to the $144 million it received. Assuming that Ohio would have received only $9 million under the class action (i.e., $144 million minus $135 million), this translates into better than a 19:1 disparity.
120. CalPERS's General Counsel described its settlement as "approximately 17 times what we would have recovered if we stayed in the class," estimating that it would have received roughly $6 million under the class action. See Chan, supra note 74 (quoting Peter Mixon, CalPERS general counsel).
121. CalSTRS estimated that it would receive only $15 million under the class action. Id.
122. Josh Gerstein, Time Warner Case Finds a Surprise, N.Y. Sun, Dec. 7, 2006, at 1, available at LexisNexis, NYSUN file [hereinafter Gerstein, Surprise]; LaCroix, Worrisome Trend, supra note 70, at 2.
123. For example, the State of Alaska settled its $60 million claim for $50 million and said that it had received "50 times more than what we would have received if we had remained in the class." Gerstein, Surprise, supra note 78, at 1;
124. see also LaCroix, Worrisome Trend, supra note 70, at 2. CalPERS claimed losses of approximately $129 million, so its settlement for $117.7 million represents a seeming ninety percent recovery rate.
125. For this estimate of the average recovery in securities class actions, see supra note 69.
126. Conceivably, if small claimants did not file, the institutional investors that did file could receive 100% or more of their claims. For studies finding that even large claimants often fail to file significant claims in a securities class action settlement, see Cox &: Thomas, Slip Through Your Fingers, supra note 68, at 412;
127. see also Cox &: Thomas, Leaving Money, supra note 68, at 855, 870-78. Formerly, class action settlements often contained a reversionary provision under which unclaimed amounts in the settlement fund reverted to the defendant, but such reversionary settlements are now rare in securities class actions because they have been discouraged by the PSLRA. See infra note 151. As a result, the beneficiaries today of a low claims filing rate are the larger institutions that do file claims. 82. The stipulation of settlement was filed on November 23, 2005. See In re Qwest Commc'ns Int'l Inc. Sec. Litig., No. Ol-cv-01451-REB-CBS, 2006 U.S. Dist. LEXIS 71267, at *8 (D. Colo. Sept. 26, 2006).
128. See Robert Elder, Retired Teachers System, Qwest Reach Settlement, Austin Am. Statesman, Dec. 6, 2007, at D-1, available at LexisNexis, AUSTIN file (noting amount received under class action settlement versus that received by opting out);
129. see also Kevin LaCroix, Opt-Outs, Claims Severity and D&O Insurance Limits, D&O Diary, Feb. 5, 2007, at http://www.dandodiary.com/2007/02/ articles/d-o-insurance/optouts-claims-severity-and-d-o-insurance-limits/ [hereinafter LaCroix, Claims Severity] (on file with the Columbia Law Review) (discussing how class action settlement was only one of several settlements Qwest has reached). In November, 2007, the Alaska Attorney General issued a press release announcing that its state's various funds had settled their $89 million in claims in the Qwest litigation for a net recovery (after attorneys' fees) of $19 million and that this recovery contrasted with a payment of only $427,000 that they would have received under the class settlement. See Press Release, Alaska Dep't of Law, Department of Law Announces $19 Million Settlement in Securities Fraud Claims Against Qwest Communications (Nov. 21, 2007) (on file with the Columbia Law Review). This is roughly a 45:1 ratio. Similarly, the Teachers Retirement System of Texas announced a $61.6 million net recovery and contrasted it with a $1.4 million payment that it would have received under the class action settlement. Elder, supra, at D-1. This would be better than a 40:1 ratio. CalSTRS settled for $46.5 million and announced that it was receiving "about 30 times more than it would have recovered if it had taken part in the class action." LaCroix, Claims Severity, supra.
130. Stuart Grant, Grant &: Eisenhofer P.A., Comments as Panelist at American Bar Association's Litigation Section Conference (May 2, 2009) (responding to question posed by author who was moderator of securities litigation panel). Grant &: Eisenhofer P.A. has represented many institutional opt-outs in securities litigation.
131. If the opt-out plaintiff can obtain a trial date prior to the resolution of the class action, it can virtually compel the defendants to settle with it on favorable terms. This is because a favorable verdict at trial in the individual action may make it difficult or impossible for the defendants to settle the class action on a cheaper per-claimant basis; in contrast, a favorable settlement with the individual opt-out can be kept confidential and so will not disrupt the class settlement. This fear that a successful plaintiffs' verdict in the optout action could disrupt the class settlement has led federal courts to enjoin such an early trial under the All Writs Act. See In re Baldwin-United Corp., 770 F.2d 328, 335 (2d Cir. 1985) (permitting state trial to be enjoined under All Writs Act where it would disrupt settlement negotiations before district court). More recently, the Second Circuit has marginally curbed this power.
132. See Ret. Sys. v. J.P. Morgan Chase &: Co., 386 F.3d 419, 429-431 (2d Cir. 2004) (reversing injunction granted by federal district court against state trial court where settlement was not clearly imminent).
133. By opting out, a state or municipal pension fund may be able to bring suit in state court in its home jurisdiction and thereby escape the jurisdiction of the Judicial Panel on Multidistrict Litigation to assign all related cases to one federal district. See infra notes 94-95 and accompanying text. In the past, once so assigned, cases rarely escaped. A state court may be a more attractive forum as local jurors may empathize more with the local plaintiffs (for example, the local teacher or police pension fund), and plaintiffs' attorneys may be able to elicit resentment against a distant corporate defendant.
134. The provisions of the PSLRA relating to pleading, discovery, and fee shifting are set forth in Section 21D ("Private Securities Litigation") of the Securities Exchange Act of 1934, 15 U.S.C. § 78u-4 (2006). Section 21D applies only to a "private action . . . that is brought as a plaintiff class action pursuant to the Federal Rules of Civil Procedure." Id. § 78u-4(a)(1). The principal exception to this generalization that opting out evades the reach of federal courts' ability to halt or deny discovery is that the Securities Litigation Uniform Standards Act (SLUSA) does authorize a federal court to stay discovery in a state court "upon a proper showing," even in a nonclass action. Section 21D(b)(3)(D) of the Securities Exchange Act of 1934 authorizes a federal court as follows: "Upon a proper showing, a court may stay discovery proceedings in any private action in a State court, as necessary in aid of its jurisdiction, or to protect or effectuate its judgments, in an action subject to a stay of discovery pursuant to this paragraph."
135. Id. § 78u-4(b)(3)(D). This provision is less automatic than the PSLRA's stay, which requires no special showing.
136. See, e.g., Oscar Private Equity Invs. v. Allegiance Telecom, Inc., 487 F.3d 261, 266-70 (5th Cir. 2007) (requiring plaintiff to establish loss causation at class certification stage by preponderance of all admissible evidence);
137. Regents of the Univ. of Cal. v. Credit Suisse First Boston (USA), Inc., 482 F.3d 372, 385-90 (5th Cir. 2007) (reversing class certification for failure to satisfy prerequisites of "fraud on the market" doctrine and adequate deceptive act);
138. In re Initial Pub. Offering Sec. Litig., 471 F.3d 24, 27 (2d Cir. 2006) (rejecting "some showing" standard and requiring district judge to make determination that each Rule 23 requirement is met).
139. Useful examples are provided by two major antitrust class actions brought against Archer-Daniels-Midland Co. (ADM). In 1997, following its criminal conviction for price fixing, ADM reached a civil settlement with class action plaintiffs in a suit alleging price fixing in the citric acid market under which ADM paid $86 million, but some of the largest purchasers-including Proctor &: Gamble, Quaker Oats, and the Kraft Foods unit of Philip Morris Co.-opted out. See P&G, Quaker Oats, Unit of Philip Morris Sue Archer-Daniels, Wall St. J., June 11, 1997, at B15. A year later, ADM was forced to reveal that it had paid $36 million to four of these opt-out firms. One careful study has estimated that the optouts received "from 2.0 to 3.5 times more damages (per dollar of citric acid purchased) than did the settling class." See John M. Connor, Archer Daniels Midland: Price-Fixer to the World 73 (Purdue Univ. Dept. of Agric. Econ., Staff Paper 00-11, 4th ed., 2000), available at http://ageconsearch.umn.edu/bitstream/28664/1/sp00-11.pdf (on file with the Columbia Law Review).
140. Earlier, ADM had also been independently sued for price fixing in the lysine acid market (lysine acid being an amino acid used by food companies and feedlots to promote quick growth in hogs and chickens). It settled the class action for $45 million (after paying a $70 million criminal fine to the Department of Justice), but again some twenty-five large customers opted out. See Thomas M. Burton, Several ADM Clients Won't Participate in a Proposed Price-Fixing Settlement, Wall St. J., July 10, 1996, at B2. While the actual amounts paid to these opt-outs is not known, one study has estimated that they would have received "about $20 million"-again outperforming class members. See Connor, supra at 73.
141. See Carlos Tejada, Some Plaintiffs Quit Settlement in Vitamin Case, Wall St. J., Feb. 18, 2000, at BlO. For a discussion of the opt-out procedures in this case, see In re Vitamins Antitrust Class Actions, 327 F.3d 1207, 1208-09 (D.C. Cir. 2003);
142. see also In re Vitamins Antitrust Class Actions, 215 F.3d 26, 28 (D.C. Cir. 2000) (discussing additional case background).
143. See Tejada, supra note 90, at B10.
144. See Scott Hensley, Wyeth Says 90,000 Opt Out of 'Fen-Phen' Pact, Wall St. J., May 16, 2003, at B4.
145. The special problems of future claimants probably best explains why the Supreme Court rejected the class settlement in Amchem Prods. Inc. v. Windsor, 521 U.S. 591, 598, 601 (1997) ("Settlement talks thus concentrated on devising an administrative scheme for disposition of asbestos claims not yet in litigation."). It has today made mass tort cases largely uncertifiable. Future claimants are inherently apathetic because each has only a low base expectancy rate of suffering injury in the indeterminate future.
146. The JPML was created in 1968, roughly the same time as the modern class action rules were formalized. See Multidistrict Litigation Act, Pub. L. No. 90-296, § 1, 82 Stat. 109, 109 (1968) (codified as amended at 28 U.S.C. § 1407 (2006)).
147. The impact of the JPML referral was diminished by the Supreme Court in Lexecon v. Milberg Weiss Bershad Hynes & Lerach, which held that, after discovery was completed, the case had to be transferred back to the original court in which it was filed. 523 U.S. 26, 28 (1998).
148. Today, this danger that the first team to settle wins has been largely mitigated by the JPML, which moves all the competing actions to the same court and thereby discourages races to settlement. But the danger of a "reverse auction" under which defendants force plaintiffs' attorneys to compete on the basis of which plaintiff's team will offer or accept the cheapest settlement still arises when overlapping class actions are pending in state and federal court.
149. For the rare case in which a federal court has awoken to the dangers of such a reverse auction, see Reynolds v. Beneficial Nat'l Bank, 288 F.3d 277, 282-83 (7th Cir. 2002) (defining reverse auction).
150. During prior decades, the average fee award in securities class actions was around thirty-two percent of the recovery.
151. See supra note 59. Hence, the plaintiff's attorneys knew that they had little chance of receiving a forty percent award.
152. For a detailed examination of one such episode (involving an antitrust class action), see John C. Coffee, Jr., Rescuing the Private Attorney General: Why the Model of the Lawyer as Bounty Hunter Is Not Working, 42 Md. L. Rev. 215, 252-61 (1983)
153. [hereinafter Coffee, Rescuing].
154. For discussions of patronage systems in which committee positions and leadership roles in the class action were awarded in return for support in the election of lead counsel, see id. at 256-59.
155. The SEC expressed its opposition to a large number of shareholders sharing the position of lead plaintiff in In re Baan Co. Sec. Litig., 186 F.R.D. 214, 216-17 (D.D.C. 1999) (noting SEC position that "'a court generally should only approve a group that is small enough to be capable of effectively managing the litigation and the lawyers' " (quoting Memorandum of the SEC as Amicus Curiae, at 16-17, In re Baan Co., 186 F.R.D. 214 (Civ. No. 98-2465))). Since that time, few law firms have attempted to assemble more than five or six shareholders to serve collectively as lead plaintiffs.
156. Stephen J. Choi &: Robert B. Thompson, Securities Litigation and Its Lawyers: Changes During the First Decade After the PSLRA, 106 Colum. L. Rev. 1489, 1520-21 (2006).
157. Id. at 1521 tbl.4.
158. Id. at 1529.
159. See Cox & Thomas, Does the Plaintiff Matter?, supra note 14, at 1611-12 (describing incident).
160. For a fuller sense of the total amount of these pay-to-play payments, see Lovett, supra note 23 (finding law firms representing New York State to have received over $518 million in fee awards in such cases over past ten years).
161. For an overview of this scandal, which resulted in the indictment of Hank Morris, a political consultant, and employees of the New York State Comptroller's office, see Mike Mclntire, Pension Inquiry Reveals a Power Broker's Web, N.Y. Times, May 14, 2009, at Al;
162. Chris Smith, Follow the Pension Money, N.Y. Mag., Aug. 3, 2009, at 18. These scandals have not involved plaintiffs' law firms, whose political contributions are fully disclosed, but they do show a norm of reciprocity at work: Those who want business must contribute.
163. Choi & Thompson, supra note 100, at 1515 tbl.3.
164. Cox &: Thomas, Does the Plaintiff Matter?, supra note 14, at 1627.
165. The mean settlement size did increase from $9,734,000 to $15,728,000, but the median settlement size rose only from $5,500,000 to $5,745,000.
166. Id. at 1624. Both these shifts may be largely the consequence of inflation and the increased size of the market capitalization of defendant corporations in the post-PSLRA period. Cox and Thomas find that overall settlement size did not change in any statistically significant way.
167. Id. at 1628-29.
168. Id. at 1624. Explaining this finding is probably the fact that institutional plaintiffs tend to serve in larger cases against larger defendants.
169. Id. at 1625-26.
170. Id. at 1627.
171. Id. at 1623.
172. Id. at 1628-29.
173. Id. at 1629.
174. Id. at 1637.
175. Id. at 1629. These largest cases may also be the ones with the highest public profile, reflecting the fact that public pension funds often have politically prominent leaders who may enjoy, and benefit from, the attention and publicity in such cases.
176. See Issacharoff &: Miller, supra note 12, at 197 (" Entrepreneurial attorneys continue to dominate securities class action practice today as they did in prior years.").
177. Once, federal courts could and did enjoin state court trials involving opt-outs where the federal court believed that the state trial might disrupt a pending settlement of the class action in federal court. See, e.g., In re Baldwin-United Corp., 770 F.2d 328, 335 (2d Cir. 1985) (discussing authority for issuance of injunction). More recently, this broad power has been curtailed.
178. See Ret. Sys. v. J.P. Morgan Chase & Co., 386 F.3d 419, 430-31 (2d Cir. 2004) (rejecting claim that All Writs Act authorizes district court to enjoin state court trial merely because state action would delay scheduled trial in federal court).
179. See Teamsters Local 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 198 (2d Cir. 2008) (finding predominance requirement not satisfied in case involving debt securities). The Second Circuit has also found that the market for initial public offerings (IPOs) is generally not efficient and so individual reliance may also need to be established in this context. This would again encourage institutional investors to turn to individual actions rather than seek class certification.
180. See In re Initial Public Offering Sec. Litig., 471 F.3d 24, 42-44 (2d Cir. 2006) ("In the first place, the market for IPO shares is not efficient.").
181. For representative critiques of the "one size fits all" fallacy, see Steven M. Davidoff, Regulating Listings in a Global Market, 86 N.C. L. Rev. 89, 97, 160 (2007) (criticizing fallacy of "one size fits all");
182. Gerard Hertig, On-Going Board Reforms: One Size Fits All and Regulatory Capture, 21 Oxford Rev. Econ. Pol. 269, 269-70 (2005) (criticizing tendency of European countries to follow United States on a "me too" basis).
183. Professors Issacharoff and Miller, two of the most astute scholars of class actions, fall into this camp. See Issacharoff &: Miller, supra note 12, at 180-81 (applauding current European reforms partly based on "American experience" and "simple economics of incentive systems").
184. Professor Richard Nagareda, also a respected scholar, seems to fall into this camp.
185. See Nagareda, supra note 10, at 2 (noting widespread view that U.S. litigation rules "generate a considerable and undesirable drag on the U.S. economy").
186. Professor Rubenstein has described a similar model as an "expertise model." See Rubenstein, Divided We Litigate, supra note 49, at 1663-1668 This Essay will not adopt that term, because Rubenstein's primary concern is with ideological and political divisions within public interest litigation, which are less likely to arise in consumer, antitrust, securities, and environmental litigation. These latter areas are those in which Europe wishes to modestly encourage aggregate litigation. Unlike Rubenstein's proposal, which allows those with expertise to censor insurgents, the model here proposed intends greater competition.
187. See supra note 37.
188. See Issacharoff & Miller, supra note 12, at 203-206
189. Id. at 203-204
190. Id. at 206-207
191. See Eisenberg &: Miller, Opt-Outs, supra note 50, at 1532.
192. See Henrik Lindblom, The Globalization of Class Actions (Dec. 6, 2007) (unpublished manuscript, on file with the Columbia Law Review), available at http:// www.law.stanford.edu/display/images/dynamic/events-media/Sweden- National-Report. pdf.
193. Id. at 21-22 (discussing Bo Åberg v. Elfeterios Kefales, Stockholm Dist. Ct., T 3515 (2003), transferred, Nacka Dist. Ct., T 1281 (2004), also known as the "Air Olympic" case).
194. Id. at 22-33 (discussing Grupptalan mot Skandia v. Försä kringsaktiebolaget Skandia, Stockholm Dist. Ct., T 97 (2004)).
195. Id. at 22.
196. Id. at 23-24 (discussing The Consumer Ombudsman v. Kraftkommission i Sverige AB, Umel Dist. Ct., T 5416 (2004)).
197. Id. at 23.
198. Id. at 25 (discussing Carl de Geer et al. v. The Swedish Airports &: Air Navigation Service, Nacka Dist. Ct, M 1931 (2007)).
199. Id.
200. See Matthias E. Storme &: Evelyne Terryn, Belgium Report on Class Actions 6 (Dec. 13, 2007) (unpublished manuscript, on file with the Columbia Law Review), available at http://www.law.stanford.edu/display/images/dynamic/ events-media/Belgium- National-Report.pdf.
201. Of these five thousand shareholders that did opt in, some two thousand were members of the consumer organization Test-Aankoop, and the rest were clients of the consulting firm Deminor.
202. The consumer organization had greater success in convincing its members to join the action because its members did not have to bear legal fees, which were advanced by their organization. In contrast, the clients of Deminor were required "to advance a limited sum to cover the costs of the proceedings." Id. Thus, although the overall opt-in rate was slightly over forty-five percent (i.e., 5,000 out of 11,000), only about one-third of the clients of Deminor ultimately authorized the firm to represent them in the action. Id. This disparity suggests that any obligation to advance legal costs will deter opt-ins even more than the tendency of small claimants to be rationally apathetic. It also underscores the obvious utility of using consumer organizations in the role of lead plaintiff in jurisdictions that forbid contingency fees, at least to the extent that such organizations are willing to bear all or most of the litigation's expenses.
203. See Issacharoff &: Miller, supra note 12, at 204-205
204. (describing workplace litigation as "specialized context" where "class members have many opportunities to communicate with one another about the litigation"). But see infra note 147 (noting optin rate in workplace litigation could be unrepresentatively low because of deterrent effect of employer-employee relationship).
205. See 29 U.S.C. § 216(b) (2006) (specifying opt-in requirement for FLSA).
206. See id. § 626(b) (similar opt-in requirement for ADEA).
207. See id. § 216(d) (mandating that opt-in procedure for FLSA also apply to Equal Pay Act).
208. Thomas Willging et al., Fed. Judicial Ctr., Empirical Study of Class Actions in Four Federal District Courts 54 (1996), available at http://www.fjc.gov/public/pdf.nsf/ lookup/rule23.pdf/$File/rule23.pdf (on file with the Columbia Law Review).
209. See Andrew C. Brunsden, Hybrid class actions, dual certification, and wage law enforcement in the federal courts, 29 Berkeley J. Emp. &: Lab. L. 269, 294 (2008) (discussing low opt-in rate of 15.71% in FLSA cases).
210. Id. at 294 n.129.
211. Id. at 293 (displaying opt-in rate of 89.5% in Lindsay v. Gov't Employees Ins. Co., 448 F.3d 416, 425 n.12 (D.C. Cir. 2006)).
212. Attorneys at the Texas law firm of Starzyk &: Associates, P.C., which specializes in FLSA cases, inform me that they use a website for each FLSA action to alert employees of a defendant company and that they send emails to all employees in the relevant job descriptions that are the subject of the FLSA litigation. They also alerted me to one recent California class action in which 256 out of the 364 persons eligible to participate in a "wage and hour" settlement elected to do so; this represents a 70.33% participation rate. See Declaration of Monica Balderrama in Support of Motion for Final Approval of Class Action Settlement at 16, Winzelberg v. Liberty Mutual Co., No.CV 07-460 GAF (CD. CaI. Sept. 15, 2008) (on file with the Columbia Law Review). In this light, historic participation rates in the fifteen to twenty-three percent range may understate the potential for the opt-in class when internet solicitation techniques are utilized.
213. Id.; see also Issacharoff &: Miller, supra note 12, at 204-205 (discussing specialized context of workplace litigation and factors contributing to increased participation rates).
214. This claim that potential class members in workplace litigation will be more prepared to sue, either because they know each other or because they share some common identity, overlooks the fact that in workplace litigation plaintiffs are typically suing their employer. Few will be deterred from suing a manufacturer of a defective product with whom they have no contact, but many more will be cautious about suing their employer. Hence, the opt-in rate in workplace litigation could even be unrepresentatively low.
215. Cox &: Thomas, Slip Through Your Fingers, supra note 68, at 413.
216. Extremely low participation rates have been reported in the case of coupon settlements, where the rate appears to fall below one percent. See, e.g., James Tharin & Brian Blockovich, Coupons and the Class Action Fairness Act, 18 Geo. J. Legal Ethics 1443, 1448 (2005). For recent examples, see Yeagley v. Wells Fargo & Co., No.C 05-03403, 2008 U.S. Dist. LEXIS 5040, at *5 (N.D. Cal. Jan. 18, 2008) (finding that less than one percent of class responded to settlement notice); Figueroa v. Sharper Image Corp., 517 F. Supp. 2d. 1292, 1327 (S.D. Fla. 2007) (rejecting proposed settlement, in part because "very low numbers of class members" expressed interest in $19 coupon, which was principal component of proposed settlement).
217. Of course, the most obvious hypothesis is that class counsel in opt-out class actions are not motivated to pay for, or accept the delay incident to, the use of more costly notice procedures that will elicit higher participation rates. Because some federal circuits base the fee award on the settlement size and not on the portion of the settlement actually paid out, the plaintiffs' attorney in these circuits has little economic incentive to encourage class members to file claims. See Masters v. Wilhelmina Model Agency, Inc., 473 F.3d 423, 437 (2d Cir. 2007) (favoring fee award based on total settlement fund). Alternatively, institutional investors may have little interest in expending funds to attract retail investors to participate and thus share the recovery with them. The lead plaintiff may thus have a conflict with smaller class members. I have discussed the issue of participation rates with the professional "claims administrators" of class action settlements (including the largest of these, which is the Garden City Group, Inc.) and have been advised that participation rates are highly variable, often falling below ten percent, but can reach much higher percentages when individual class members will receive a cash payment of several hundred dollars or more. Nonpecuniary relief, such as a coupon settlement, elicits much lower levels of participation.
218. Where the settlement contains no reversion provision, all of the settlement goes to those class members who do file. Reverter clauses entitling defendants to unclaimed settlement funds were once common. Cf. Geoffrey P. Miller &: Lori S. Singer, Nonpecuniary Class Action Settlements, Law &: Contemp. Probs., Autumn 1997, at 97, 106 (describing reverter funds settlements, including Boeing Co. v. Van Gemert, 444 U.S. 472 (1980), a "well-known reverter fund case"). But they are today disfavored because sophisticated courts have come to recognize that reverter clauses produce illusory settlements. Such clauses were discouraged by the PSLRA, which requires that the fee award to the class attorneys be based only on damages "actually paid to the class." See Securities Exchange Act of 1934 § 21D(a)(6), 15 U.S.C. § 78u-4(a)(6) (2006). This reversed (but only for securities class actions) the prior rule of Boeing Co., 444 U.S. at 481-482, which permitted attorney fees to be paid based on the total funds made potentially available.
219. See Securities Exchange Act of 1934 § 21D(c)(3), 15 U.S.C. § 78U-4(c)(3) (creating "presumption in favor of attorneys' fees and costs" in certain circumstances). In particular, section 21D(c)(3)(A)(ii) makes fee shifting presumptive for any "substantial failure of any complaint to comply with any requirement of Rule 11(b)." Because plaintiffs, not defendants, draft the complaint, the risk of such a sanction falls more heavily on plaintiffs.
220. Section 21D(c) (2) expressly states that the court can shift fees and impose them on either the party or the party's attorney for failure to satisfy its pleading standard. See id. § 78u-4(c)(2).
221. Indeed, because directors and officers insurers typically insure the litigation costs of corporate defendants and their officers and directors, it would make obvious economic sense for them also to insure the plaintiffs against the risk that these costs would be shifted to them under a "loser pays" rule. Economically, the same insurer might want to insure both sides, thereby receiving premiums from both sides for the same risk. A single insurer might also increase the pressure to keep litigation costs down to reasonable levels.
222. See supra note 3 and accompanying text.
223. England and Wales have allowed the "conditional fee," which allows the attorney to waive attorney fees if the action is unsuccessful and permits an enhanced fee if the action is successful. But this fee is not based on a percentage of the recovery and is typically much more modest than American class action fee awards. This change was largely motivated by a desire to finance personal injury litigation, not class actions.
224. For an overview, see Richard L. Abel, An American Hamburger stand in St. Paul's cathedral: Replacing legal aid with conditional fees in english personal injury litigation, 51 DePaul L. Rev. 253 (2001).
225. Sweden also permits "risk agreements" between the representative plaintiff and the class counsel. In one recent case, this agreement provided that class counsel would receive twice its usual hourly fee if successful and only one half that fee if the case lost.
226. Lindblom, supra note 127, at 21-22
227. This is a far milder enhancement or reduction than is common in the United States, where the fee is fully contingent on success and may sometimes exceed thirty percent of the recovery. See supra note 59 and accompanying text.
228. Cf. Issacharoff &: Miller, supra note 12, at 197-202 (highlighting usefulness of contingent fee system).
229. This author tends to agree that this criticism has often been overstated, but also believes that there is a case for insulating the class attorney, as the key professional, from the conflicts of interest that arise when the attorney finances the litigation.
230. Under Australian law, an attorney may receive a normal fee plus an agreed uplift, which is a percentage of its normal fee that is contingent on winning the case. Recent legislation has capped this uplift at twenty-five percent in most Australian jurisdictions.
231. See Stuart Clark &: Christina Harris, The Push to Reform Class Action Procedure in Australia: Evolution or Revolution?, 32 Melb. U. L. Rev. 775, 789-90 (2008)
232. (" [L] egislative caps have now been introduced on the allowable uplift, with most Australian states imposing a maximum uplift of 25 per cent of legal costs (excluding disbursements).").
233. For overviews of this trend, see Susan Lorde Martin, Litigation Financing: Another Subprime Industry That Has a Place in the United States Market, 53 Vill. L. Rev. 83, 107-12 (2008);
234. Virginia Marsh, Australian Company Pioneers Approach, Fin. Times, Jan. 5, 2007, at 3;
235. Nikki Tait, Lawyers Test Litigation Funding Waters, Fin. Times, Jan. 5, 2007, at 3.
236. See Martin, supra note 160, at 107-108 (describing Australia's largest litigation financing company). The largest, IMF, went public in 2001.
237. Id. at 107.
238. Id. For the comparable percentage fees charged by U.S. plaintiffs' attorneys in securities cases, see Martin et al., supra note 59, at 141 (finding thirty-two percent to be average fee award during 1990s).
239. Martin, supra note 160, at 109. Professor Martin estimates that the IMF, the largest Australian litigation funding firm, earned A$8 million between 2002 and 2006, or a roughly seven percent return per annum on its capital. Id. at 108.
240. For a discussion of U.S. litigation funding practices involving third-party lenders, see Julia H. McLaughlin, Litigation Funding: Charting a Legal and Ethical Course, 31 Vt. L. Rev. 615, 617-624 (2007)
241. (describing U.S. litigation funding companies' practices). The major litigation funding firms in the United States entered into a settlement with the New York Attorney General in 2005, which established a number of protections for the consumer/litigant.
242. Id. at 654-55; see also Martin, supra note 160, at 109 (noting in United States, "litigation funders agree not to make any decisions about the lawsuits").
243. See Martin, supra note 160, at 109 (noting in Australia "litigation funders provide the same services as insurance companies," who typically make key litigation decisions for defendants).
244. Id.
245. Id. at 112-113
246. Id. at 113.
247. Id.
248. The leading Australian decision is Campbells Cash &: Carry Pty. Ltd. v. Fostif Pty. Ltd. (2006) 229 A.L.R. 58.
249. See Lee Aitken, Before the High Court-'Litigation Lending' After Fostif. An Advance in Consumer Protection, or a License to 'Bottomfeeders'?, 28 Sydney L. Rev. 171, 171 (2006)
250. (describing potential problems with allowing litigation funding). In the United Kingdom, the Civil Justice Council, an advisory public body, has concluded that litigation funding provides a necessary role in assuring access to justice. See Martin, supra note 160, at 112 ("In the UK, the Civil Justice Council concluded in April 2007 that litigation funding played an important role in facilitating access to justice . . . ."). The Supreme Court of South Africa in 2004 upheld a litigation funding agreement against the claim that it was contrary to public policy.
251. See Price Waterhouse Coopers, Inc. v. Nat'l Potato Coop. Ltd. 2004 (6) SA 66 (SCA) at 79-80 (S. Afr.); see also Martin, supra note 160, at 113-114 (analyzing decision).
252. Professor Christopher Hodges of Oxford University advises that insurance companies in Europe are prepared to finance litigation costs on a "before the event" basis for both plaintiffs and defendants, and he views insurers as the most likely candidates to become third party funders of litigation in Europe.
253. Hodges, supra note 10, at 227-228
254. Some argue that this is a desirable structure, because the self-interest of the agents motivates them to watch each other. See Bernard S. Black, Agents Watching Agents: The Promise of Institutional Investor Voice, 39 UCLA L. Rev. 811, 817-818 (1992) (assessing potential for oversight of corporate managers by institutional money managers).
255. For representative arguments, see Coffee, Reforming the Securities Class Action, supra note 4, at 1536,1547-56 (discussing deterrence rationale for securities class actions); Mitchell, supra note 4, at 246-248 (same).
256. When the tide began to turn against broad class certification in the mid-1990s, one of the most influential decisions arguing against liberal certification standards was written by Judge Richard Posner. See In re Rhone-Poulenc Rorer Inc., 51 F.3d 1293, 1299, 1304 (7th Cir. 1995) (granting mandamus order to decertify multibillion dollar class action because of intense settlement pressures and danger that single jury verdict could destroy industry). He stressed that class certification in a mass tort case placed defendants "under intense pressure to settle," even when the litigation merits seemed weak. Id. at 1298. He added, however, that this pressure had to be "balanced against the undoubted benefits of the class action." Id. at 1299. Presumably, he was referring chiefly to the problem of "negative value" claims. This Essay agrees that a balance needs to be struck, and neither the opt-in nor the opt-out class action inherently achieves such a balance.
257. For a fuller discussion of this circularity problem, see Coffee, Reforming the Securities Class Action, supra note 4, at 1556-1561
258. The term "private attorney general" was coined by Judge Jerome Frank in Associated Industries of New York State, Inc. v. Ickes, 134 F.2d 694, 704 (2d Cir. 1943) ("Such persons, so authorized, are, so to speak, private Attorney Generals.").
259. For a historical and contemporary overview of the term's use and misuse, see William B. Rubenstein, On What a "Private Attorney General" Is-And Why It Matters, 57 Vand. L. Rev. 2129 (2004).
260. Despite the relatively recent origin of the term, the concept of private enforcers using civil litigation to supplement public enforcement dates back to the early days of the Republic, and the qui tam action (in which an enforcer who has not suffered an injury sues to recover primarily on behalf of the government, who will in turn compensate it with a portion of the recovery) can be traced back to the fourteenth century in England. For a review of the historical antecedents to the modern "private attorney general"
261. concept, see Trevor W. Morrison, Private Attorneys General and the First Amendment, 103 Mich. L. Rev. 589, 599-602 (2003). The "bounty hunter" of the nineteenth century American West is but one example of the concept, as is the treble damages action authorized by the federal antitrust laws. Few parallels, if any, to the private attorney general exist in Europe.
262. Some American legal scholars, however, do reject the private attorney general concept on the ground that it has not been legislatively authorized (in most cases) and so represents a usurpation of legislative power. See Martin H. Redish, Class Actions and the Democratic Difficulty: Rethinking the Intersection of Private Litigation and Public Goals, 2003 U. Chi. Legal F. 71, 81-83 (objecting to absence of legislative authority for "bounty hunter" model of private enforcement). This Essay does not take this position, but acknowledges the connection between one's constitutional theory and one's view of the private attorney general.
263. Professor Mark Tushnet has distinguished two models of constitutionalism that have evolved over the last century: (1) a legislative model of parliamentary supremacy; and (2) the U.S. model of "constrained parliamentarianism," in which a written constitution limits both the legislature and the majority. Mark Tushnet, New Forms of Judicial Review and the Persistence of Rights- and Democracy-Based Worries, 38 Wake Forest L. Rev. 813, 813-14 (2003). The private attorney general fits much less comfortably within the former system (in part for the reasons stressed by Professor Redish).
264. See Redish, supra note 178, at 84-93 (discussing ways in which class actions conflict with "precepts of democratic theory").
265. This appears to have already happened to a degree in Australia, the United Kingdom, and Sweden.
266. See supra notes 157, 159 and accompanying text. But the riskadjusted contingent fee in these countries is not based on a percentage of the recovery and hence is far more modest than the typical fee award in U.S. securities litigation.
267. See Martin et al., supra note 59, at 141 (finding average fee award in securities class actions to be thirty-two percent of recovery during 1990s).
268. Historically, public interest organizations, such as the NAACP Legal Defense Fund, did formulate long-term litigation strategies, sometimes fashioning plans that took a decade or more to implement. See Rubenstein, Divided We Litigate, supra note 49, at 1627-44 (discussing "disputes among group members and lawyers in civil rights campaigns");
269. see also Mark V Tushnet, The NAACP's Legal Strategy Against Segregated Education 1925-1950, at 138-166 (1987) (noting legal tactics NAACP lawyers used in complex litigation).
270. The plaintiffs' attorney in a class action is a form of "gatekeeper" on whom dispersed class members must rely for services that they cannot perform or closely monitor themselves. Such gatekeepers are often relied upon in the financial markets and they essentially function as "reputational intermediaries" who pledge a reputational capital that they acquire over many years and many clients to assure their clients of their loyal performance in an individual case. The classic example would be the auditing firm, for whom the gain in auditing fees from acquiescing in its client's fraud cannot normally exceed the loss in reputational capital and future business. See John C. Coffee, Jr., Gatekeepers: The Professions and Corporate Governance 2-5 (2006) (explaining role of gatekeepers in capital markets). Thus, the bottom line assertion here is that a representative plaintiff with greater reputational capital than the typical lead plaintiff would make a superior gatekeeper.
271. See Rubenstein, Divided We Litigate, supra note 49, at 1663-1664 (noting danger of "elitist subversion of democratic equality" and "infringement on individual liberty"). The fear here is largely that the representative group, for political or ideological reasons, may decline to represent legally meritorious claims or may refuse to assert legally valid, but politically objectionable, legal theories, thus leaving potential class members without effective representation.
272. In this sense, this model does not realize the goals that Professor Rubenstein favors, because his "expertise model" allows the experts to screen out and censor legal arguments that the experts consider either frivolous or premature. Instead, under the variant proposed here, all that would be required is that one of potentially multiple representative groups would be willing to adopt the particular legal theory.
273. This could make litigation funding more expensive, as the third-party funder might demand a larger percentage of the proceeds for being denied control rights. The goal is not to restrict the funder's profit, but its control, as control would typically remain with the nonprofit organization.
274. Institutional shareholders face a number of conflicts that often make them reluctant to serve. Sometimes, for example, they hold stock in both the bankrupt corporation which is the subject of the litigation and other defendants. Thus, their recovery may be offset by the reduction in value of other stocks in their portfolio. Professors Cox and Thomas provide an example of such a case in the WorldCom litigation (where Citigroup was the principal defendant). See Cox & Thomas, Does the Plaintiff Matter?, supra note 14, at 1600-1601 (discussing frustrations experienced by those with continuing ownership).