Indiana Law Review


Fundamental Principles for
Class Action Governance

Alexandra Lahav*

Table OF Contents

Introduction 65

L Problems in Class Action Governance 75

A. A Misapplication ofthe Traditional Adjudication Model 75
B. The Simultaneous Expansion and Disintegration ofthe

Class Action 76

C. Barriers to Participation 79

D. The Limits ofJudicial Policing 90

11. Proposed Models of Class Action Governance: The Last
Twenty Years 92

A. Market Solutions 92

1

.

Attorneys' Fees 93

2. Opt Outs 95

B. Democratic Solutions 99

1

.

Representation 101

2. Deliberation 103

C. The Judicial Administration Model 1 06
1

.

The Entity Model 106
2. Relying on Judges: Justice Breyer and Judge Weinstein .... 110

in. A Principled Approach to Governance 115
A. Strong Disclosure Requirements 118

B. An Actively Adversarial Process 125
C Expertise ofDecisionmakers and Agents 131
D. Independence ofDecisionmakersfrom Influence

and Self-Interest 133
Conclusion 140

Introduction

Class actions face a crisis of governance. The form of governance provided

by Rule 23, governance by representative parties, is both vague in theory and

ignored in practice.' Instead, by a combination of procedural rules, judicial

interpretation and common practice, the class is governed by attorneys with
limited judicial oversight. This regime neither reflects the basic insight that the

* Teaching Fellow, Stanford Law School. J.D., Harvard Law School. Thank you to
Barbara Babcock, Amanda Frost, Pamela Karlan, Mark Kelman, Nick Lundgren, Martha Minow,
Henry Monaghan, Alan Morrison, and Allan Stein for valuable comments on previous drafts ofthis

article. I am also grateful to Janet Alexander, Richard Freer, Barbara Fried, Tom Grey, Joseph
Grundfest, Deborah Hensler and Paul Lomio for their time and insights.

1. See Fed. R. Crv. P. 23(a)(4) (requiring representative parties who will "fairly and

adequately protect the interests of the class" as a prerequisite to bringing a class action). My
discussion of Rule 23 addresses the Rule as amended in December 2003.



66 INDIANA LAW REVIEW [Vol. 37:65

class and attorney do not have a traditional attorney-client relationship nor

performs the task of transforming the inchoate collectivity of the class into an

organization that protects and is responsive to the interests of class members.^

This Article proposes an alternative regime of governance for 23(b)(3) small

claims class actions^ that accomplishes both these things, based on four

fundamental principles: mandatory disclosure ofmaterial information, an actively

2. As a recent RAND study observed, procedural rules "provide only a weak bulwark
against self-dealing and collusion." Deborah R. Hensler et al.. Class Action Dilemmas:

Pursuing Public Goals for Private Gain 1 20 (RAND 2000).
3. This Article focuses on small claims class actions, sometimes referred to as "negative

value claims" because litigating the claim will result in a loss to the claim-holder. These are the

quintessential class actions and encompass mostly consumer class actions. I define these as claims

with a predicted individual award of under $10,000. This Article does not address mass tort class

actions, or aggregative litigation, where there are often substantial money damages sufficient to

justify individual suits as well as large differentials in potential recovery, punitive damages class

actions, or non-opt out class actions brought under Rules 23(b)(1) and 23(b)(2), although the

principles articulated here may also be applicable to these cases. I have chosen to address the

quintessential class action, which, according to the Supreme Court and other commentators, is the

least problematic. Consumer class actions constitute one third ofthe class actions brought, and thus

are a significant presence in the class action world. See Hensler ET AL., supra note 2, at 49-1 23

(describing that in 1995-96, class actions were mostly damages 23(b)(3) class actions, not civil

rights or social reform, and estimating that one third of class actions against businesses are

consumer class actions not including securities class actions). Many scholars consider consumer

damage class actions to be prototypical. See Jules Coleman & Charles Silver, Justice in
Settlements, 4 PHIL. &LAW 1 02, 1 22 (Autumn 1 986) (describing consumer class actions with small
recoveries as "'paradigm class actions'" because no one would bring them if the class action device

was unavailable); Hensler et AL., supra note 2, at 68 (describing actions for money damages as

the "traditional paradigm" ofclass actions). One reason these cases are thought paradigmatic is that

they are considered to be a more straightforward use of the class action mechanism than mass tort

actions. For example, in Amchem Products, Inc. v. Windsor, the Court opined in passing that the

"predominance test," so difficult to meet in the mass tort context, is "readily met in certain cases

alleging consumer or securities fraud or violations ofthe antitrust laws." 521 U.S. 591,625(1997).

See also Richard A. Nagareda, Autonomy, Peace, and Put Options in the Mass Tort Class Action,

1 1 5 Harv. L. Rev. 747, 750 (2002) (observing that in the context of securities, antitrust, and

consumer litigation "the prospect of effecting a comprehensive peace through a class settlement

poses little threat to the autonomy of class members"); John C. Coffee, Jr., Class Action

Accountability: Reconciling Exit, Voice, and Loyalty in Representative Litigation, 100 COLUM. L.

Rev. 370, 386 (2000) [hereinafter Coffee, Accountability] (arguing that internal conflicts are not

typically present in "commercial class actions because ... in these cases, some objective measure"

of damages permits automatic allocation without discretion). Among other things, I hope to show

that the same intractable problems faced in mass tort and other more complex contexts are also

present in consumer class actions. I hope this Article will illustrate how very problematic small

claims or "negative value" class actions are. The law and policy concerning these different types

of class actions is so varied that no one system of governance—never mind a single procedural
rule—should be applied across the board to all types of class actions.



2003] CLASS ACTION GOVERNANCE 67

adversarial process, expertise of decisionmakers, and independence of
decisionmakers from influence and self-interest.

In a class action, absent persons, who may or may not want to be part of a
lawsuit, receive notice that they are part of a class action or that they are part of

a class action settlement/ They are denied the power to define the parameters
of their group. ^ They do not pick their representatives,^ and have no power to

4. Fed. R. Civ. P. 23(c)(2) provides:

In any class action maintained under subdivision (b)(3), the court shall direct to the

members of the class the best notice practicable under the circumstances, including

individual notice to all members who can be identified through reasonable effort. The

notice shall advise each member that (A) the court will exclude the member from the

class ifthe member so requests by a specified date; (B) thejudgment, whether favorable

or not, will include all members who do not request exclusion; and (C) any member who

does not request exclusion may, if the member desires, enter an appearance through

counsel.

5. In many cases, the attorneys will first seek certification ofa class with specific parameters,

then change the parameters of the class as part of a settlement or in preparation for trial. See

Hensler ET AL., supra note 2, at 2 1 7 (discussing Selnick v. Sacramento Cable, where the attorney

filed an amended complaint extending the class definition in preparation for trial); see also Brian

Wolfman & Alan B. Morrison, Representing the Unrepresented in Class Actions Seeking Monetary
Relief, 71 N.Y.U. L. REV. 439, 444-45 (1996) (discussing the problem of ensuring proportional

allocation where there is no "natural class"). Because global peace—or the ability to lay to rest the
most possible claims—is ofspecial interest to defendants, the practice ofenlarging the scope ofthe
class as part of settlement can smack of collusion.

6. The observation that the class does not pick its attorney and that the class representative

is merely a figurehead has become commonplace among scholars. See, e.g. , Coffee, Accountability,

supra note 3, at 406 ("Commentators have generally agreed that the representative in a class action

is more a figurehead than an actual decision-maker."); John C. Coffee, Jr., The Regulation of

Entrepreneurial Litigation: Balancing Fairness and Efficiency in the Large Class Action, 54 U.

Chi. L. Rev. 877, 877 (1987) [hereinafter Coffee, Entrepreneurial Litigation] (referring to client

control of litigation decisions in the class action as a "noble myth"); Samuel Issacharoff, Class

Action Conflicts, 30 U.C. DAVIS L. REV. 805, 805 (1997) (noting that courts determine adequacy

of representation based on "representations of counsel who have little if any connection to the

parties to be bound"); Henry Paul Monaghan, Antisuit Injunctions and Preclusion Against Absent

Nonresident Class Members, 98 COLUM. L. REV. 1 1 48, 1 1 49 n. I ( 1 998) ("Once enmeshed in a class

action, class members cannot shape their own claims, and their individual rights to participate in

the class proceeding are quite limited."). The acceptance of the proposition that the actual

relationship between the class and the class counsel is irrelevant is illustrated by the proposed Rule

23(g), which provides a series of factors for court approval of class counsel: the one factor

conspicuously absent from the list is client approval. See also Stewart v. Gen. Motors Corp., 756

F.2d 1285, 1295 n.5 (7th Cir. 1985) (quoting Alpine Pharmacy, Inc. v. Chas. Pfizer & Co., 481
F.2d 1045 (2d Cir. 1973)).

One accepting employment as counsel in a class action does not become a class

representative through simple operation of the private enterprise system. Rather, both

the class determination and designation ofcounsel as class representative come through



68 INDIANA LAW REVIEW [Vol. 37:65

dismiss those representatives.^ Instead, as a practical matter, class counsel

defines the group membership, manages the litigation, makes unilateral strategic

decisions, oversees the accrual of fees and costs, and shapes the outcome of a
mysterious process class members neither launched nor agreed to resolve.^ It is
not surprising, then, that class actions have been criticized as mechanisms by

which attorneys exploit disunited class members and garner enormous fees, whi le
claimants recover minimal damages.^ Nor is it surprising that 47% ofAmericans
believe that consumer class actions benefit plaintiffs' lawyers most and 20%

judicial determinations, and the attorney so benefited serves in something ofa position

of public trust.

Id. at 1050. There are, however, exceptions to this. Empirical studies have found that class

representatives sometimes seek out attorneys, rather than the other way around. See Hensler ET

AL., supra note 2 (describing examples of different relationships between class counsel and class

representatives). Institutional investors may act as lead counsel in securities class actions. 1

5

U.S.C.A. § 78u-4 (West 2003). Cf. Berger v. Compaq Computer Corp., 257 F.3d 475, 483 (5th Cir.

200
1 )

(holding that "PSLRA's requirement that securities class actions be managed by active, able

class representatives who are informed and can demonstrate they are directing the litigation" raises

the standard adequacy threshold in Rule 23). Thus, the class representative is a representative in

the formal sense, that is, that he or she theoretically has the authority to act on behalf of the class,

but no conception of popular sovereignty or accountability is imbedded in the mechanisms of

representation.

7. This is because after certification, class counsel's ethical obligations run to the class, not

to the individual class member. See In re Gen. Motors Corp. Pick-Up Truck Fuel Tank Prods. Liab.

Litig., 55 F.3d 768, 801 (3d Cir. 1995) ("Beyond their ethical obligations to their clients, class

attorneys, purporting to represent a class, also owe the entire class a fiduciary duty once the class

complaint is filed."); Jonathan R. Macey & Geoffrey P. Miller, The Plaintiffs' Attorney's Role in
Class Action and Derivative Litigation: Economic Analysis and Recommendationsfor Reform, 58

U. Chi. L. Rev. 1, 20 (1991) (noting that "the attorney must act for the benefit of the class as a

whole and therefore is not obliged to follow the unilateral wishes ofany individual class member").

8. Although there may be a presumed requirement that the attorneys consult with the class

representatives, if these representatives have been hand-picked by the attorney, such consultation

may have little meaning. In any event, the Rules create no obligation for class counsel to consult

with the class or answer to class representatives prior to notifying the class or presenting a

settlement to the court for approval pursuant to Federal Rule of Civil Procedure 23(e).

9. See, e.g., Jesse J. Holland, Official: Put Class Actions in U.S. Court, AP ONLINE, May
1 5, 2003, available at 2003 WL 553721 25 (stating that Justice Department spokesperson asserted
that trial lawyers benefit more from class actions than plaintiffs they represent); see also Deborah

R. Hensler, Revisiting the Monster: New Myths and Realities of Class Action and Other Large

Scale Litigation, 1 1 DUKE J. COMP. & Int'l L. 1 79, 1 80 (200
1 )

[hereinafter Hensler, Revisiting the

Monster] (noting popular belief that "monsters are loose in the land"); Susan P. Koniak & George
M. Cohen, Under Cloak ofSettlement, 82 Va. L. Rev. 1051 (1996) (arguing that attorneys often

abuse their position as class counsel and take advantage ofclass members in order to increase their

fees); Deborah R. Hensler & Thomas D. Rowe, Jr., Beyond "It Just Ain 't Worth It": Alternative
Strategies for Damage Class Action Reform, 64 LAW & CONTEMP. PROBS. 137, 137-39 (2001)
[hereinafter Hensler & Rowe, Beyond "It Just Ain 't Worth It"] (collecting similar criticisms).



2003] CLASS ACTION GOVERNANCE 69

believe that they benefit attorneys for defendants most, while only 9% believe
that class actions benefit plaintiffs most.'^

A few procedural protections purport to lend legitimacy to this system by
enlisting judicial oversight to prevent gross abuses of the class action. In

certifying a class, the court reviews the adequacy of the representatives and

counsel.'^ After the class is constituted and settlement has been reached, class

members are given opportunities to opt out or to speak out at a fairness hearing.'^
Nevertheless, an absent class member, not represented by counsel and armed

with minimal information, has a very limited ability to take advantage of these

mechanisms.'^ In the best case, class members' interests are represented by

responsible objectors. But for the most part, the class is left to rely on the

expertise of class counsel and the wisdom of judges to guarantee fairness.''^
Judges approve the class parameters, decide whether or not a settlement is fair,

and push the litigation in the direction of settlement or adjudication.'^

Notwithstanding the judge's role in overseeing class actions, fair results are far

from guaranteed.

The dispute over whether the benefits of the small claims class action

10. Opinions on Class-Action Lawsuits, USA Today, Mar. 24, 2003, at Bl, available at
2003 WL 5307581.

11. See Fed. R. Crv. P. 23(a) (setting forth the requirements for class certification).

Certification is no easy feat for plaintiffs counsel to pull off, but nevertheless, meeting the criteria

of 23(a) provides no guarantee of good governance.

12. See Fed. R. Civ. P. 23(c)(2) (requiring notice and opportunity to opt out at certification

to (b)(3) class members); 23(e) (class action may not be dismissed or compromised withoutjudicial

approval); 23(e)(1)(B) (requiring notice ofsettlement); 23(e)(1)(C) (court may approve settlement

upon finding that it is "fair, reasonable, and adequate"); 23(e)(3) (opportunity to opt out at

settlement is at court's discretion).

13. One Federal Judicial Center study found that only seven to fourteen percent of fairness

hearings were attended by class members who were not objectors or named plaintiffs and that in

forty-two to sixty-four percent of fairness hearings there were no objectors at all. Koniak & Cohen,
supra note 9, at 1 105-06.

14. See William B. Rubenstein, Divided We Litigate: Addressing Disputes Among Group

Members and Lawyers in Civil Rights Campaigns, 106 YALE L. J. 1623, 1663-68 (1997)

(discussing the benefits and drawbacks of an expertise-based model of group governance). Cf.

Coffee, Accountability, supra note 3, at 406 (characterizing the possibilities of voice under the

current Rule 23 regime as a choice "between the uninformed democracy of class members versus

the often self-interested professionalism of plaintiffs' attorneys").

15. See Fed. R. Crv. P. 23(c) (power of court to determine when class action may be

maintained); 23(c)(3) (to determine membership in class); 23(c)(4) (to require sub-classification);

23(d) (to issue procedural orders); 23(e) (to approve dismissal or compromise). See generally

Judith Resnik, Whose Judgment? Vacating Judgments, Preferences for Settlement, and the Role

ofAdjudication at the Close ofthe Twentieth Century, 41 U.C.L.A. L. REV. 1471 (1994) (arguing

that litigants do not expect to go to trial and that judges, by using rules, doctrine, and practices try

to get them to settle).



70 INDIANA LAW REVIEW [Vol. 37:65

outweigh its costs is longstanding, much discussed, and unresolved.'^ Among
other things, class actions solve the collective action problems faced by
individuals with claims too small to be economically adjudicated individually,

and address certain small private wrongs with substantial public effects,
especially in the absence of governmental intervention.'^ Thus, class actions

serve a regulatory function.'^

There are two substantive justifications for permitting groups to litigate
through the class action mechanism: compensation and deterrence.'^ Because the

1 6. "For all our effort, we do not know whether this is a good or a bad thing. The great big

question is whether the social utility of the large class action outweighs the limited benefits to

individuals, the aroma of gross profiteering, and the transactional costs to the court system." John

Frank, 1966 Civil Rules Advisory Committee Member in a memorandum to the chair of the 1995

Committee (quoted in Hensler ET AL., supra note 2, at 401). See also Hensler & Rowe, Beyond
"ItJustAin 't Worth It, " supra note 9, at 1 37-39 (describing ongoing debate over merits ofdamages

class actions); Hensler, Revisiting the Monster, supra note 9, at 1 96 (arguing whether class actions

are directed at doing social good or ill "depends substantially on how well judges control the

litigation process"). But see Elizabeth J. Cabraser, Enforcing the Social Compact Through

Representative Litigation, 33 CONN. L. REV. 1239 (2001) (arguing that class action litigation is a

beneficial social good).

1 7. In situations where claims are small, most ofthe individual class members "would have

no realistic day in court if a class action were not available." Phillips Petroleum Co. v. Shutts, 472

U.S. 797, 809 (1985). See also Deposit Guar. Nat'l Bank v. Roper, 445 U.S. 326, 338 n.9 (1980)

(describing negative value class action claims); Coleman & Silver, supra note 3, at 121 n.60 ("A
rational person will refuse to sue on a claim when the opportunity cost of a lawsuit exceeds its

expected return (the payoff given that the lawsuit may be won, lost or settled times the probability

of each outcome)."). As Arthur Levitt, former Chairman ofthe SEC, observed, "private litigation

is 'the primary vehicle for compensating defrauded investors.'" HENSLER ET AL., supra note 2, at

69. The class action mechanism seems to have competing objectives: administrative efficiency and

enabling litigation in pursuit of larger social goals. See id. at 49. Hensler points out that the

"distinction in the public debate between the efficiency and enabling goals of class actions for

money damages is illusory. In practice, any change in court processes that provides more efficient

means for litigating is likely to enable more litigation." Id. In recent years, there has been an

increase of involvement of state attorneys general in consumer issues. This development is an

interesting one and perhaps should affect the way policy-makers look at the regulatory role of

consumer class actions. On the other hand, this development may be too personality-driven to spur

a change in policy towards class actions.

1 8. "[M]any see class actions as a powerflil regulatory enforcement tool and view those who

bring them as 'private attorneys general.'" Hensler, Revisiting the Monster, supra note 9, at 1 82-

83.

1 9. Courts have interpreted the class action mechanism as a way to solve the collective action

problem of harms that are too small for individuals to prosecute. At the core of this goal is

compensation for those individuals. Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 617 (1997)

("the very core ofthe 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



2003] CLASS ACTION GOVERNANCE 71

aggregation of these small claims has a substantial effect on the defendant,
deterrence may have a larger influence than compensation.^° It may be that the
smaller the likely compensation to the individual, the more important the
deterrence justification becomes.^' To the extent that the class action's primary
goal is compensation, the objective of its governance scheme should be to
maximize claimholder vaiue.^^ To the extent that the primary goal is deterrence,
the governance regime should be concerned that the amount extracted from
defendants is sufficient to further that goal and is properly distributed among
class members, the public, and class counsel. This Article assumes that the

something worth someone's (usually an attorney's) labor." (quoting Mace v. Van Ru Credit Corp.

109 F.3d 338, 344 (7th Cir. 1997))); 5 JAMES Wm. Moore et al., Moore's Federal Practice

§ 23.02 (3d ed. 1998). Torts scholars have moved towards the view that tort norms, such as

deterrence, are a more appropriate way to view class actions. See David Rosenberg, Individual

Justice and Collectivizing Risk-BasedClaims in Mass-Exposure Cases, 7 1 N.Y.U. L. Rev. 2 1 0, 2 1

4

(1996) (proposing a functional approach to mass tort actions, viewing them from point of view of

tort policy of minimizing costs of accidents and maximizing deterrence). Scholars recognize a

larger social goal for securities class actions. See Janet Cooper Alexander, Rethinking Damages

in Securities Class Actions, 48 STAN. L. REV. 1 487 ( 1 996) (observing that the purpose ofsecurities

class actions is to protect the public interest and the integrity of capital markets). These

justifications necessarily exclude frivolous class actions brought by attorneys intent on obtaining

fees to "blackmail" defendants. See, e.g., Monte Morin, Lawyers Who Sue to Settle, L.A. TIMES,

Oct. 6, 2002, at Al (describing attorneys who file suits with the intent to settle for attorneys fees).

These types of "blackmail" class actions are not an internal governance problem of the type

addressed here, but are an important issue in class action policy generally. For mechanisms to

screen class actions, see Hensler & Rowe, Beyond "It Just Ain 't Worth It,'' supra note 9. For a
discussion of blackmail class actions and suggestions for reform, see Bruce Hay & David
Rosenberg, "Sweetheart " and "Blackmail " Settlements in Class Actions: Reality and Remedy, 75

NOTRE Dame L. Rev. 1 377 (2000).
20. A pure deterrence perspective does not provide a basis for distributing the proceeds

among the class members, but only requires that they be disgorged from the defendant. Cf.

Coleman & Silver, supra note 3, at 1 37 (arguing that the principle ofcompensation does not require
that the settlement fund be paid out to class members). It nevertheless seems fundamentally unfair

that compensation should mostly go to attorneys at the expense of wronged class members, even

if the wrongs themselves are small. When claimants are difficult to find, a charitable ftind is one

solution. See In re Mex. Money Transfer Litig., 164 F. Supp. 2d 1002 (N.D. 111. 2000) (approving

settlement that included cypres fund where claimants were illegal immigrants and thus difficult to

find), aff'd, 267 F.3d 743 (7th Cir. 2001).

21

.

See David L. Shapiro, Class Actions: The Class as Party and Client, 73 NOTRE DAME
L. Rev. 913, 923-31 (1998) (arguing that the main purpose of small claims class actions is

deterrence).

22. This is with the caveat that the amounts meted out in settlement should reflect the value

of the claim at litigation. See Janet Cooper Alexander, Do the Merits Matter? A Study of
Settlements in Securities Class Actions, 43 STAN. L. REV. 497, 566 (1991) (showing that in many

securities litigation settlements the amounts paid out cohere to a "going rate" of settlement rather

than an amount reflecting the strength of plaintiffs' claims).



72 INDIANA LAW REVIEW [Vol. 37:65

regulatory device of the class action attempts to achieve deterrence through

individual compensation. Whether this device is effective in achieving these
linked goals—and whether they need to be linked at all—is an open question. ^^

One might ask why we should care about class action governance at all.
Scholars viewing class actions from an economic perspective generally view the

central problem in class actions to be an agent-principal problem, and have a

correspondingly narrow view of the governance regime required to resolve it.^'*

Other scholars assume that participation values have a place in class actions, but

without providing the conceptual underpinning for this assumption beyond the

representation requirement in Rule 23P The due process requirements ofnotice,
hearing, and opportunity to opt out in 23(b)(3) class actions, combined with the

Rule 23(a)(4) representation requirements,^^ indicate that the rule makers saw
some autonomy value for individual claimants in class action litigation, but the
strength of that value is uncertain and the reasons for it are not clarified in the

case law.^^ The smaller the individual claim at issue, as in consumer class

23. Arguably the smaller the individual claim the lower the autonomy value associated with

that claim. See, e.g.. Gammon v. GC Servs. Ltd. P'ship, 162 F.R.D. 313, 321-22 (N.D. 111. 1995)
(holding that cypres distribution of damage award is appropriate where class of four million had

claim valued at thirteen cents each). In such cases, the compensation goal may be a barrier to

extracting the correct amount from the defendant and thus a barrier to deterrence.

24. The underlying assumption of economic analysis of this type appears to be that

compensation is the central goal of consumer class actions. See, e.g., John C. Coffee, Jr.,

Understanding the Plaintiffs Attorney: The Implications of Economic Theory for Private

Enforcement of Law Through Class and Derivative Actions, 86 COLUM. L. RJEV. 669 (1986)

[hereinafter Coffee, Understanding the Plaintiff's Attorney]; Macey & Miller, supra note 7. See
also Judith Resnik et al., Individuals Within the Aggregate: Relationships, Representation, and

Fees, 71 N.Y.U. L. REV. 296, 374 (1996) (describing how economic analysis of litigation has been

based on the agent-principal problem).

25. See, e.g., Owen M. Fiss, The Political Theory ofthe Class Action, 53 WASH. & LEE L.
Rev. 21, 25 (1996) (criticizing the self-selected representative in class actions as an "anomalous"

form of governance); Shapiro, supra note 21, at 958-59 (advocating for fuller representation of

class members as a group).

26. Federal Rule of Civil Procedure 23(a)(4) requires that the class be "adequately"

represented in order to be certified as a class.

27. For an enlightening discussion on the basis for and critique of this individual choice

rationale, see Robert G. Bone, Rethinking the "Day in Court " Ideal and Nonparty Preclusion, 67

N.Y.U. L. Rev. 193 (1992). See also Patrick Woolley, Rethinking the Adequacy ofAdequate

Representation, 75 Tex. L. Rev. 571 (1997) (arguing that due process requires notice, the right to

be heard, and opt out in the class action context and that these are inadequately met under the

current regime). Autonomy values, the traditionaljustification for due process protections and good

governance, are less powerful in this context. See Frank Michelman, Formal and Associational

Aims in Procedural Due Process, DUE PROCESS: NOMOS XVIII 126-28, 154 n.4 (J. Pennock & J.
Chapman eds., 1977) ("[R]espect for individual dignity, autonomy, and self-expression demands

that those with rights directly at risk have adequate means of registering their concerns."). Since

the claim is small, some might argue, as long as there are mechanisms to avoid fraud or self-dealing



2003] CLASS ACTION GOVERNANCE 73

actions, the more limited the autonomy values that can be placed on that
individual claim.

Despite small individual recoveries, the internal governance of small claims

class actions deserves attention.^^ Both the regulatory function ofthe class action

mechanism and the deterrencejustification are promoted by a robust governance

regime. Furthermore, a strong governance regime will benefit class members by
increasing agent monitoring to prevent exploitation. Good governance will
improve outcomes in class actions by multiplying the sources of protection

against agency problems and by giving advocates of alternative solutions an

opportunity to be heard.^^ Furthermore, because consumer class actions seek to

in payouts, the internal governance structure does not matter because the small size of the claims

renders autonomy values relatively weak. But autonomy values are not the only values for support

of participatory governance. Viewed collectively, the size of these cases is significant.

Communitarian values may be a more fitting place to look for class action governance. Cf.

Michael Sandel, Liberalism and the Limits of Justice 1 73-74, 1 83 (contrasting "liberalism"

with a "constitutive conception of community"). Although it is difficult to describe the consumer

class as a "community" in the way that community is typically defined in political science literature,

it nevertheless is possible to see the consumer class as a group with a public-oriented interest

beyond their individual interest in compensation. Cf. Frank 1. Michelman, Conceptions of

Democracy in American Constitutional Argument: Voting Rights, 4 1 Fla. L. Rev. 443, 445 (1 989)

(comparing liberal, individually oriented conceptions with the republican conception of an

autonomous public interest).

28. The reasons for the importance of the internal governance structure of class settlements

presented here are admittedly incomplete. I hope to develop a more rigorous analysis of the

organizational theory ofthe class action and to explore more fully the extent to which norms other

than litigant autonomy can serve as a basis for a robust governance regime in the future.

29. Some alternative options include the use of cy pres funds in addition to or instead of

compensation directly to class members. See 3 Newberg ON CLASS ACTIONS § 10:17 (4th ed.

2003) (describing use of cypres distributions of class damages); Kerry Barnett, Equitable Trusts:

An Effective Remedy in Consumer Class Actions, 96 YALE L. J. 1591 (1987). Other alternatives

may be the provision of repairs or retrofitting. See In re Ford Motor Co. Bronco II Prod. Liab.

Litig., 1995 U.S. Dist Lexis 3507 (E.D. La. Mar. 15, 1995) (noting objector suggestion that

retrofitting would significantly improve stability and safety of vehicle at issue in litigation).

Although there is no empirical evidence that good governance, or that this governance proposal in

particular, will yield better substantive results to class action suits, what seems clear is that the

current system of attorney control does not yield excellent substantive results. Furthermore,

vociferous criticism of plaintiffs' side class action attorneys indicates the current system of

governance has been difficult to justify on a results basis. This utilitarian argument has been one

basis for liberal political philosophy. See JOHN LOCKE, SECOND TREATISE OF GOVERNMENT 56

(Richard H. Cox, ed. 1985).

[A]s if when men quitting a state of nature entered into society, they agreed that all of

them but one should be under the restraints of laws, but that he should still retain all the

liberty ofthe state of nature, increased with power, and made licentious with impunity.

This is to think that men are so foolish that they take care to avoid what mischiefs may

be done to them bypolecats orfoxes, but are content, nay think it safety, to be devoured



74 INDIANA LAW REVIEW [Vol. 37:65

deter wrongdoing, they are a public good. Thus, the inclusion of deliberative

process, accountability and responsiveness, to claimants and the public, in class

action governance is justified for the same reasons notice and comment
provisions are integral to administrative and regulatory law.^° Finally, good
governance would give class action architects greater public legitimacy because
they can point to a process that resembles other legitimate processes for reaching

decisions in our society, in contrast to the current regime that increases the

perception that attorneys are exploiting class actions unchecked.^'

In pursuit of this goal, this Article looks at the class as a group in need of

governance and proposes a comprehensive set of principles through which to

judge the efficacy of class action mechanisms. This Article focuses on

governance in the settlement phase because it is one of the most contentious

moments in class action litigation. Part I illustrates the problems in the current
regime ofclass action governance for small claims class actions. Part II analyzes

and critiques three solutions to these problems that have been proposed over the

last twenty years: market mechanisms, democracy-based solutions, and judicial

administration. Part III proposes a comprehensive model of governance based

on four fundamental principles: mandatory disclosure, an actively adversarial

process, expertise, and independence of decision-making.

by lions.

Id.

30. The Administrative Procedure Act, for example, provides for notice and comment on

rules and thus indicates support for participation values in the administrative state. See a/^oNLRB

V. Wyman-Gordon Co., 394 U.S. 759, 764-65 (1 969) (opining public participation in administrative

rulemaking proceedings ensures that regulation is responsive to the needs ofthose regulated). Both

the defendants, who are represented directly in the proceedings, and class members, have concrete

interests in the outcomes and distributive effects ofclass actions. The public has a more attenuated

interest in these outcomes. In some circles, scholars are cynical about the relationship between

administrative and regulatory agencies and notions of popular sovereignty. Although there is

insufficient room to develop this issue further, there is significant work to be done on the

relationship between democratic values in regulatory litigation and the regulatory state.

3 1

.

The pervasive criticisms against class actions—which fall mostly in the lap of plaintiffs'
side class action attorneys—are damaging to the legal profession and to the regulatory purpose of
the class action mechanisms. Commentators who oppose class action litigation often point to

lawyer control as the reason for eliminating class actions. See, e.g. , WALTER OLSON, The Rule OF

Lawyers: HowtheNew Litigation Elite Threatens America's Ruleof Law(2003). Studies
and opinion polls have found that the public believes that lawyers are greedier and more dishonest

than other professionals. See Lynn A. Baker& Charles Silver, Introduction: CivilJustice Fact and
Fiction, 80 Tex. L. Rev. 1 537, 1 539 (2002) (describing studies). The bad press concerning class

action litigation has not helped this image. Good governance will lend legitimacy to the class

action process and by extension to the legal profession charged with overseeing these cases.



2003] CLASS ACTION GOVERNANCE 75

I. Problems [N Class Action Governance

A. A Misapplication ofthe Traditional Adjudication Model

Class action governance problems are rooted in the misapplication of the

traditional model of adjudication to the group context. While it may seem
obvious that a class action is more than simply a suit with multiple complainants,
such has been the power of the traditional model of adjudication that very few
procedural mechanisms were deemed necessary to ensure fair and efficient class
governance.^^ In reality, the contrast between the traditional model and class
action litigation is sharp.

In the traditional model of adjudication, the client is a voluntary and active

participant in the lawsuit, dictating its course." The attorney is bound by
specific, enforceable ethical duties to zealously advocate on behalfofthe client's

interests, to keep the client informed, and to consult the client at every critical

juncture in the litigation. The client's leverage over his agent is ensured by his
power to hire or fire the attorney at will. Finally, thejudge is a neutral arbitrator
of the dispute.^'*

Class actions set the traditional model of adjudication on its head.^^ The
class action effectively herds absentee plaintiffs into a lawsuit without their

consent and often without their knowledge in part because the right to opt out is

illusory. Class counsel fills the resulting power vacuum by proposing the
parameters of the class, recruiting named representatives, and making every

32. Class members are protected by a determination that the representation is adequate,

notice, a hearing, and the opportunity to opt out. See FED. R. CiV. P. 23. On the history of the

development of Rule 23, see Hensleretal., supra note 2, at 9-37.

33. See Abram Chayes, The Role ofthe Judge in Public Law Litigation, 89 Harv. L. Rev.

1281, 1282 (1976) (characterizing the traditional model of civil litigation as having the following

defining features: (i) it is bipolar, (ii) litigation is retrospective, (iii) right and remedy are

interdependent, (iv) the lawsuit is a self-contained episode, (v) the process is party initiated and

party controlled). This traditional model is an ideal type even when applied to individual litigation,

but has powerful resonance that has been particularly damaging in the class action context. See

generally Lon L. Fuller, The Forms and Limits ofAdjudication, 92 Harv. L. Rev. 353 (1978)

(basing legitimacy of adjudication on highly individualized participation).

34. See generally Marvin E. Frankel, The Searchfor Truth: An Umpireal View, 123 U. PA.

L. Rev. 1031 (1975) (discussing the role of the judge in the search for truth).

35. See Samuel Issacharoff, Preclusion, Due Process and the Right to Opt Out of Class

Actions, 77 NOTRE Dame L. Rev. 1057, 1058 (2002) ("A class action is simply, when all else is
stripped away, a state-created procedural device for extinguishing claims ofindividuals held at quite

a distance from the 'day in court' ideal of Anglo-American jurisprudence."). Some scholars have

made proposals from the point of view that the role of class action governance is to aggregate

individual preferences with a strong tilt towards autonomy values. See, e.g. , Coffee, Accountability,

supra note 3, at 379-80 (criticizing the "entity theory" as "a legal fiction" that "provides a

justification for the attorney making judgments in the 'best interests' of the class and allocations

among class members in precisely the manner Amchem seemed to forbid").



76 INDIANA LAW REVIEW [Vol. 37:65

important decision, including whether to accept or reject proposed settlements.^^

Class counsel are not required to attempt to identify or build majority support for

a settlement or survey class members to determine their interests. Far from being

a neutral or disinterested arbiter, the judge must often assume an active role in

facilitating settlement, in the process becoming interested in its approval.
^^

Despite these differences, few procedural modifications have been made to
the class action litigation process to accommodate the collective nature of class

action litigation. Instead of dictating the course of the suit, class members'

interests are pursued by a self-appointed or court-appointed "governor" who
decides what is in their best interest, often without input from his clients. As
Owen Fiss points out, "self-appointment is an anomalous form ofrepresentation,
only justified, if at all, by the most exceptional circumstances."^^ This form of

government resolves the problem of organizing a group into a "client" by

reducing the group to a few representatives of dubious legitimacy. The results
have been occasional instances of egregious corruption on the part of attorneys

who take advantage of class members^^ and a perception that consumer class
actions are not a public good, but a money making scheme for unscrupulous
lawyers.

B. The Simultaneous Expansion and Disintegration ofthe Class Action

Two forces work to expand the numbers of claimants included in class
actions. First, preclusion rules, anti-suit injunctions, and removal statutes

36. Coffee, for example, likens class counsel to a joint-venturer with the class, who has a

financial stake in the outcome ofthe litigation because ofthe considerable investment a class action

requires. He argues that the "leading cause of 'cheap settlements' may not be collusion between

class counsel and defendants . . . but rather a basic differential in the level ofrisk aversion" because

class counsel, with its significant financial investment in the action, is more likely to be risk averse

than the class. Coffee, Accountability, supra note 3, at 390-93. See also In re Cendant Corp. Litig.,

264 F.3d 201, 254-55 (3d Cir. 2001) (explaining that "[bjecause of this conflict (and because 'the

class' cannot counteract its effects via counsel selection, retention, and monitoring), an agent must

be located to oversee the relationship between the class and their lawyers. Traditionally, that agent

has been the court."). While not quite collusion, this perverse incentive does look like a type of

self-dealing.

37. Owen Fiss discussed this phenomenon in the context of what he perceived to be the

bureaucratization of the judiciary, predicting that where the judge serves "as both architect and

structural engineer" ofan institution (and, analogously, a settlement), he is likely to view challenges

to that institution as personal challenges. Owen M. Fiss, The Social and Political Foundations of
Adjudication, 6 LAW & HUM. Behav. 121, 126 (1982).

38. 5ee Fiss, 5wpra note 25, at 25.

39. See generally Koniak & Cohen, supra note 9, at 1058-69 (describing litigation in which
class members who received no benefit had attorneys fees deducted from their escrow accounts);

Wolfman & Morrison, supra note 5, at 473-77 (describing coupon settlements that did not benefit
classes).



2003] CLASS ACTION GOVERNANCE 77

encourage the consolidation of rival actions in a single forum."^^ Second, class

counsel has incentives to settle the most comprehensive action first, because only

one action, if any, will ultimately stand. Defendants play rival class counsel

against each other to settle for the lowest price, a process referred to as a reverse

auction."^' The phenomenon ofreverse auctions, or the threat ofreverse auctions,
pushes class counsel towards sub-optimal settlements and more global
settlements to ensure their own compensation."*^ Enterprising counsel may try to
bolster their settlement potential by buying out competitors."*^ When competing
attorneys know that they will be precluded by a settlement, a buy-out is an
attractive alternative.

The trend towards the expansion of small claims class actions over an ever-
increasing number of geographically widespread claims renders the internal
governance structure ofclass actions even more important."*"* The proposed Class
Action Fairness Act sought to further expand this trend by granting federal

diversity jurisdiction to state law-based small claims class actions."*^ Multi-

District Litigation panels consolidate multiple class actions to a single district for

discovery and settlement purposes."*^ The right to collateral attack has been
limited so that settlement in one forum will extinguish claims in other fora, even

claims that could not have been brought in the settling forum."*^ Settling parties

40. This in part because rival attorneys know ex ante that these procedural tools will be used

against them.

4 1

.

See John C. Coffee, Class Wars: The Dilemma ofthe Mass Tort Class Action^ 95 COLUM.

L. Rev. 1343, 1370 (1995) (describing the reverse auction phenomenon as "a jurisdictional

competition among different teams ofplaintiffs' attorneys in different actions that involve the same

underlying allegations").

42. See Coffee, Accountability, supra note 3, at 390-93 (discussing risk aversion of class

counsel).

43. For example, in the Bausch & Lomb class action discussed in Hensleretal., supra note
2, at 1 58, tag along class actions were settled for undisclosed, but reportedly nominal amounts.

44. The numbers of nationwide class actions are reportedly rising. See id. 2X6^-66. There

also is a countervailing trend towards localization ofclass actions. See, e.g.,K.]. Reynolds Tobacco

Co. V. Engle, 649 So.2d 245 (Fla. Dist. Ct. App. 1996) (denying certification of nationwide class

and approving certification of class of Florida citizens in tobacco class action).

45. See, e.g.. Class Action Fairness Act of2003, S.274, 1 08th Cong. § 4 (2003) (among other

things, proposing amendment to 28 U.S.C.A. § 1332 (West 2003) permitting original jurisdiction

of class action where the amount of the entire controversy exceeds two million dollars). Such an

amendment would dbvog?i\.Q Snyder v. Harris, 394 U.S. 332 (1 969) and Zahn v. International Paper

Co., 414 U.S. 291 (1973).

46. ^ee 28 U.S.C. § 1407 (West 2003).

47. See Matsushita Elec. Indus. Co. v. Epstein, 516 U.S. 367 (1996) (holding that the Full

Faith and Credit Act requires federal courts to give preclusive effect to state judgments even when

state court judgment at issue incorporates class action settlement releasing claims solely within

jurisdiction of federal courts, so long as they comply with due process). Subsequently, on remand

the 9th Circuit held that the Supreme Court's decision determined that the state court judgment

comported with due process. See Epstein v. Matsushita Elec. Indus. Co., 179 F.3d 641 (9th Cir.),



78 INDIANA LAW REVIEW [Vol. 37:65

can obtain an anti-suit injunction preventing class members from bringing suits
in any other forum.'*^ Or, they may seek a determination that a rival action is
barred in the forum where a competing suit was filed/^

For example, in the Mexico Money Transfer Litigation,^^ a nationwide
consumer class action arising out of allegations that money transfer companies
overcharged customers by manipulating the exchange rate, a total of eight

lawsuits were initially filed.^' Most ofthese were dismissed, dropped or stayed

by the federal court in the Northern District ofIllinois that ultimately decided the

case.^^ After this consolidation. National Class Counsel, consisting of the

attorneys who had filed class actions in Texas and Illinois, negotiated with the
defendants." This process eliminated potential objections ex ante.^*

cert, denied, 120 S. Ct. 497 (1999).

It is now settled that ajudgment pursuant to a class settlement can bar later claims based

on the allegations underlying the claims in the settled class action. This is true even

though the precluded claim was not presented, and could not have been presented, in

the class action itself.

In re Prudential Ins. Co. ofAm. Sales Practice Litig., 261 F.3d 355, 366 (3rd Cir. 2000). See also

Marcel Kahan & Linda Silberman, The Inadequate Search for "Adequacy" in Class Actions: A
Critique ofEpstein v. MCA, Inc., 73 N.Y.U. L. REV. 765 (1998) (criticizing the right to collaterally

attack a settlement for inadequacy of representation as excessively broad). The right to collateral

attack in state court appears to be expanding somewhat. See State v. Homeside Lending, Inc., 826

A.2d 997 (Vt. 2003).

48. See Monaghan, supra note 6, at 1150, 1151-52 n.l3. (discussing Baker v. General

Motors' Corp., 118 S. Ct. 657, 665 (1998), which distinguished between enforcement and

preclusive effects of antisuit injunctions). As Monaghan points out, the nature ofthe relationship

between the enjoining court and the enjoined court has yet to be worked out.

49. See SyngentaCrop Prot. Inc. v. Henson, 123 S. Ct. 366, 379 n.l (2002).

50. 164 F. Supp. 2d 1002 (N.D. 111. 2000), affd, 267 F.3d 743 (7th Cir. 2001), cert, denied,

Garcia v. W. Union Fin. Servs., Inc., 535 U.S. 1018 (2002).

51. In In re Mexico Money Transfer Litigation, two suits were filed in federal court in

California, After the federal claims of the initial suits were dismissed, these were refiled in

California state court on behalf of a California class with solely California state claims. A third
federal suit was filed in California federal court naming a nationwide class. Two class actions were

filed in federal court in the Northern District of Illinois. These were subsequently consolidated.

Two class actions were filed in federal court in Texas on behalf of a Texas class. Another class
action was filed in Texas state court also on behalf of a Texas class. See 164 F. Supp. 2d at 1008-

09.

52. Id.

53. 164 F. Supp. 2d at 1010. The district court noted that the parties' declarations in support

of settlement described a year-long, arms-length negotiation and that attorney's fees were only

negotiated after a preliminary settlement had been reached.

54. The objectors who were represented by counsel consisted of California class members

who alleged that they would have stronger claims under California law. The remaining objectors

were not represented by counsel. While it is still possible for the remaining class members to

object, many potential committed objectors will opt out.



2003] CLASS ACTION GOVERNANCE 79

Concurrent with the expansion of class actions is a disintegration of the

courts' power to consolidate and control them. This is a federalism issue as well

as one of manageability. Blatant forum shopping is rife in class actions. Class

counsel will file national actions in state court in order to gain the leverage of

hospitable state courts. Both class counsel and defendants will push class actions

towards state court to obtain settlement approval. ^^ A settlement rejected in
federal court on fairness grounds may be re-filed in state court and approved
there.^^ Both class counsel and defendants will pull towards federal court when
they believe they can gain more leverage or obtain global settlement in that
forum. Objectors and rival class counsel can also use procedural rules to their

advantage.^^ Even when a state suit disrupts the settlement already reached and
approved in federal court, it is up to the state court to determine whether or not

the suit is barred. Collateral attacks in state courts on personal jurisdiction

grounds based on denial ofdue process protections are also gaining momentum.^^

All the while, the courts work to prevent the multiplication of class actions to
conserve the efficiency of the class action mechanism and judicial resources.

A viable governance regime for class actions must recognize the prevalence of
forum shopping and the corollary phenomena ofbuy-outs ofcompetition by class
counsel and defendants.

C. Barriers to Participation

Significant systemic barriers block direct class member participation in class
actions. Under the current governance regime, class members have a minimal
role in the litigation.^^ Class counsel's role as entrepreneur encourages counsel

5 5

.

The reasons for state court's apparent deference to settlements are not clear, but anecdotal

evidence suggests that this is the case.

56. For example, when the Third Circuit overturned a somewhat notorious settlement in the

In re General Motors Pick-Up Truck Fuel Tank Products Liability Litigation, the parties re-filed

the settlement in Louisiana state court, where it was approved. 134 F.3d 133 (3d Cir. 1998). See

Public Citizen Litigation Group, Public Citizen 's Involvement in Class Action Settlements (Feb, 28,

2002) at http://www.citizen.org/litigation/briefs/Class_Action/articles.cfm?ID=552.

57. For example, rival attorneys or objectors can turn to state courts and the All Writs Act

cannot furnish removal jurisdiction where a federal court has no original jurisdiction. See Syngenta

Crop Prot. Inc. v. Henson, 537 U.S. 28 (2002). In Syngenta, a suit was settled in a consolidated

federal court proceeding with plaintiff's counsel participating. That same counsel then brought a

competing class action in state court and argued that the settlement did not preclude the subsequent

state court action. The Supreme Court held that the state court would decide whether or not the

subsequently filed state court action violated the federal court settlement. Id

58. See State v. Homeside Lending, Inc., 826 A.2d 997 (Vt. 2003) (holding that class action

settlement did not preclude suit by Vermont residents because Alabama court lacked personal

jurisdiction due to defects in due process protections).

59. Although class members have a formal right to participate in the class action,

participation is difficult and costly. Structural barriers prevent class members from exercising their

right to participate, particularly in the absence of attorney representation.



80 INDIANA LAW REVIEW [Vol. 37:65

to keep information from the class in order to retain full control of the

negotiation and settlement process. Communication with far-flung class
members is expensive, and will likely lead to the discovery of class member
preferences or internal class conflicts that class counsel would rather ignore.

Were there a requirement that class counsel solicit class member views, or that
a certain percentage ofthe class "opt-in" to the class action by submitting claims,

this relationship would change dramatically.

Class members are not repeat players and they cannot create ongoing
relationships with class counsel, therefore they lack leverage over their attorney.

Because there is no ongoing relationship, even the type of leverage that a

consumer has over a manufacturer is lost in the class action context. In a
commercial enterprise, a consumer's decision not to buy a product has a direct

effect on the bottom line, even if a small one. In the class action context, opt outs

will not affect attorney remuneration unless they are so massive as to scuttle the

entire settlement. Because of the ongoing relationship between consumers and

manufacturers, a decline in services or the product in the commercial enterprise

may activate a consumer response to alert the firm to the problem and lead to
recovery.^^ There is no such ongoing relationship in the class action.

Structural realities and perverse incentives inhibit participation to an even

greater extent in settlement classes. The first notice the class receives of the
litigation is also a notice of settlement. If they are dissatisfied with the

settlement, they may opt out or protest the settlement in the fairness hearing. The
hearing is usually a non-adversarial proceeding because both parties' attorneys

want approval of the settlement they have worked hard to formulate.^' Worse
yet, the client—that is, the class—exists at the whim ofthe defendant because the

60. See ALBERT HlRSCHMAN, EXIT, VOICE ANfD LOYALTY: RESPONSES TO DECLINE IN FiRMS,

Organizations and States 15 (1970). In order for exit to be an effective means of

communicating consumer dissatisfaction, enough individuals must defect to another firm in order

to affect the company's bottom line. But unless the company is vigilant, which a firm in decline

is unlikely to be, by the time the company understands consumer dissatisfaction, the decline may

already be too far underway for correction.

For competition (exit) to work as a mechanism ofrecuperation from performance lapses,

it is generally best for a firm to have a mixture of alert and inert customers. The alert

customers provide the firm with a feedback mechanism which starts the effort at

recuperation while the inert customers provide it with the time and dollar cushion

needed for this effort to come to fruition.

Id. at 24. Several prominent class action scholars have discussed the problems of class action

governance using Hirschman's work as a guide or inspiration. See Coffee, Accountability, supra

note 3, at 376 n. 1 7 (using the corporate governance paradigm of"exit, voice, loyalty" but asserting

no intent to adopt specific conclusions from Hirschman's work); Samuel Issacharoff, Governance

and Legitimacy in the Law ofClass Actions, 1999 SUP. Ct. Rev. 337, 367 (1999).

61

.

The fairness hearing is non-adversarial where there are no objectors. Where there are

objectors represented by sophisticated counsel, the objectors may bring an adversarial aspect into

the proceeding.



2003] CLASS ACTION GOVERNANCE 81

court has not certified the class action.^^ Where a certified class is expanded
during settlement negotiations, usually to assist defendants in obtaining global

peace, the same problem arises." Thus, even a settlement reached subsequent to

certification may have a "settlement class"—a number of people v^hose first
connection to the class action is through settlement and whose claims did not go
through the certification process.^'*

Those due process protections that seem on their surface to encourage some
class member participation have proven ineffective. The fundamental due
process requirements for 23(b)(3) class actions are notice, the right to be heard,

and the right to opt out.^^ Opt out, or exit, is an ineffective means of realizing

participation values because small claims class members cannot bring
independent claims. Notices are difficult for ordinary people to understand and

hearings are non-adversarial and, therefore, fail to address inequities in

settlements. The structure of notices and fairness hearings does not facilitate the
possibility of persuasive objectors who are able to point out the inadequacy of
settlements.

The means by which a class member may exit a class action is by opting out
ofthe class either at the certification or settlement stage.^^ The central argument

62. See IssacharofT, supra note 60, at 348 (discussing court's treatment of settlement classes

\n Amchem Products, Inc. v. Windsor, 521 U.S. 591 (1997), and arguing that even though Supreme

Court in Amchem warned against the evils of settling without reference to testing of claims in the

adversarial process, that is just what settlement classes are). Because no certification motion has

been approved in a settlement class, if defendants pull out ofthe settlement then the class will cease

to exist. In that case, class counsel will have to litigate a difficult and costly certification motion.

63. See Hensleretal., supra note 2, at 157-58 (discussing the expansion ofthe class after

settlement in the Bausch & Lomb contact lens pricing litigation). That RAND study notes the
expansion of class parameters in several of the case studies.

64. To the extent that we believe that the expansion of the class for settlement purposes does

not raise notice problems, it is because we do not believe that the initial notice itself was

particularly effective.

65. See Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 812(1 985) (holding that for personal

jurisdiction over absent class members, due process requires notice, the opportunity to be heard,

and the opportunity to opt out); see also Patrick Woolley, Rethinking the Adequacy ofAdequate

Representation, 75 Tex. L. Rev. 571 (1997) (arguing that due process requires notice, the right to

be heard, and the right to opt out in the class action context and that these are inadequately met

under the current regime); Monaghan, supra note 6, at 11 66 (discussing due process requirements

under Shutts).

66. See Fed. R. Civ. P. 23(c)(2)(B) (requiring "the best notice practicable under the

circumstances, including individual notice to all members who can be identified through reasonable

effort" and requiring that this notice be in "plain, easily understood language" and include "the

nature of the action, the definition of the class certified, the class claims, issues, or defenses, that

a class member may enter an appearance through counsel if the member so desires, that the court

will exclude from the class any member who requests exclusion, stating when and how members

may elect to be excluded, and the binding effect ofthe class judgment on class members under Rule

23(c)(3)"). Opt outs are not required in limited fund class actions brought under Rule 23(b)(1) or



82 INDIANA LAW REVIEW [Vol. 37:65

in favor of opt outs is that they permit the realization of individual choice.^^

Courts understand opt outs as a pure expression of claimant autonomy, one that

cannot be collectivized.^* The autonomy rationale for opt outs assumes that
claimants make knowing and intelligent decisions to opt out or stay in the class
action, which we know is often not the case in small claims class actions.
Individual opt outs make the most sense where a claimant does not want to be
part of a lawsuit in the first place and/or has a positive value claim that can be

brought independently, because opt outs place the claimant in the same position
she would have been in absent the class action mechanism.^^

Exit is not a viable means ofrealizing autonomy in a non-competitive space,
however, because practically speaking, opting out simply removes that class

member from the debate. Opt outs are not viable solutions to an inadequate
settlement because once a class member has opted out, he or she no longer has
standing to participate in the class action.^^ The opt out, as an expression of
autonomy, could be valuable where there is a realistic alternative to the

settlement; otherwise, the right to opt out is a mere formality.^' A class member
who has opted out of a small claims class action will not be able to file a claim
on her own. From the perspective of that class member, the right to opt out is
meaningless because there is nowhere to opt out to.

Theoretically, massive opt outs might be an effective means of
communicating client dissatisfaction since courts consider not opting out to be

injunctive class actions brought under Rule 23(b)(2).

67. See Coffee, Accountability, supra note 3, at 418 (arguing that exit through opt outs

maximizes individual choice).

68. "The right to participate, or to opt out, is an individual one and should not be made by

the class representative or the class counsel." Hanlon v. Chrysler Corp., 1 50 F.3d 1011, 1 024 (9th

Cir. 1998). Hanlon argued that the right to opt out must be an individual right not only for due

process concerns of the individual class members "who have the right to intelligently and

individually choose whether to continue in a suit as class members," id., but also because any other

rule "'would lead to chaos in the management ofclass actions.'" Id. (quoting Berry Petroleum Co.

V. Adams & Peck, 518 F.2d 402, 412 (2d Cir. 1975)). Collective opt outs, which have been
rejected by courts, make more sense where the choice of law renders one group of claimants more

valuable than others.

69. But see Rubenstein, supra note 1 4, at 1 648-49 (arguing that the class device is inadequate

because a litigant cannot get a case un-filed, and the opportunity to participate in remedy not useful

for one who does not want the case at all). There is a possibility, albeit unlikely, that a class of opt

outs could seek certification and mount a rival class action, and that the courts might harness such

potential rivalries to yield more fair results. John Coffee proposes mechanisms for encouraging

such competition. His solution is discussed infra pp. 32-33.

70. See Mayfield v. Barr, 985 F.2d 1090, 1093 (D.C. Cir. 1993) (finding that class members

who have opted out of a damages class action have no standing to object to a subsequent class

settlement because by opting out they "escape[] the binding effect of the class settlement").

71. See, e.g.. Coffee, Accountability, supra note 3, at 378 (discussing small claims class

action where "right to exit will mean little" in contrast to the mass tort or large claims case where

right is critical because at least some claimants hold high value claims).



2003] CLASS ACTION GOVERNANCE 83

tacit consent/^ A near-total opt out would be something like a boycott of the
settlement and would make possible a second, perhaps better, class action on
behalf of opt outs. But in a world where numerous objections have not been

sufficient to spur courts to reject settlements, a campaign ofopt outs is not likely

to be effective^^ Courts may read massive opt outs as a protest, or may see them
as an exercise in client autonomy that legitimates acceptance of the offending

settlement for the remaining class members. This creates a double-bind for class

members, who lack standing to intervene in the fairness hearing if they have
opted out and are trapped in the settlement they disapprove of if they stay in,

object, and lose. Even under the revised rule, which permits opt outs at
settlement, the timing of the opt out is such that class members must choose
whether to opt out before the fairness hearing.^"* Courts have rejected opt out

attempts to appeal settlements for lack of standing, perhaps because courts

sympathize with class counsel's arguments that attorneys representing opt outs

are attempting to hijack the intervention process to obtain leverage over class

counsel and defendants in order to procure a better settlement for the claimants

who opted out.^^
One may ask why we should be concerned about the exclusion ofdissatisfied

class members with small value claims. Some argue that we should not because
absent the class action collectivization mechanism, these dissenters would have

no recourse at all,^^ and therefore any recovery is a windfall for them. There are

several responses to this view. First, because some dissatisfied class members

72. See, e.g.. Bell Atl. Corp. v. Bolger, 2 F.3d 1304, 1314 n.l5 (3d Cir. 1993) (noting that

"silence constitutes tacit consent," but recognizing that this assumption may, as a practical matter,

understate potential objectors); In re GNC S'holder Litig., 668 F. Supp. 450, 45 1 (W.D. Pa. 1 987)
(noting that "[i]n the class action context, silence may be construed as assent").

73. See, e.g.. Reed v. Gen. Motors Corp., 703 F.2d 170, 174-75 (5th Cir. 1983) (approving

settlement over objections of twenty-three of twenty-seven named plaintiffs and forty percent of

class); TBK Partners, Ltd. v. W. Union Corp., 675 F.2d 456 (2d Cir. 1982) (approving settlement
over objections ofapproximately fifty-six percent ofclass); Cotton v. Hinton, 559 F.2d 1 326, 1333-

34 (5th Cir. 1977) (approving settlement despite objections of counsel purporting to represent

nearly fifty percent ofclass); Bryan v. Pittsburgh Plate Glass Co., 494 F.2d 799 (3d Cir.) (approving

settlement despite objections oftwenty percent ofthe class), cert, denied, 4 1 9 U.S. 900 ( 1 974). But

see In re Corrugated Container Antitrust Litig., 643 F.2d 1 95, 2 1 7- 1 8 (5th Cir. 1 98 1 ) (recognizing

that "a low level of vociferous objection is not necessarily synonymous with jubilant support");

Pettway v. Am. Cast Iron Pipe Co., 576 F.2d 1157 (5th Cir. 1978) (overturning approval of

settlement where seventy percent of class objected).

74. See FED. R. Crv. P. 23(e)(3) ("[T]he court may refuse to approve a settlement unless it

affords a new opportunity to request exclusion to individual class members who had an earlier

opportunity to request exclusion but did not do so.").

75. See, e.g.. In re Brand Name Prescription Drugs Antitrust Litig., 115 F.3d 456 (7th Cir.

1997) (barring organized opt outs from appealing settlement).

76. For example, Monaghan approves ofthe result in "situations where the absent plaintiffs

lack independently viable claims and stand only to 'win' by the lawsuit." Monaghan, supra note

6, at 1169.



84 INDIANA LAW REVIEW [Vol. 37:65

opt out for good reasons, the argument for ignoring opt outs is in effect an

argument against the policy of vindicating and deterring small wrongs through

class actions. The "windfall" argument also ignores the regulatory function of
class action suits. Because legislatures have made consumer protection through
lawsuits a staple of our political landscape, it is important that these suits

produce fair results, whether that means individual compensation or deterrence.
The deterrence value of small claims class actions further requires that the
system ensure that the amount extracted from defendants is appropriate.^^ The
deterrence value alone cannot determine distribution ofthe fund, however. There

may be better justifications for excluding dissenters from decisions about how
settlement funds are distributed, because they do not have a stake in that

distribution.^^ On the other hand, dissenters will often be those who are most
aware ofthe inequities ofthe distribution scheme developed by the attorneys, and
their input will benefit class members who do stand to gain from the fund. In the
context where the state has successfully consolidated claims through the class

action mechanism in the hands of class counsel, dissenting voices become more
important as indicators ofa failure of process because competition is not a viable

outlet for dissent.^^

The two other procedural protections meant to create some limited class
member participation, notice, and hearing, are also inadequate as currently
enforced. As a form of communication with informational intermediaries and
monitors, such as objectors' attorneys and judges, notices provide insufficient

information. As a form of communication with class members, notices are
difficult to understand and therefore ineffective. The types of notice used in

most class actions indicate that courts rarely take the notice requirement

seriously. Notices are generally written in a form of legalese that is difficult, if

not impossible, for the ordinary claimant to understand.^^ Some are even

77. See generally Alexander, supra note 22 (arguing in favor of settlements that reflect

outcomes at trial rather than the "going rate" of settlement in securities context).

78. This argument does not apply to claimants who stand to benefit from the fund. The only

justification to exclude small claims class members from participating in such distribution decisions

is that this type of input is too expensive or inconvenient to obtain.

79. Sometimes class members opt out because they oppose the action altogether as harmful

to the interests of the class as a whole. In that case, opting out will not help them because they

cannot get their case "un-filed" by opting out. See Rubenstein, supra note 14, at 1648-49 (noting

that the class action device is an inadequate solution to the problem of the limits of individual

autonomy in the group litigation context because class members cannot get a case "un-filed" and

the opportunity to participate is not a useful remedy where plaintiffs do not want the lawsuit

brought in the first place). See, e.g.. In re Matzo Food Prods. Litig., 156 F.R.D. 600, 608 (D. N.J.

1994) (rejecting settlement and noting that plaintiffs who have business relationships with

defendant may be reticent to bring claim against defendant).

80. "It is beyond the experience or expectation of reasonable citizens that the failure to

respond to what looks like a slightly unusual piece ofjunk mail constitutes assent to the solicitation

. .
." Paul D. Carrington & Derek P. Apanovitch, The Constitutional Limits ofJudicial Rulemaking:

The Illegitimacy ofMass-Tort Settlements Negotiated Under Federal Rule 23, 39 ARIZ. L. REV.



2003] CLASS ACTION GOVERNANCE 85

inaccessible to a reader trained as an attorney.^' Even the notice requirements of
the revised Rule 23(c), which provide some specific disclosures, are
insufficient.^^ Rule 23 does not require notices to include the amount an
individual class member will receive or the amount of attorneys' fees.^^

Fairness hearings are equally flawed. They are often too short to achieve the
goal of genuine debate over the merits of the settlement.^"* The timing of filings
further limits the usefulness of fairness hearings. Objectors are often required

to file their opposition motions before class counsel and defendants file their

motions in support of settlement. ^^ This timing, combined with the limits on
objector discovery, leaves objectors at a disadvantage because they must develop

their objections without the information possessed by class counsel and

defendants.^^ Such hearings provide little opportunity for meaningful objection

and allow class counsel and defendants to push their settlement through without

461,467-68(1997).

81. See, e.g., Kamiiewics v. Bank of Boston Corp., 100 F.3d 1348, 1352 (7th Cir. 1996)

(Easterbrook, J., dissenting) ("The notice not only didn't alert the absent class members to the

impending loss but also pulled the wool over the state judge's eyes.")

82. See Fed. R. Civ. P. 23(c)(2)(B) (requiring that notices to class members in 23(b)(3) class

actions provide the following information in "plain, easily understood language": ( 1 ) nature ofthe

action, (2) the definition of the class certified, (3) the class claims, issues or defenses, (4) that a

class member may enter an appearance through counsel, (5) that any class member may request

exclusion from the class, (6) the binding effect of the class judgment).

83. See Hensler ET al., supra note 2, at 451-53. There is some movement towards more

rigorous notice requirements. See State v. Homeside Lending, Inc., 826 A.2d 997 (Vt. 2003)

(finding lack of personal jurisdiction over Vermont plaintiffs in part because notice was

substantively inadequate); infra note 270 (citing cases holding that disclosure of attorneys fees is

a required part of class notice).

84. See, e.g. , Wolfman & Morrison, supra note 5, at 489 (describing the Mustang convertible
coupon settlement where objectors obtained no information prior to the fairness hearing and the

judge approved settlement after a thirty minute hearing with no evidentiary support in favor ofthe

settlement presented).

85. See id. at 480-90 (discussing the difficulties of obtaining information as objectors); see

also BriefofAmicus Curiae Public Citizen in In re Prudential Ins. Co. ofAm. Sales Practices Litig.

Agent Actions, No. 99-5960 (3d Cir.), available at http://www.citizen.org/litigation/briefs/

ClassAction/ articles. cfm?ID=693 (detailed discussion of legal and practical difficulties faced by

settlement objectors).

86. Potential objectors are kept out ofthe negotiation process and are permitted very limited

discovery. Courts have barred objectors from cross-examining fairness experts and from obtaining

discovery of side-settlements. See, e.g., Rutter& WilbanksCorp. v. Shell Oil Co., 314 F.3d 1 180,

1 186 (10th Cir. 2002) (affirming settlement and holding that hearing did not violate due process

where objectors were forbidden from cross examining parties' fairness expert, not permitted to

present live testimony, and not permitted to present rebuttal affidavits); Duhaime v. John Hancock

Mut. Life Ins. Co., 183F.3d 1 (1st Cir. 1999) (denying objectors discovery ofside-settlement). The

revised Rule 23(e)(2) requires parties to file a statement identifying side deals but not the substance

of those deals.



86 INDIANA LAW REVIEW [Vol. 37:65

debate.

In their Rule 23(e) review of settlements for procedural fairness, courts will

generally look to whether "(
1
) the negotiations occurred at arms length; (2) there

was sufficient discovery; (3) the proponents ofthe settlement are experienced in
similar litigation; and (4) only a small fraction of the class objected."^^ All but

one of these criteria are fundamentally flawed. First, even parties engaging in

arms-length negotiations can face incentives that lead them to self-serving
settlements that do not adequately compensate the class. This is especially true

where the "client" cannot monitor the attorney's actions. Second, in evaluating

discovery, courts only look at discovery as between defendant's and plaintiffs

counsel. Discovery is only sufficient if both class counsel aw^objectors have an

adequate opportunity to obtain material information. Third, the absence of

objectors has no necessary relationship to the fairness of a settlement. Few
objectors can mean that there is little to object to because the settlement is fair.
But it can also mean that class members lacked sufficient opportunity to object.
For example, there may be few objections because notices were too confusing,
class members had insufficient information, or class members were not given an
adequate opportunity to voice their objections. In addition, objections may be
limited because even though a settlement is unfair, class members have made the
cost-benefit calculation that their potential individual recovery is too small to

merit involvement. Other types of objectors, such as public interest groups or

state attorneys general, may not object to unfair settlements because they did not
know of the settlement or lacked the resources to intervene. Because the costs
of objecting will exceed the value of any individual claim, unscrupulous class

counsel can take advantage of the same type of collective action problem that

was to be remedied by the class action device in the first place. Only the
experience of the attorney in similar litigation (essentially a measure of quality

of leadership), is a good procedural measure. Even this measure is problematic,

however, because a self-dealing counsel may have substantial experience in
similar litigation, but not have reached adequate results for her clients.

Compounding the flaws in these procedural requirements is the problem of
ascertaining class member preferences. The structure of Rule 23 implies that
class member preferences can be ascertained through the attorney-class
relationship or, as a last resort, at a fairness hearing, but these provisions are

inadequate to the task. There are numerous practical barriers to ascertaining

class member preferences. Reaching out to class members takes attorney time
and money. Voting, for example, would exponentially increase the costs

associated with notice as each additional communication means substantial

additional expense. Community meetings are also difficult to organize when
class members are geographically disbursed. Even effective mechanisms might
not guarantee high response rates.^^ Even if costs were not an issue, there are

87. See In re Gen. Motors Pick-Up Truck Fuel Tank Litig., 55 F.3d 768, 785 (1995).

88. There is little empirical data on claimant response rates, although commentators believe

that they are low. See Rubenstein, supra note 14, at 1657-58 ("[F]ew class members respond to

court mailings and those who do are not representative.").



2003] CLASS ACTION GOVERNANCE * 87

other difficulties in ascertaining preferences. The standard economic assumption
that class members' preferences are static and that claimants seek to maximize

the payout on their claims is not entirely appropriate where claimants are not
educated about their claims or where claimants are not engaged in repeated,

similar transactions. Moreover, the settlements themselves are confusing and

difficult to understand. Even if a decisionmaker was assured that claim
maximization was the goal, substantive preference questions still arise because
the possibilities for structuring payouts make for difficult choices. For example,
reasonable claimants may differ as to whether a coupon or the promise ofin-kind
services or a cypres fund is preferable to a cash payout of less value.

^^

The substance of preferences collected depends on the level of education
provided to class members and the methods by which preferences are
ascertained. For example, class members may wish to realize values other than
claim maximization; they may prefer to support consumer advocacy groups rather
than obtain minimal individual compensation.^*^ Ifthe formulator ofthe question

assumes that claim maximization is the only possible value, however, class

members will not be able to express this preference. Because class members'
preferences will depend on how informed they are, the institutional designer
needs to determine how much education to provide, balancing the cost of
education against the value of better decision making by class members.^'

Even minimal participation demands informed class members, but under
current practices, as a result ofincomprehensible notices, uninformative fairness

hearings, and alienated representatives, class members are uninformed. Thus,
even if class members were willing to monitor their agents, they could not do so
effectively, and courts do not fill the breach. For example, neither the Rules nor

most courts require the parties to report on the ultimate payout at the end of

settlement administration.^^ They do not bar reversion ofsettlement funds to the

89. As one political scientist explained: "[u]nder the standard assumptions of modern

economic theory, public choice theorists have demonstrated the logical impossibility ofconstructing

any attractive, consistent procedure for making collective choices from unrestricted sets ofthree or

more alternatives." LARRY M. Bartels, PRESIDENTIAL Primaries and the Dynamics of Public

Choice 296 (Princeton 1988). For an interesting discussion of nonpecuniary settlements and

recommendations for evaluating such settlements, see Geoffrey P. Miller & Lori S. Singer,
Nonpecuniary Class Action Settlements, 60 LAW «& CONTEMP. Probs. 97 (1997). The consumer
class action settlements that have been most criticized tend to be coupon settlements in which class

members receive the dubious opportunity to obtain more services from the defendant who cheated

them in the first place. See, e.g., Wolfman & Morrison, supra note 5, at 472-77 (discussing
inadequate coupon settlements).

90. See Patricia Sturdevant, Using the Cy Pres Doctrine to Fund Consumer Advocacy, 33

Trial 80 (November 1997) (arguing in favor of distributing residue of class settlements to

consumer advocacy organizations); Kerry Barnett, Note, Equitable Trusts: An Effective Remedy in

Consumer Class Actions, 96 YALE L. J. 1591 (1987).

9 1

.

Bartels argues that students ofgovernance should be concerned not with "summation of

preferences" but with "formation of preferences." Bartels, supra note 89, at 299.

92. For example, although the court in the In re Mexico Money Transfer Litigation retained



88 INDIANA LAW REVIEW [Vol. 37:65

defendant to the extent that class members do not collect payment. Nor are
"smooth sailing" agreements, in which defendants agree not to oppose class
counsel's attorney's fee motion, forbidden.

Courts should ensure that an absence of objections is the result of well-

educated class members being satisfied, rather than the result ofobjections being
silenced.^^ This means encouraging objections, because class counsel cannot be
relied upon to provide the court with necessary information to evaluate the

settlement.^'* For example, in the General Motors Pick-Up Truck Fuel Tank
Litigation, a settlement was nearly approved that would have provided class
members with essentially non-transferable coupons for $ 1 000 offofthe purchase
of a new GM truck.^^ These coupons disfavored the two groups most likely to
own the offending truck, class members who lack the funds to buy a new truck.

jurisdiction over the settlement to review the injunction consented to by the defendants, the court

did not order the parties to return to demonstrate that the settlement had in fact achieved the

predicted results. 164 F. Supp. 2d 1002, 1033-34 (N.D. 111. 2000). Such decisions can create an

incentive to exaggerate the predicted settlement amount to increase class counsel's fee award where

attorney's fees are calculated on a percentage of the fund. For example, in Roberts v. Bausch &
Lomb, Inc., No. CV-94.C-1 144-W (N.D. Ala. filed Nov. 1, 1994) discussed in Hensler et al.,

supra note 2, at 1 45-73, the total settlement was presented to the court as approximately $68 million

depending on how many class members claimed rewards. The attorneys were paid $8 million based

on that representation, calculating a fee of a little less than fifteen percent. The parties were not

required to report the ultimate payout to the court, but based on SEC filings, RAND researchers
deduced that the defendant never allocated more than $37.7 million to pay out all expenses relating

to the litigation, including attorney's fees and administration costs. Thus there was likely a

substantial overpayment to class counsel.

93. For example, in Roberts v. Bausch & Lomb, Inc., there were no objectors. Id. Two other
lawsuits had been filed in federal court in California and state court in New York, but rather than
yielding objectors, the competing suits were settled for undisclosed amounts ofattorneys' fees. See

Hensler et al., supra note 2, at 158-61. In that case, absent class members had no means of

organizing an opposition or improvement to the settlement. This lack ofparticipation may have had

an effect on the settlement and on the award of attorneys' fees. The RAND study estimated that
class members received a little over 30% ofthe fund. Id. at 429. See also Koniak & Cohen, supra
note 9, at 1 083-84 (describing BancBoston settlement with only one objector where class members

were required to pay attorneys' fees in excess of their individual recoveries).

94. Class counsel and defendants' counsel cannot be relied on to point out injustice—they
are equally likely to "put one over on the court." Kamilewicz v. Bank of Boston Corp., 100 F.3d

1348 (7th Cir. 1996) (Easterbrook, J., dissenting from denial of rehearing en banc), quoted with

approval in Amchem Prods., Inc. v. Windsor, 1 17 S. Ct. 223 1, 2249 (1997). See In re Ford Motor

Co. Bronco II Prod. Liab. Litig., 1995 U.S. Dist. Lexis 3507, at *19 (E.D. La. Mar 15, 1995)

(settlement rejected as having a value of "effectively zero" based on involvement by objectors);

Bloyed v. Gen. Motors Corp., 881 S.W.2d 422 (Tex. App. 1994), aff'dandremanded,9\6S.'^.2d

949 (Tex. 1996) (Texas Supreme Court's rejection of settlement as unfair based in large part on

brief filed by objectors. Public Citizen Litigation Group and Center for Auto Safety).

95. See In re Gen. Motors Pick-Up Truck Fuel Tank Prod. Liab. Litig., 55 F.3d 768 (3d Cir.

1995).



2003] CLASS ACTION GOVERNANCE 89

and local governments that had fleets of trucks but were not likely to buy more
during the applicable period.^^ The involvement of objectors and their appeal
resulted in the settlement being overturned.^^ Thus, where responsible objectors

have a serious and active role in the fairness hearing, they can often influence the

result substantially.

Although in most cases the problem is too little objector involvement,

objectors can also be the source ofproblems. Objectors' counsel have a perverse

incentive to drop legitimate objections or soft-pedal them in order to obtain
remuneration from the settlement because, to the extent that they are paid,

objectors' counsel only receive payment if a settlement is approved.^^ As a
result, objectors' counsel may try to use their leverage to extract undeserved rent
from class counsel. Other times, objectors only marginally improve settlements

but in a manner that also resembles self-dealing.^^ Fears ofobjectors multiplying

proceedings or using their leverage to extract rent from class counsel have led

some courts to impose significant sanctions against objectors. '°^ As an empirical

96. See Wolfman & Morrison, supra note 5, at 472-77 (discussing the GM coupon
settlement). Coupon settlements raise significant problems beyond the scope of this paper.

Opponents of coupon settlements see coupon settlements as subject to serious abuse, especially

when the coupons appear to be an advertisement for the defendant, and argue that consideration of

whether a secondary market can be created in coupons should be required. Id. Other scholars

believe that problems with coupon settlements can be overcome and that non-monetary settlements

can be fair and efficient. See Miller & Singer, supra note 89. Miller testified in support of the In
re Mexico Money Transfer Litigation coupon settlement on behalfofthe defendants. 1 64 F. Supp.

2d at 1 1 8- 1 9 ("As Professor Miller explained to this court, a well designed coupon settlement can

provide class members with more value than a cash settlement because the defendant is likely to

be much more generous in its coupon offer.").

97. See supra note 95. The case was subsequently re-filed in Louisiana state court and the

settlement affirmed with a revision that encouraged a secondary market in the coupons. See White

V. Gen. Motors Corp., 835 So.2d 892 (La. Ct. App. 2002).

98. Most courts have held that objectors may obtain fees to the extent that their objections

benefited the class. See, e.g., Gottlieb v. Barry, 43 F.3d 474, 490-91 (10th Cir. 1994) (objector's

fees appropriate where objectors "benefitted" the class).

99. For example, the objecting attorney in the Mexico Money Transfer Litigation was able

to increase the ultimate settlement, but it is not clear how meaningful this increase was. This

objecting attorney convinced the negotiating parties to make some substantive changes to the

settlement, including adding a choice of one $6 coupon instead of only two coupons valued at

$4.25, and doubling the cypres donation from approximately two million dollars to four million

dollars. See In re Mex. Money Transfer Litig., 164 F. Supp. 2d 1002, 1011 (N.D. 111. 2000)

(approving settlement that included cy pres fund where claimants were illegal immigrants thus

difficult to find), discussed supra note 20.

100. See Vollmer v. Publishers Clearing House, 248 F.3d 698, 711 (7th Cir. 2001). In

Vollmer, the district court levied $50,000 in sanctions—to be donated to a charity—against
objecting attorneys on the basis of evidence that the objecting attorneys had no opportunity to

dispute. Many ofthese sanctions are brought under 28 U.S.C. § 1927 (2003), which provides that

[a]ny attorney or other person admitted to conduct cases in any court of the United



90 INDIANA LAW REVIEW [Vol. 37:65

matter, these fears may be exaggerated and such sanctions may over-deter
objector involvement or chill valuable objections.

D. The Limits ofJudicial Policing

"[C]ourts have gone so far as to term the district judge in the settlement

phase ofa class action suit a fiduciary ofthe class, who is subject therefore to the
high duty of care that the law requires of fiduciaries.'"^' Nevertheless, as one
prominent class action scholar has noted, "[p]erhaps in no other context do we
find courts entering binding decrees with such a complete lack of access to

quality information and so completely dependent on the parties who have the
most to gain from favorable court action.'"^^ Ifjudges approve a settlement in
a case that they fundamentally believe is unsound, '^^ or approve a poor settlement

in a case they believe is strong,'^'* then we may rightly be concerned about
whether the judge is acting as an independent decisionmaker or grantingjudicial

States or any Territory thereof who so multiplies the proceedings in any case

unreasonably and vexatiously may be required by the court to satisfy personally the

excess costs, expenses, and attorneys' fees reasonably incurred because ofsuch conduct.

Id. See also In re Prudential Ins. Co. Am. Sales Practice Litig. Agent Actions, 278 F.3d 1 75, 1 93

(3d Cir. 2002) (upholding $100,000 in sanctions against objecting counsel and reversing on due

process grounds a "scarlet letter" sanction requiring that objecting counsel attach documentation

ofthe sanctions proceeding to all futurepro hac vice applications). The Court upheld the sanctions

against the objecting counsel despite the lack of express finding of bad faith below. See id. 194

(Rosenn, J., dissenting).

101. Reynolds v. Beneficial Nat'l Bank, 288 F.3d 277, 280 (7th Cir. 2002) (Posner, J.)

(overturning settlement of class action on grounds that district court did not scrutinize sufficiently

whether the settlement was collusive and failed to quantify the net expected value of continued

litigation). See also In re Warner Communications Sec. Litig., 798 F.2d 35, 37 (2d Cir. 1986) ("a

district court has the fiduciary responsibility ofensuring that the settlement is fair"); In re Cendant

Corp. Litig., 264 F.3d 286, 296 (3d Cir. 2001) (same).

102. Issacharoff, .ywpm note 6, at 808.

103. See, e.g.. In re Mex. Money Transfer Litig., 267 F.3d 743, 748 (7th Cir. 2001) ("Were

the class's claims worth more than $40 million, plus cypres relief, plus the value ofthe injunction?

Like the district court, we think not—indeed, we think that the claims had only nuisance
value ").

104. Compare In re Rhone-Poulenc Rorer Inc., 51 F.3d 1293 (7th Cir.), cert, denied, Grady

v. Rhone-Poulenc Rorer Inc., 516 U.S. 867 (1995) (denying certification of blood products class

as unmanageable), with In re Factor VIII or IX Concentrate Blood Prods. Litig., 159 F.3d 1016,

1020 (7th Cir. 1998), cert, denied. Mull & Mull, PLC v. Rhone-Poulenc Rorer, Inc., 526 U.S. 1081
(1999) (affirming approval of class-wide settlement of blood products claims and labeling

settlement of individual cases with different levels of damages "downright weird"). This case is

thoroughly described in Hensler et al., supra note 2, at 293-317. The contrast between Judge

Posner's opinion in the Blood Products Litigation, and Judge Posner's more recent decision in

Reynolds v. Beneficial National Bank, 288 F.3d 277, 280-81 (7th Cir. 2002), in which he asserted

that judges have a fiduciary duty towards class members, is stark.



2003] CLASS ACTION GOVERNANCE 9

1

imprimatur to an inadequate result, perhaps in violation of her fiduciary duty to

the class.

Thejudge's role in class actions ranges from the traditional responsibility to
evaluate and determine cases, to the more modem responsibilities to manage the
interaction of various actors, monitor agents, determine the extent of plaintiff

participation, and encourage the case's resolution. '^^ Checks on district judges

making fairness determinations are limited by the deferential standard of review
employed by the appellate courts. '^^ Although judges are required to accept or
reject settlements as they are, there is the possibility for judges to reject

settlements and provide guidance to the parties as to what a fair settlement might

look like.'°^ The very things that make ajudge a more experienced arbiter ofthe
fairness of a class action settlement—her experience in the difficulty of crafting
it—may influence her to approve the settlement in order to complete a long and
arduous litigation and/or negotiation.'^^ If a judge is inclined to involve herself

and assist in the crafting of a fair settlement, this may create a bias on the part of
the judge in favor of the settlement that she worked to create.

Under the current regime, judicial accountability for the settlement results
in an unreliable governance structure. The extent to which judges police class
actions is largely a matter of the individual judge's choice, influenced by
traditional views of the attorney-client relationship and settlement as a form of

private ordering. Not only will judges not do enough to review settlements, there
is little incentive for continued judicial oversight in the administration phase.

There is no requirement in the Federal Rules for judicial review of settlement

administration. Although judges retain jurisdiction over settlements of class

actions until administration is complete, whether the settlement administration

will be reviewed or a final report issued is at the discretion of the individual

judge or party request. '^^ This lack of uniformity is compounded by incentives
forjudges to approve settlements and not to maintain active oversight over their

subsequent administration. Judges understandably may choose to focus their

105. One example of this is Judge Weinstein's case resolution methods, described in PETER

SCHUCK, Agent Orange On Trial: Mass Toxic Disasters in the Courts ( 1 987).
106. 5'eeJoelA. V.Giuliani, 218 F.3d 132, 139 (2d Cir. 2000) (standard of review of fairness

determination is abuse of discretion).

107. "[I]t is not a district judge's job to dictate the terms of a class settlement; he should

approve or disapprove a prosed agreement as it is placed before him and should not take it upon

himself to modify its terms." In re Warner Communication Sees. Litig., 798 F.2d 35, 37 (2d Cir.

1986). See also Hanlon v. Chrysler Corp., 150 F.3d 101 1, 1026 (9th Cir. 1998) ("The settlement

must stand or fall in its entirety.").

108. &e, e.g., Twelve John Does V. District ofColumbia, 1 17F.3d571, 576(D.C. Cir. 1997)

(approving a settlement under the abuse of discretion standard and finding that "the district court's

experience overseeing the case for nearly two decades" gave it "a unique familiarity with the issues

and the performance of class counsel").

1 09. See Tyson v. City ofNew York, 97 Civ. 3762 (JSM), Submission ofProposed Stipulation

of Settlement (Jan. 5, 2000) (on file with author) (requiring that administration of claims be

completed before attorneys fees are awarded and providing for interim fees).



92 INDIANA LAW REVIEW [Vol. 37:65

limited resources on cases currently being adjudicated. The prospect of
continuing oversight when the parties have agreed to a settlement may strike
some judges as superfluous.

Accordingly, judges are both a source of hope for class action governance,

in that they are a ready-made monitor of self-interested class counsel, and a

source of frustration, in that the protections afforded class members by judicial
oversight are dependent on the discretion of the individual judge.

II. Proposed Models OF Class Action GOVERNANCE:
The Last Twenty Years

In the past twenty years, scholars have proposed various mechanisms for
solving the problems of class action governance, particularly in the settlement

phase where agent-principal problems are most acute and where judicial
discretion plays a significant role. The proposed solutions to the various
problems posed by class actions fall into three general categories: market

solutions,"^ democracy-based solutions,'" and judicial administrative

solutions."^ Each of these offers significant contributions to aspects of class

action problems, but each is fundamentally incomplete.

A. Market Solutions

Scholars have proposed numerous market mechanisms to solve the problems

posed by small claims class actions. In looking for an analogy to class actions,

some scholars have turned to corporate law."^ As in the corporate context, this
theory maintains, the problem with class actions is an agent-principal problem, '

'"^

no. See, e.g.. Coffee, Accountability, supra note 3, at 370 (advocating additional opt out

mechanisms, i.e., "exit").

in. See, e.g., Deborah L. Rhode, Class Conflicts in Class Actions, 34 STAN. L. REV. 11 83,

1 185 (1982) (proposing "a theory of representation mandating full disclosure of, although not

necessarily deference to, class sentiment").

112. See, e.g. , Wolfman & Morrison, supra note 5, at 439 (proposing amendments to Rule 23
that increase judicial scrutiny of settlements and create more substantive guidelines for evaluating

settlements); Shapiro, supra note 21, at 917, 922-23 (advocating an "entity" approach to class

actions).

113. See, e.g., Charles Silver & Lynn Baker, / Cut, You Choose: The Role of Plaintiffs'
Counsel in Allocating Settlement Proceeds, 84 Va. L. Rev. 1465 (1998) (making an explicit

analogy to the corporate context).

1 1 4. See Resnik et al., supra note 24, at 374 (describing how economic analysis of litigation

has been based on the agent-principal problem); Coffee, Accountability, supra note 3, at 375

(describing class action as an "organizational form" involving a "principal/agent relationship").

But see Coffee, Understanding the Plaintiffs Attorney, supra note 24, at 683-84.

Yet for analytical purposes, one better understands the behavior of the plaintiffs

attorney in class and derivative actions ifone views him not as an agent, but more as an

entrepreneur who regards a litigation as a risky asset that requires continuing investment

decisions. Furthermore, a purely fiduciary perspective is misleading because it assumes



2003] CLASS ACTION GOVERNANCE 93

arising from the fact that the class members' ownership of their claims is
separated from control over those claims. Accordingly, scholars urge that class

action reform requires policy-makers to "consider market-based remedies and

checking mechanisms that have worked in related contexts to align the interests
ofthe principal and the agent."^'^ The two most prominent market mechanisms
proposed in the class action literature concern the structure of attorney fee

awards and the availability of opt outs."^

1. Attorneys ' Fees.—Monitoring class counsel is central to class action
governance. Legal economists have focused their attention on the incentive

structure for class counsel to "pursue their own interests at the expense of the
class" resulting in cheap settlements."^ As courts and commentators have noted,
class counsel have an incentive to engage in self-dealing regardless of their

ethical obligations and duties to the class. "^ Some scholars propose creating

that the cHent's preferences with respect to when an action should be settled are

exogenously determined, when, in fact, they are largely influenced by the fee award

formula adopted by the court.

1 15. Coffee, Accountability, supra note 3, at 371 . Under this view, class members may be

deemed to have consented to representation by the attorney where "the agency costs associated with

the relationship have been minimized." Id. at 376. There have been numerous propositions for

methods of aligning class counsel's interests with those of the class. See, e.g., Christopher R.

Leslie, A Market-BasedApproach to Coupon Settlements in Antitrust and Consumer Class Action

Litigation, 49 UCLA L. Rev. 991, 1088 (2002) (advocating that class counsel be paid in the same
currency as the class).

1 16. Some scholars have even proposed to do away with the class members altogether by

auctioning offtheir claims to the highest bidder. See Macey & Miller, supra note 7, at 6 (proposing
a class action regime whereby claims are auctioned off to attorneys). This proposal has generated

an interesting scholarly discussion. See, e.g., Randall S. Thomas & Robert G. Hansen, Auctioning
Class Action and Derivative Lawsuits: A Critical Analysis, 87NW. U. L. REV. 423, 424-25 (1993).

Thomas & Hansen only address auctions of the right to represent the class, not auctions for
purchase of class claims.

1 17. Staton v. Boeing Co., 313 F.3d 447, 467 n.l2 (9th Cir. 2002), withdrawn and reh'g

denied, 327 F.3d 938 (9th Cir. 2003).

Even when there is no direct proof of explicit collusion, there is always the possibility

in class action settlements that the defendant, class counsel, and class representatives

will all pursue their own interests at the expense of the class. For that reason, the

absence of direct proof of collusion does not reduce the need for careful review of the

fairness of the settlement, particularly those aspects of the settlement that could

constitute inducements to the participants in the negotiation to forego pursuit of class

interests.

Id. See also Coffee, Understanding the Plaintiff's Attorney, supra note 24, at 669.

118. Courts have held that class counsel owes a fiduciary duty to the class as a whole. See In

re Gen. Motors Corp. Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 801 (3d Cir. 1 995)

("Beyond their ethical obligations to their clients, class attorneys, purporting to represent a class,

also owe the entire class a fiduciary duty once the class complaint is filed."); Stewart v. Gen.

Motors Corp., 756 F.2d 1285, 1294 n.5 (7th Cir. 1985) (citation omitted) ("[B]oth the class



94 INDIANA LAW REVIEW [Vol. 37:65

competition among attorneys for the privilege of representing a class. "^ Others
propose giving a powerful plaintiffthe ability to negotiate a market rate on behalf

of the class. '^^ Still others propose fee regimes that reward class counsel

incrementally, based on the actual benefits conferred to the class,'^' or attempt

to solve the problem by subsidizing rival class actions amongst which class

members can choose. '^^

Ofthe proposals put forward, the one most consistent with accepted attorney
compensation norms is to link attorneys' fees to the amount of benefit the
attorney provides the class, with a marginally increased percentage of the fund

going to the attorney to encourage the maximization of claimholder value.
'^^

Large fee awards are beneficial because they create an incentive for attorneys to

bring small claims suits as private attorneys general, and it makes sense to link
that incentive directly to the benefits conferred to the class. '^"^ Yet, it also makes
sense to limit percentage fee awards so that they bear some relationship to the
effort expended by the attorneys and do not become a windfall to attorneys at the
expense ofthe class. The problem is that there is little empirical evidence to help
judges determine the appropriate percentage of the fund to give to attorneys to

attract responsible and experienced attorneys to bring class action suits and to

compensate for the risk that they incur in taking on such suits, without offending

sensibilities. Setting a loadstar cap on attorneys' fees would mitigate the concern

that attorneys receive a windfall, but even setting such a cap is difficult without

more information on how the balance should be struck.
Another solution that has been much discussed in the literature is auctioning

determination and designation of counsel as class representative come through judicial

determinations, and the attorney so benefited serves in something of a position of public trust.").

1 19. See Macey & Miller, supra note 7, at 6 (proposing auctions for the position of class
counsel); Koniak & Cohen, supra note 9, at 1 1 1 3- 1 4 (advocating same and arguing that the absence
of auctions may constitute an antitrust violation).

1 20. Jill E. Fisch, Lawyers on the Auction Block: Evaluating the Selection ofClass Counsel

By Auction, 102 COLUM. L. Rev. 650 (2002) (criticizing auctions and proposing stronger lead

plaintiffs as an alternative solution to the agent-principal problem).

121. See Coffee, Understanding the Plaintiff's Attorney, supra note 24, at 725. Compare

Third Circuit Task Force Report on Selection ofClass Counsel, 74 TEMP. L. REV. 689, 704-05

(200
1 ) (advising that "[t]he traditional methods ofselecting class counsel, with significant reliance

on private ordering, are preferable to auctions in most class action cases. In using those traditional

methods, however, the court must guard against overstaffing by lawyer groups."), with John C.

Coffee, Jr., Litigation Governance: A Gentle Critique ofthe Third Circuit Task Force Report, 74

Temp. L. Rev. 805, 808 (2001) [hereinafter Coffee, Litigation Governance] (arguing that "private

ordering works least well when agency costs are high and competition is limited." Also noting that

"as a general rule, professional groups have little incentive to seek increased competition or lower

fees, and professional ethical norms generally are enforced only against outliers and insurgents.").

1 22. See Coffee, Accountability, supra note 3, at 424.

1 23. See Coffee, Understanding the Plaintiff's Attorney, supra note 24, at 725.

124. Seeid2Ml\'25.



2003] CLASS ACTION GOVERNANCE 95

off the right to represent a class.
'^^

In essence, the auction proposal entails

potential class counsel submitting proposals for representation and fees at the

beginning of the suit, and the court awarding the case to the lowest bidding

counsel. '^^ This proposal may reduce attorneys' fees, but will do little to
eliminate agency costs. Attorneys may "settle early in order to obtain a larger
profit on the fee'"^^ or, if they are obtaining a percentage of the total fund,

exaggerate the value of the settlement to increase their compensation. An
auction may produce lower attorneys' fees, but may not obtain the best quality
or most experienced counsel.

'^^ As their proponents concede, they cannot
eliminate agency costs associated with the attorney-class relationship.'^^ Thus,

on their own terms, attorneys' fee mechanisms are an incomplete solution to class
action problems.

The agent-principal analysis of class actions has led to an overemphasis on
mechanisms to control attorneys' fees. An attorneys' fee solution to class action
governance is analogous to a corporate governance scheme based solely on
controlling executive compensation. The observation that various forms ofself-
dealing, including "cheap settlements," are the likely result of a flawed fee

regime is accurate, but it captures only one aspect of the weak governance
structure of class actions.

2. Opt Outs,—Opt outs are the only exit mechanism available to class
members. Recognizing that the absence of competition diminishes the

significance ofopt outs, John Coffee has proposed the creation ofa market in opt

125. See Macey & Miller, supra note 7, at 105-18 (describing proposals for auctioning ofT
claims and for auctioning rights to represent the class).

126. See, e.g.. In re Oracle Sees. Litig., 131 F.R.D. 688, 689-90 (N.D. Cal. 1990) (ruling that

competitive bidding would determine selection ofclass counsel and counsel's compensation). See

also Koniak & Cohen, supra note 9, at 1202-06 (arguing that market for representation of class
should be competitive and that current system may be an antitrust violation). The revised Rule

23(g) offers something in this regard by permitting the court to choose class counsel. In a twist on

this idea, Geoffrey Miller has also proposed auctions for class counsel after settlement has been

reached, permitting rival class counsel to post a bond for the settlement amount and attempt to

negotiate a better deal for the class. See Geoffrey P. Miller, Competing Bids in Class Action

Settlements, 31 HOFSTRA L. REV. 633 (2003).

127. Macey & Miller, supra note 7, at 1 13.
128. M at 1 13; Lucian Arye Bebchuk, The Questionable Case for Using Auctions to Select

Lead Counsel, 80 WASH. U. L.Q. 889, 892 (2002); Coffee, Litigation Governance, supra note 121.

See also Fisch, supra note 1 20, at 652 (arguing that auctions are poor tools for selecting firms based

on multiple criteria, that auctions compromise thejudicial role, and that they are unlikely to produce

reasonable fee awards as an empirical matter).

1 29. It does not seem rational that an attorney who has purchased the rights to a class action

at a slightly lower fee will take the nuances of class interests into account any more than the

attorney entitled to potentially exorbitant fees. One might even think that the auction attorney is

less likely to take class desires seriously, so long as they do not affect the ability ofthe class action

settlement to be approved or otherwise interfere with the total recovery, because he has an even

greater sense of ownership over the lawsuit than the "joint venturing" class action attorney did.



96 INDIANA LAW REVIEW [Vol. 37:65

outs rather like a hostile takeover in the corporate context.
'^^ When a settlement

has been proposed, attorneys who wish to represent dissenting class members
would be permitted to include a "counter-solicitation" in the notice ofsettlement

sent to class members, inviting them to opt out ofthe settlement.'^' If successful,

these competing attorneys would organize dissatisfied class members and file a
competing class action.'^" According to Coffee, this approach would solve the

abandonment problem ofopt outs without threatening "the fmality ofsettlements

nor encourag[ing] collateral attacks on the class settlement.'"" It would be

relatively cost efficient because the "counter-solicitation" could be included with

the regular settlement notice.
'^"^

Finally, it would provide greater opportunity for

the substantive exercise of class member autonomy at least to the extent that the
choices presented are reasonably attractive to the class members.

To create a viable market in opt outs, the potential gain from competition
would have to be greater than the cost of the new attorney and the rival
solicitation. We might predict that competition would be limited where a
settlement is inadequate but the value that the rival attorney is able to extract

from the sub-class is not high enough to merit the investment in organizing a

rival solicitation. Without empirical analysis, it is difficult to tell whether that

result is optimal in terms of the regulatory function of small claims class actions

and whether the cost of mounting a rival action would be too high for attorney

entrepreneurs.'-^^ It is worth noting that in the securities context, when
sophisticated class members are dissatisfied with class counsel, instead of
creating rival class actions by organizing opt outs, these potential lead plaintiffs

prefer to opt out and file independent actions in state court.
'^^

Coffee recognizes some of the problems with this creative proposal,
'^^

yet

130. See CoffQQ, Accountability, supra note 3, 3X422.

131. /^. at 423.

132. Id.

133. /^.at424.

134. /^. at 423-24.

135. Even if rival attorneys are permitted to solicit opt outs in settlement notices, the cost and

risks ofpursuing such a rival class action may be too high for attorney entrepreneurs in small claims

actions. Interview with Joseph Grundfest, W.A. Franke Professor of Law and Business, Stanford

Law School (Feb. 13,2003).

1 36. See, e.g. , Cal. State Teachers' Ret. Sys. v. Qwest Communications Int'l, No. 4 1 5546 (Gal.

Super. Gt., San Francisco County filed Dec. 10, 2002), available at http://www.sftc.org/ (state

lawsuit with no class or federal claims filed to avoid class litigation); UC Sues Firms with
WorldCom Ties, L.A. TIMES, Feb. 14, 2003, at C4 (Regents of the University of California, which

had purchased 10.2 million shares of Worldcom between 1998 and 2000, pulled out of nationwide

federal class action and commenced suit on their own.). These plaintiffs bring only state court

claims to avoid removal and class adjudication.

1 37. Coffee notes that counter-solicitations create the danger that the class will become too

fragmented as a result of significant number of proxy proposals, and proposes that the problem of

too many proxy proposals might be solved by requiring insurgent counsel to pay reasonable costs

of printing and mailing proposals (as the SEC handles hostile proxy solicitations). Counter-



2003] CLASS ACTION GOVERNANCE 97

leaves many questions unanswered. How would class members intelligently
choose among the multi-factored proposals likely to arise and the associated
risks? More than likely, such rival solicitations would require opt outs to give
up the proverbial bird in the hand for a bird in the bush.'^^ Would there be
disclosure requirements that enabled class members to make these choices and,
if so, how much disclosure or education would be necessary and in what form?'^^

Second, this market model would suffer the same agent-principal problems
observed under the current regime. The incentives of class counsel and rival

attorneys would continue to be the same: to maximize their own payout by
settling early or for inadequate amounts. Rival attorneys may be bought out or
co-opted, as happens in competing class actions under the current regime. Thus,

even if there were such competition, agent monitoring would still be necessary.

Moreover, this competitive model would further hamper the ability of the court

to act as such a monitor and to evaluate the fairness of settlements because the

process of soliciting opt outs would siphon objectors away to the new action
rather than encourage them to voice their objections to the proposed settlement.
In addition, judges would have a difficult time determining whether in any
particular case a lack of competition for opt outs is the result of market

inefficiencies or an indication that the settlement is indeed fair.''*^

A rising trend that bears some relationship to the rival solicitation proposal

solicitation could also be made contingent on rival counsel getting enough support, and the court

could refuse to permit rival solicitation if too few class members sign on or otherwise fail tests of

Rule 23. See Coffee, Accountability, supra note 3, at 426. Coffee also suggests that a court could

arbitrarily limit the number of competing solicitations (to two or three) "leaving others to contact

class members on their own if they wished." Id. at 426. On the other hand, if the number of rival

solicitations are limited to avoid fragmentation, then the class may be denied the full panoply of

choices that a true market would offer. Id. Rival attorneys may experience judicial hostility to a

proposal intended to delay and complicate an existing settlement process. Id. at 427. Additionally,

there may be substantial reputational costs for prestigious class action firms to mount counter-

solicitations. In a world where class action attorneys are repeat players, it is not difficult to imagine

that there may be costs meted upon dissident counsel. As a result, the type ofexperienced attorneys

one would ideally want to run a rival class action may not be willing to create such a rivalry unless

the gain is substantial enough to outweigh the potential reputational costs.

1 38. This differentiates the proposal from the hostile takeover context where the shareholders

are offered specific sums of money for their shares: "the class member is still very much subject

to risk, while the shareholder who accepts a cash tender offer is no longer subject to the risks ofthe

enterprise." Coffee, Accountability, supra note 3, at 423 n. 140.

139. Such a rival solicitation may also face other hurdles, such as when class dissatisfaction

is based on non-monetary factors, including class counsel's relationship with the class, a desire for

greater voice, or the desire for non-monetary relief out of which it would be difficult to take a

contingency share.

140. Similarly, in the current regime it is difficult for courts to determine whether a small

number of objectors is the result of the fact that there are too many barriers to objection or of the

settlement being genuinely fair.



98 INDIANA LAW REVIEW [Vol. 37:65

is the devolution ofclass actions to state courts."*' Charles M. Tiebout's familiar
model of local government provides a useful analogy.'"*^ Tiebout theorized that
the distribution of public goods in a given community reflects local preferences
because a consumer-voter will pick the community "which best satisfies his
preference pattern for public goods.""*^ The natural corollary ofthis proposition
is that "[t]he greater the number of communities and the greater the variance
among them, the closer the consumer will come to fully realizing his preference
position.""*"* If compensation through a class action is a public good, then it

would seem that permitting class members a choice would maximize their ability
to obtain a preferential benefit from the class action mechanism. But if the

choice is a stark one between obtaining nothing (which is the normal result of

opting out in most small claims class actions) and joining an inadequate

settlement, the "choice" is only in name.

Applying Tiebout's theory to class actions, we can envision a regime that
would encourage class actions on a local scale instead of collecting claims into
nationwide class actions."*^ Local class actions would be more manageable and
actions based on state laws would not suffer from choice of law problems.
Localization would make public monitoring by the states' attorneys general
easier. It would also create a kind of competitive environment, allowing courts
to compare the terms of settlements with those of similar class actions in other
states. On the other hand, this type of "market" in class actions, demarcated
along state boundaries, does not offer class members greater choice and
autonomy. Moreover, localization would reduce the efficiencies achieved by

national class actions. Finally, further localization may not be feasible at a time
when policy makers, attorneys and judges, are increasingly enlarging class
actions, expanding defendants' ability to remove such actions to federal court and

settle them as nationwide class actions rather than permitting them to be split off

by jurisdiction."*^

The central limitation ofexit mechanisms as a sole solution to unsatisfactory
settlements is the same problem as all exit mechanisms in markets with limited
competition share: limited and ineffective opportunities to opt out benefit the

141. See supra note 44 and accompanying text (describing state class actions limited to

citizens of forum state).

142. See Charles M. Tiebout, A Pure Theory of Local Expenditures, 64 J. POL. ECON. 416

(1956). See, e.g., William W. Bratton & Joseph A. McCahery, The New Economics of
Jurisdictional Competition: Devolutionary Federalism in a Second-Best World, 86 Geo. L. J. 201

(1997) (criticizing Tiebout's model); Gerald E. Frug, City Services, 73 N.Y.U. L. Rev. 23, 25-33

(1998) (criticizing Tiebout's model among other things for the unrealistic assumption that some

people choose poor service communities and noting limits of mobility in the real world).

143. Tiebout, 5Mpra note 142, at 418.

144. Id : y :••• -:-./:.; •,•..,.

1 45. See supra note 44 (discussing nascent trend towards localized class actions).

146. One example is the Class Action Fairness Act of 2003, S. 274, 108th Cong. § 4 (2003),

discussed supra note 45. It seems clear that legislation advocating unchecked removal of class

actions to federal courts is not necessarily protective ofclass members, but rather favors defendants.



2003] CLASS ACTION GOVERNANCE 99

decisionmakers, who are spared the trouble of addressing genuine concerns
because their most vociferous opponents have left the stage. ''^^ It is more
beneficial for the architects ofa class-wide settlement to permit some opt outs as
a kind of safety valve to get rid of dissenting voices, than to allow a realistic

option that might scuttle settlement. Thus we often see courts validating
settlements on the basis that class members had an opportunity to opt out and
chose not to do so, without looking to the role ofthe opt out as a mechanism for
minimizing dissent.''*^ Like attorneys' fee regimes, exit mechanisms alone are
not a satisfactory solution to the governance problems in class actions.

These types ofmarket mechanisms ignore a central difference between agent

principal problems in the class action context as opposed to the corporate

context. For claimants the class action is a single, unique transaction, and

subsequent market penalties for singular instances of mismanagement are
inadequate. '"^^ The complex problems in class action governance cannot be
resolved by resort to a single incentive mechanism because class actions are not
merely a function of the plaintiffs' attorney. A simple market model would not
work even if there were vigorous competition between class counsel bringing
identical class actions—which in itself creates obvious inefficiencies—because
the panacea of attorney competition overlooks the role of other actors in class

action governance, including judges, class members, objectors, and opt-outs.

These observations represent a significant hurdle to any market solution to class

action problems.
'^^

B. Democratic Solutions

Another source for class action governance mechanisms is democratic

theory. Rule 23(b)(3) provides a number of mechanisms that resemble political

1 47. "Those who hold power in the lazy monopoly may actually have an interest in creating

some limited opportunities for exit on the part of those whose voice might be uncomfortable."

HiRSCHMAN, supra note 60, at 60. Some limited competition may actually comfort monopolist by

taking away the most vocal customers. Id. at 59.

1 48. See discussion supra at note 73 (discussing court treatment ofopt outs and objections in

settlement context); see also Devlin v. Scardelletti, 536 U.S. 1, 122 S. Ct. 2005, 2011 (2002)

(noting that because class member was denied opt out, appeal gained greater importance for him

to protect himself).

149. See FRANK H. Easterbrook & Dandel R. Fischel, The Economic Structure of
Corporate Law 103 (1991) (discussing the ineffectiveness of market mechanisms on singular

breaches of fiduciary duty in the corporate context). In the corporate context, voting and

purchasing mechanisms, as well as boards ofdirectors, discipline management. Shares are fungible,

and regulation attempts to create transparency of information on which shareholders base their

investments. That is simply not the case in non-fungible small claims class actions where claimants

do not "opt in"; there are no disclosure mechanisms and only the judge, who is not elected by the

claimants, disciplines class counsel.

1 50. This is not to say that market solutions cannot cure some of what ails class actions, but

that market mechanisms alone do not provide a complete governance system.



100 INDIANA LAW REVIEW [Vol. 37:65

devices. The class as a group has a representative in the proceedings.'^' The
fairness hearing provides a forum where the merits of settlement proposals are
presented and discussed prior to the judge's decision. In deciding whether to
approve a settlement, judges take into account certain displays of popular will

from the class, such as numerous opt outs or objections to the settlement. '^^

Scholars have proposed democratic governance structures for class actions

in the context of civil rights injunctive actions.
'^^

However, deliberative

solutions have not been a prominent part of the discourse concerning consumer
class actions. This may be because consumer class actions are viewed more as
an economic matter than a political one, whereas civil rights actions relate
directly to the political realm. Viewed in this light, it may seem hypocritical to
permit experts to run a civil rights action in order to vindicate democratic values,

particularly when the underlying claim relates to some form of
disenfranchisement from the political or social sphere. Perhaps deliberation in

the civil rights context is seen to "rectify the antidemocratic exclusion of
chronically disadvantaged groups from the theatre of politics.'"^'*

Democratic values have relevance, however, even to complaints based on
economic harms. The observation that "[d]emocratic decisionmaking is an
attractive alternative to unrestrained liberty because it provides a means for

reining in the self-appointed community representative; it also checks the
alienation and disempowerment that result from overreliance on lawyers'" ^^

applies just as well to the consumer class action context. The mandatory nature
of injunctive civil rights actions is not a sufficient reason to limit democratic

proposals to them because opt outs do not provide a true alternative in the small

claims context. Furthermore, there is nothing inherent in consumer issues, as

opposed to civil rights issues, that render them less amenable to class member
input into their resolution. Although consumer class actions generally address

small individual claims, they are nevertheless significant protective and

preventive mechanisms in a world where interactions with economically and

politically powerful corporations are such a pervasive aspect of people's daily

151. See Fed. R. Crv. P. 23(a)(4). Although it is not clear what "representation" means in this

context, as discussed below, this provision indicates that some importance is attached to the voice

of class members.

1 52. See supra text accompanying note 65-70 (discussing the judicial approach to opt outs)

and note 72-73 (discussing the judicial approach to objectors).

153. Examples of discussions of the role of democratic decision making processes in civil

rights class actions include Rhode, supra note 1 1 1, at 1 191-92 and Rubenstein, supra note 14, at

1663-68. My research has not revealed substantial scholarship making a consistent argument for
a democratic approach of any kind to class actions outside the area of civil rights injunctive class

actions. Cf. Elizabeth J. Cabraser, Enforcing the Social Compact Through Representative

Litigation, 33 CONN. L. REV. 1239 (2001) (discussing the representative litigation as an aspect of

the realization of the American social contract).

154. Rubenstein, supra note 14, at 1659 (quoting Jane S. Schacter, Metademocracy: The

Changing Structure ofLegitimacy in Statutory Interpretation, 1 08 Harv. L. Rev. 593, 6 1 9 ( 1 995)).

155. /^. at 1659-60.



2003] CLASS ACTION GOVERNANCE 101

life.

In the mass tort context, scholars' discussions of participation values have

been limited to support or rejection of the idea of universal, particularized

adjudication.'^^ However, the argument that small claims (or even mass tort)
plaintiffs do not have a process right to individual participation that trumps the

right of the public or defendant to conserve resources,'" as one prominent

scholar has argued, should not necessarily mean that policy makers should ignore
methods for collective input from the class into the settlement process. The
question is whether it is possible to realize democratic values in the small claims

class action in the absence of individualized adjudication.

It may be inherently good for collective settlements to reflect democratic
processes because they are political; their regulatory role makes them so.'^^
Furthermore, settlements can be preference forming as well as preference

expressive vehicles. They can create expectations as to the class' specific claims
and future suits.

'^^
Just because the class action settlement is a transaction does

not mean that it needs to be viewed as a purely private transaction with no
political dimension or as limited to class counsel and defendants.

7. Representation.—The Rule 23(a)(4) requirement of "adequate
representation" raises an obvious comparison to representative democracy.

Although self-appointed representatives are more in keeping with dictatorial
systems of government than with representative democracy, '^^ as political

theorists remind us, representation "need have nothing to do with popular self-

government.'"^' A representative may merely be someone with the authority to
act on behalf of her constituency.'"

Scholars and courts have struggled to create a thicker description of what

1 56. See Rosenberg, ^wpra note 19, at 21 1-12 (arguing against the idea that self-determination

requires individualized litigation); see also Judith Resnik, From "Cases " to "Litigation, " 54 LAW
& CONTEMP. Probs. 5, 9-14 (1991) (discussing the 1966 Advisory Committee's concerns about
applying class action to mass torts).

157. See Rosenberg, supra note 19, at 213 n.6.

158. Frank Michelman, Law'^7?^/7m6//c, 97 Yale L.J. 1493, 1503 (1988) (discussing political

participation as a positive human good).

1 59. See Alexander, supra note 22, at 500 (criticizing settlements in securities class actions

that are settled at a "going rate" rather than based on the strength of the parties' claims).

160. See Fiss, supra note 25, at 25.

161. Hanna F. PrPKiN, The Concept of Representation 2 ( 1 972).

1 62. To the extent one can argue that class representatives represent at all, it is in the Burkean

conception of the representative as a trustee without accountability. As Burke wrote, "[t]he king

is the representative of the people; so are the lords; so are the judges. They are all trustees for the

people." Quoted in PiTKiN, supra note 1 6 1 , at 1 29. Under the current damages class action regime,

the representative appears to lack both the authority to act, because the lawyer is in fact the

representative with authority, and has no accountability to the class membership. It is not clear,

under this scenario, where the content of the notion of "class representative" is to be derived. A
class representative chosen by an attorney or self-appointed, without the power to make decisions,

is a "representative" only in the thinnest sense.



102 INDIANA LAW REVIEW [Vol. 37:65

role representatives take in the class context and of the source, if any, of their

legitimacy. '^^ Some scholars have defined the legitimacy of the representative
in the class context as inhering in her ability to be a "mirror" of the class. '^"^ She
is a representative in the sense that she has suffered the same injury or asserts the
same claims as absent class members. This view stands behind the logic of the
cohesion theory of representation.'^^ Cohesion theory assumes that ifthe claims

are identical, the class representative will necessarily follow the intent of the

class as a whole.
'^^

This conception is unsatisfactory for several reasons. First, it assumes that

the representative is imbued with the power to control the litigation. Under our

163. See Rhode, supra note 111, at 1191-92. Rhode describes how, because of class

certification, the requirement that representative parties will fairly and adequately represent the class

is of constitutional dimension, but that this requirement does not tell litigants what it means to

"adequately protect" the "interests" ofthe class and asks, "Do the named representative and counsel

serve primarily as 'instructed delegates' pursuing objectives to which a majority of class members

subscribe?" Id. at 1 192. Or, in the alternative, "does the representative role track Edmund Burke's

notion of an 'enlightened trustee,' who makes an independent assessment of class concerns?" Id.

at 1 1 92-93. She concludes that courts have "done little more than acknowledge the absence ofany

'clear principles governing the allocation ofdecisionmaking authority between the attorney and the

class.'" Id. at 1193. Twenty years later, this appears to still be the case. See also Coffee,

Accountability, supra note 3, at 373 (critiquing the Supreme Court's treatment of the notion of

representativeness and class cohesion as overly formalistic). The Burkean notion of the

representative as a trustee with special expertise (that is, the lawyer) is often rejected as overly

paternalistic and not in keeping with ideas of litigation as an autonomous act. For a sustained

discussion of the expertise model in this context, see Rubenstein, supra note 14, at 1 163-68. For

a thoughtful historical discussion of the relationship between representation and participation, see

Robert G. Bone, Personal and Impersonal Litigative Forms: Reconceiving the History of

Adjudicative Representation, 70 B.U. L. REV. 213, 297-98 ( 1 990) (reviewing STEPHEN C. Yeazell,

From Medieval Group Litigation to the Modern Class Action) (concluding that 23(b)(3)

class actions in some ways resemble participation-based representative suits).

1 64. See Coffee, Accountability, supra note 3, at 375 (discussing the representative as "mirror

image").

165. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 621 (1997) (whether class should

be divided into multiple, independently represented sub-classes depends on whether the class "has

sufficient unity so that absent members can fairly be bound by decisions of class representatives").

This conception originates in the requirements ofcommonality and typicality found in Rule 23(a).

Although on its facts Amchem is a mass torts case, its language provides an overarching ruling

concerning the basis of representation in class actions generally.

1 66. It is not clear how proponents ofthe cohesion principle believe that the translation from

the individual representatives interests to group interests occurs—whether the representative's
decisions will reflect the decisions ofeach individual member or will reflect the range ofdecisions

that the groups many members would have made. What is clear is that the cohesion principle is

based on the theory that the right of representation is "not a day in court but the right to have one's

interest adequately represented." Owen M. Fiss, The Allure ofIndividualism, 78 lOWA L. Rev. 965,

970-71 (1993).



2003] CLASS ACTION GOVERNANCE 103

current regime, the real authority to make decisions on behalf of the class rests
with the class' attorney. '^^ Significant additional oversight mechanisms would
have to be instituted to make the class representative an effective monitor of
attorney conduct. Second, it requires a very high level of identity between the

representative and the class. Such a level of identity may not be entirely
possible, nor entirely necessary to good governance. It is difficult to draw the
1 ine between those individual qual ities ofclass members that require identity and
those qualities that do not matter for purposes of representation. As a corollary,
the Rule 23 typicality requirement, together with the representation requirement,

places reliance on the court's ability to determine that the class representative is

in fact like the class in all material ways and will make rational choices on their
behalf This determination is, to put it mildly, an inexact science and no

provision is made for situations in which the representatives' differences from
other members of the class subsequently become material.'^* Finally, the
cohesion principle attempts to define away the problem of inter-class discord by
removing from the discourse the individuals who may disagree.

It is difficult to imagine a class achieving this strong coherence requirement

and still retaining the efficiency of the class action device.'^^ Cohesion and
efficiency are in tension with one another. The looser the concept of cohesion,
the more efficiently claims can be disposed of through the class action

mechanism. The tighter the concept of cohesion, the more difficult it is to forge
a class action. There are two equally unpalatable ways to achieve cohesion:
remove individuals with different interests from the class, thereby narrowing it
to a near impossibility, or redefine their interests in a cribbed and unrealistic

way. Either way, it does not provide an underlying rational for the legitimacy of

any one particular group identity or for the particular representative appointed to
defend its interests.

2. Deliberation.—Some scholars have turned to democratic deliberation as
a model for listening to class members. In exploring the idea of deliberation in

the class context, Deborah Rhode presented two models: pluralism and
majoritarianism.'^^ The pluralism model consisted of individualized
representation for discrete constituencies within the class. '^' As commentators
have noted, successive sub-classification has the potential to fragment the class

1 67. See supra note 6 (describing the view that class counsel is the true architect of class

actions).

168. For example, what if the representative is lazy, risk averse or a risk seeker? See Coffee,

Accountability, supra note 3, at 375 (discussing potential problems with representative).

1 69. See discussion supra, accompanying notes 1 7-25 (discussing the efficiency basis for class

action litigation).

1 70. See Rhode, supra note 1 11 , at 11 85.

171. See id. This is similar to the model that was eventually adopted by the Supreme Court

in mass tort cases under the guise of adherence to the cohesion principle. See Ortiz v. Fibreboard

Corp., 527 U.S. 815, 856 (1999) (requiring sub-classification where sub groups within class have

conflicts of interests).



104 INDIANA LAW REVIEW [Vol. 37:65

action out of existence. '^^ On the other hand, this model also has the potential
to create small, more manageable and responsive, class actions. Without more
development as to what governance of sub-classes means, this proposal suffers
from the same problems ofthe cohesion principle; it assumes that a smaller scale
will cure the multiple problems of class actions.

The majoritarian model would ascertain members' preferences through polls
or plebiscites.'^^ But within this model, some entity must still shape and ask the
questions and oversee the referendum process. '^"^ Majoritarianism also assumes

that preferences are clear and ascertainable, not contentious, disparate and

dynamic, and just need to be "collected," although this conflicts with most
people's experience. '^^ Furthermore, there are practical barriers to plebiscites,

such as the natural apathy of class members with small stakes in the litigation '^^

and the cost ofvoting mechanisms as well as voter education. '^^ Commentators

172. See Coffee, Accountability, supra notQ 3, 2A 314.

173. 5^eRhode,5Mp/'a note 111, at 1185.

1 74. Not every aspect of a case is amenable to voting or deliberation, such as purely strategic

legal decisions that would be made by the attorney alone in a traditional attorney-client relationship.

See Rubenstein, supra note 14, at 1654 ("Democratic values would lend significant legitimacy to

goal-based decisions, but would have less applicability to the more technical decisions about legal

strategies."). Furthermore, as Rhode herself points out, public choice models demonstrate how

agenda setting can shape results. See Rhode, supra note 1 1 1, at 1 236. See also Rubenstein, supra

note 14, at 1656-57 (discussing hurdles to voting). In addition, there is what Rhode called the

"unarticulated premise" against majority vote, that "[t]he class as an entity has interests beyond

those expressed by its current constituents." Rhode, supra note 1 1 1, at 1241. In consumer class

actions, third parties may have an interest in the outcome of the class action. For example, the

public has an interest in the adequacy of deterrence and in avoiding over-deterrence.

1 75. See discussion supra note 88-89 (describing the difficulty of ascertaining class member

preferences).

1 76. One of the most important and largely unresolved hurdles to a plebiscite is participation.

Scholars have recognized the difficulties of getting class members to respond to communications

from class counsel, especially where the stakes are relatively small. "Few class members respond

to court mailings and those who do are not representative." Rubenstein, supra note 1 4, at 1 657-58.

See also Paul D. Carrington & Derek P. Apanovitch, The Constitutional Limits of Judicial
Rulemaking: The Legitimacy of Mass-Tort Settlements Negotiated Under Federal Rule 23, 39

Ariz. L. Rev. 461, 466 n.35 (1997) ("The likelihood that a class member will actually receive and

comprehend the notice ofthe action is in every case very small. Frequently, the cost ofreading and

understanding the notice exceeds the benefits, and not infrequently, the notice is impenetrable by

the average citizen."). These scholars have had difficulty resolving the lack ofparticipation without

resorting to a model based on attorney expertise. Rubenstein, supra note 14, at 1663-68.

Rubenstein argues that the expertise model has been given short shrift, and that attorney expertise

has its place in group litigation, such as in making procedural or technical decisions. Id.

1 77. One cannot easily disaggregate the difficulty ofobtaining class member participation from

the types of participation class members have been permitted. Attorneys have incentives to keep

class members out ofthe negotiation and settlement process and notices are often mind-numbingly

difficult to read. These factors must be added to the cost-benefit analysis of class member



2003] CLASS ACTION GOVERNANCE 105

have also expressed the paternalistic concern that class members will not know
what is best for them, or will choose the wrong result out of shortsightedness.'^^

Another approach is consultation-based deliberation.'^^ One concrete
proposal is a procedural 'community consultation' requirement for class

actions. '^^ Such a procedural requirement may help to alleviate concerns about
coercion in the beginning of the class action lawsuit, although it would require

procedural reform and would likely be difficult to implement.'^' By what criteria
will decisionmakers determine the sufficiency of consultation? This type of

solution is particularly difficult to implement in consumer class actions because

of the lack of a coherent "community" with which to consult.

There are some analogies to community consultation in the context of small

claims class actions. Courts might seek out consumer advocacy groups or, ifone

demographic is particularly affected, "community representatives" ofthat group,

although that concept is problematic as well. This has occasionally occurred in

the context of cypres settlements. In In re Mexico Money Transfer Litigation,
for example, a case concerning a class made up almost entirely of Mexican-
Americans and Mexican residents ofthe United States who transferred funds to
their families in Mexico, the district court heard testimony from a number of
representatives ofMexican communities in the affected areas concerning the four

million dollar cypres fund.'*^ In that case, both the objectors and class counsel

brought forward community representatives, the end result of which was

participation.

1 78. See Rhode, supra note 1 1 1 , at 1 237 (articulating concerns that class members' positions

may be due to posturing or reflect peer pressure, or that class members views may not result in the

best outcome because they are too risk averse or risk seeking).

1 79. In seeking an alternative to exclusive attorney decision-making in the civil rights context,

Rhode advocated a "theory ofrepresentation mandating full disclosure of, although not necessarily

deference to, class sentiment." Id. at 1 185.

1 80. Rubenstein seeks to use procedural and ethical rules to "promote more democratic means

of client goal-setting and more expertise-driven norms of attorney decisionmaking in group

litigation." Rubenstein, supra note 14, at 1668. He proposes rules requiring lawyers to show that

they had some community interaction prior to their decision to file, or to show community

participation or that democratically elected representatives approve ofthe action. Id. at 1659. For

example, he suggests a kind of"community derivative suit," in which the representative would have

to file a special pleading showing that the case "grew out of some prefiling democratic processes"

and if the action did not "flow from a democratically produced group decision" it could be

dismissed. Id. at 1670-71. Rubenstein recognizes that this deliberative model, which offers only

limited participation, would be difficult to implement and would potentially create an overwhelming

barrier to suits.

181. Rubenstein notes the problems with the "community derivative" idea are that there is no

literal "community" with a principal place of business, the idea of community likened to

corporations could lead to less democracy because some sub-groups might get undue power and

the procedural hurdle would be costly and might undermine civil rights suits generally. See id. at

1672-73.

182. 164 F. Supp. 2d 1002, 1020-21 (N.D. 111. 2000), aff'd, 267 F.3d 743 (7th Cir. 2001).



106 INDIANA LAW REVIEW [Vol. 37:65

something similar to a battle of the experts. '^^ Although such community
involvement can lead to consensus, it may also reflect or even increase
differences within the group. Such consultation can deteriorate into a battle of

competing community leaders, taking the role of experts rather than authentic
voices ofthe community. Nevertheless, disputes should not always be construed

as negative. The inability to create consensus around the settlement may not be
the product of failed judicial leadership, but a reflection of political reality.'^"*

C The Judicial Administration Model
Another group of scholars and judges subscribe to a model of class action

governance based on judicial monitoring.'*^ Many ofthese proposals have been
inspired by mass tort cases. '^^ The subtext of most proposals based on judicial
monitoring is that the wrongs vindicated through the class action mechanism are
best addressed through governmental regulation, but in the absence of legislative

action, the class action requires courts to act like administrative agencies.'*^ The
central feature of these proposals, strengthening the hands of the court, is not

limited to the mass tort context.

1. The Entity Model.—One group of scholars has described an "entity
theory" of class actions where the class is understood as an entity rather than an

aggregate of individuals.'** Entity theory is not a specific governance

mechanism, but rather a reconceptualization ofclass structure that would change

the way procedural rules are applied in class actions. It proposes to abandon
autonomy-based approaches to class actions and shift the focus ofthe Rule 23(a)

inquiry on the class as a whole. Scholars have suggested that the entity theory

would improve the current law on class actions by rationalizing the mootness

183. Id.

184. In Hart v. Community School Board, 383 F. Supp. 699 (E.D.N.Y. 1974), Judge

Weinstein did not order the special masters' community redevelopment plan but did exliort third

parties to follow the plan. See id. at 775; Curtis J. Berger, Awayfrom the Court House and into the

Field: The Odyssey ofa Special Master, 78 COLUM. L. REV. 707 ( 1 978) (describing his experience

as a special master in the Hart case).

185. See, e.g., Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 630-41 (1997) (Breyer, J.,

dissenting); Jack B. Weinstein, Individual Justice in Mass Tort Litigation: The Effect of

Class Actions, Consolidations andMultiparty Devices ( 1 995); Edward H. Cooper, Rule 23:

Challenges to the Rulemaking Process, 71 N.Y.U. L. REV. 13 (1996); Shapiro, supra note 21, at

920.
,

1 86. See, e.g., Shapiro, supra note 21, at 920 (focusing on mass tort because "this has been

the principal area of current debate").

1 87. See Amchem Prods., Inc., 521 U.S. at 599 ("In the face of legislative inaction, the federal

courts—lacking authority to replace state tort systems with a national toxic tort compensation
regime—endeavored to work with the procedural tools available to improve management offederal
asbestos litigation.").

1 88. The entity theory seems to have been first proposed by Edward Cooper. See Cooper,

supra note 185, at 13.



2003] CLASS ACTION GOVERNANCE 107

doctrine, focusing on the actual claim rather than the individual representative,

and encouraging a more direct focus on the class and potential conflicts requiring
sub-classification.'^^

Entity theory would strengthen the trend towards federalization and
nationalization of class actions. Under the entity theory, the amount in
controversy rules would apply to the class as a whole rather than the individual

litigants, and courts could determine whether the amount in controversy has been

met on an aggregate basis. '^^ The entity theory might also affect the way we
think about due process in the class context, making "membership in the
litigating class itself a tie to the forum.'"^' An entity approach may also dictate
that a single law should apply to all class members, thus eliminating the choice

oflaw issues.
'^^

Although it seems difficult to imagine that the procedural device

of the class action should affect the law governing an individual located in his

home state, judges approve settlements that do not account for differences in the
laws of individual states.

'^^
Most importantly, according to Cooper, an entity

approach to Rule 23 would focus the court's inquiry more acutely on the
adequacy of the attorney's representation, where it rightfully belongs.'^"* Thus,

the entity theory follows up on scholarship that observed that the attorney

entrepreneur is the true engine of class action litigation.
*^^

The entity theory was further developed by David Shapiro, who argued that
"the wisest and most efficient way of promoting individual justice" is through
collective mechanisms such as the class action lawsuit.'^^ The viability of the
entity theory in a given context depends on the cohesion of the class and the

extent to which the class' interests are coextensive.'^^ This justification for

collective treatment explicitly abandons procedural autonomy values in favor of

a rough-justice approach to individual claims.
'^^

It also depends on the purposes

189. Mat 28.
190. See id. This is in contrast to an approach to the amount in controversy requirement

permitting jurisdiction where it is met by the named plaintiff, like the rule for diversity jurisdiction

in class actions. See Supreme Tribe of Ben Hur v. Cauble, 255 U.S. 356 (1921); Class Action

Fairness Act of 2003, S. 274, 108th Cong. § 4 (2003), discussed supra note 45.

191. See Cooper, supra note 1 85, at 29.

1 92. See id. (describing a uniform choice of law mandate as too extreme and stating, "[a]s a

mere procedural device, class treatment cannot alter the conceptual substance of the individual

claim, no matter how drastically the claim is affected in fact").

193. See, e.g., Wolfman & Morrison, supra note 5, at 460-61 (discussing approval of
settlements despite substantial choice of law issues).

1 94. See Cooper, supra note 1 85, at 3 1

.

195. See Coffee, Understanding the Plaintiff's Attorney, supra note 24.

196. 5ee Shapiro, 5Mpra note 21, at 916.

197. See id dX922-2'i.

198. From a political theory perspective, this view might be described as communitarian

because greater weight is given to the common good than to the realization of individual rights,

based on the idea that justice for individuals is only realized through common good. See generally
Michael Sandel, Democracy's Discontent: America in Search of a Public Philosophy



1 08 INDIANA LAW REVIEW [Vol. 37:65

of the substantive law at issue. Where deterrence is the primary goal, Shapiro
argues, an entity approach is more palatable because compensation inherently

entails an individualist orientation.'^^

The entity theory would "severely limit[] such aspects of individual
autonomy as the range of choice to move in or out of the class or to be
represented before the court by counsel entirely of one's own selection."^^"^

Shapiro proposes that the class might be represented by a group ofmembers who
would receive notice and the opportunity to object, in place ofthe individualized
notice currently required. ^^' The focus of the district court's procedural due
process inquiry would shift radically towards the adequacy ofthe representative
and away from notice and opt out requirements.^°^ Shapiro justifies this call for
limited notice and opt outs with a very narrow view of due process rights in the
class action context.^^^ But he recognizes that this proposal would require, at a
minimum, a revision of Rule 23.^^"* Curtailing the notice requirement would
prevent the expense of notice from becoming a barrier for small claims that
"deserve[] recognition as a matter of substantive law" and "individual stakes

seem to be worth the cost of litigating" but are too small to merit the notice

(1996).

199. See Shapiro, supra note 21, at 923-31 (arguing that the main purpose of small claims

class actions is deterrence, and therefore the entity approach makes sense).

200. Id. at 919.

20 1 . Shapiro argues that the Supreme Court's interpretation of Rule 23 requires more robust

notice, but that due process concerns may be more flexible. See id. at 937 (citing Mullane v. Cent.

Hanover Bank & Trust Co., 339 U.S. 306 (1 950)). This requires a very strained reading o^Phillips
Petroleum Co. v. Shutts, 412 U.S. 797 (1985). See Monaghan, supra note 6, at 1159-63. See

generally Patrick Woolley, Rethinking the Adequacy ofAdequate Representation, 75 Tex. L. Rev.

571, 573 (1997) (arguing that individual notice, hearing and the opportunity to opt out are due

process requirements).

202. 5ee Shapiro, jwpra: note 21, at 937.

203. Shapiro's view is that because the Supreme Court did not focus on opt outs and notice

in Supreme Tribe ofBen Hur v. Cauble, 255 U.S. 356 (1921), and Hansberry v. Lee, 3 1 1 U.S. 32

(1940), but instead on adequacy of representation as a due process threshold, the former are not

truly due process requirements. See Shapiro, supra note 21, at 938, 958-59. Shapiro dismisses

Shutts as essentially dictum and argues that the notice and opt out requirements articulated in that

case may not be constitutionally required. See Shapiro, supra note 21, at 938, 954-55.

204. Although he argues that due process does not prohibit a looser view ofnotice and opt out,

Shapiro believes that using Rule 23 to move towards an entity model would overstep the Court's

bounds under the Rules Enabling Act and the Rules Decision Act. See Shapiro, supra note 21, at

953. In this regard Shapiro focuses on the requirements that individual notice be sent out upon

class certification and opt out rights, and argues that the Rule

should be framed in a way that does not place unreasonable roadblocks in the way of

movement toward an entity model by responsible policymakers, nor should it impede

recognition of the present force and effect of the model in the administration of class

actions. At present, the rule may well fail both of these criteria.

Id. at 957.



2003] CLASS ACTION GOVERNANCE 1 09

Shapiro advocates a revision of Rule 23 to create stronger judicial oversight

of the adequacy of counsel, focusing on the experience of counsel, potential

conflicts of interests, and the availability ofchannels ofcommunication between

counsel and a representative group of the class members. ^^^ Judicial review at

the 23(e) hearing would then focus on the fairness of settlements overall and as
to individual class members. It is not clear from Shapiro's vague description how
this revision would be more than a codification ofthe current regime without opt

out or notice protections.^^^ In comparison to the entities recognized in our legal

system, Shapiro's governance solution seems too simple.^^^ For example,

corporations or trade unions have governance mechanisms to select and monitor

directors and to regulate membership participation. Corporations, which share

some ofthe agent-principal problems of class actions, are heavily regulated, and
subject to mandatory disclosure requirements, proxy voting and market forces.^°^

From an implementation perspective, the entity theory faces significant
hurdles, particularly because notice and opt out requirements are well entrenched

in due processjurisprudence and in Rule 23. There are, however, some doctrinal
developments that seem to move closer to the entity model. These include the
recent revision of Rule 23 to require judicial approval of class counsel,^'^ an

increasing focus on the class cohesion by the Supreme Court,^" increased

attention to the conflicts of interests of groups within the class,^'^ and the

development of a doctrine ofjudicial fiduciary duty to the class as a reaction to

conflicts of interests problems with class counsel.^'^

205. Id. at 956.

206. See id. at 959. This approach holds promise but is insufficiently developed.

207. Shapiro supports his view with reference to other corporate bodies, such as a trade union

that may be authorized by the majority ofworkers to represent them in bargaining or litigation even

when an individual worker may not have agreed to that representation. See id. at 921. The

weakness of the entity theory as a solution to the problem of class action governance is well

illustrated by the example ofa trade union. While majority vote is an acceptable democratic means

of picking a representative, self-selection is not. As Shapiro himself points this out in another

example, that of a municipality, "a company town organized and run by one's employer has less

to be said in its favor than a truly public municipality." Id. at 922 n. 1 8. While Shapiro focuses on

this problem in the context of a defendant-defined class, it is equally a problem in a class defined

by class counsel.

208. See John Leubsdorf, Pluralizing the Client-Lawyer Relationship, 11 CORNELL L. REV.

825, 828 ( 1 992) (criticizing entity theories ofattorney-client relations that do not provide decision-

making arrangements for groups).

209. Recent experience with the failure of corporate governance seems to indicate that more,

not less, watchfulness is necessary in the corporate sphere.

2 1 0. See proposed Fed. R. Crv. P. 23(g), described supra note 6.

211. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 621 (1997).
212. /^. at 626.

213. See Reynolds v. Beneficial Nat' 1 Bank, 288 F.3d 277, 279-80 (7th Cir. 2002) (overturning

settlement of class action on grounds that district court did not scrutinize sufficiently whether the



no INDIANA LAW REVIEW [Vol. 37:65

Despite its problems, the entity theory has its appeal. Thinking of the class

as a group, rather than as individual litigation multiplied, brings us a step closer

to developing a coherent theory ofclass action governance. But the entity theory
needs to do more than reinforce the dictatorial system currently in place.^^"* Its
focus on the client, a reference to the traditional system of individual litigation,

is also a call for a coherent system ofgovernance that will serve the same control
and monitoring functions in the class context that the client does in the individual

litigation context. As currently described, however, entity theory does not fulfill
its promise.

2. Relying on Judges: Justice Breyer and Judge Weinstein.—While no
judge has advocated adherence to the entity theory of class actions described

above, there are examples of judicial support for a more collectivist or even
communitarian view of the class action. Judicial focus in this regard is more on
fair outcomes than fair process of group litigation. Two prominent judges have
in their opinions and writings adopted an administrative view ofgroup litigation

that departs from the traditional atomistic approach: Justice Breyer and Judge

Weinstein. Although these judges have considered the administrative model in

the mass tort context, their thinking has implications for consumer class actions.
Justice Breyer' s concurrence in part and dissent in part mAmchem Products,

Inc. V. Windsor,
^^^

argued that, despite the flaws the majority found in the

structure ofthat settlement, because the tort system was not working and because
atomized litigation caused "[d]elays, high costs and a random pattern of
noncompensation," a judicially created system of compensation through

settlement was appropriate for asbestos claims.^ '^ He essentially advocated the
creation of a temporary administrative agency with limited appellate review.^'^

Accordingly, he was prepared to review settlements on an abuse of discretion
standard that would be significantly more deferential to the district court's

decisions than the majority.^'^ Justice Breyer's view acknowledges the extent to

settlement was collusive and failed to quantify the net expected value of continued litigation and

stated that "[w]e and other courts have gone so far as to term the district judge in the settlement

phase ofa class action suit a fiduciary ofthe class, who is subject therefore to the high duty of care

that the law requires of fiduciaries.").

214. As Cooper notes, it is hard to pretend that the class is a "real" client. "The need for a

client is most real in cases that aggregate large numbers of small claims and do not win the

involvement of any class members with substantial stakes." Cooper, supra note 185, at 31-32.

215. 521 U.S. 591 (1997).

216. /c^. at 632 (Breyer, J., dissenting in part).

217. See generally Martha Minow, Judgefor the Situation: Judge Jack Weinstein, Creator of

Temporary Administrative Agencies, 97 COLUM. L. REV. 2010 (1997) (proposing that Judge

Weinstein's work on mass tort settlements essentially creates temporary administrative agencies).

218. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 631 (1997). In reaching this

decision. Justice Breyer underscored the special nature ofthis settlement as a compensation system

preserving money for future illnesses and characterizes the settlement as "unusual in terms of its

importance, both to many potential plaintiffs and to defendants, and with respect to the time, effort

and expenditure that it reflects." Id. at 633. But despite this mitigating language, he appears to



2003] CLASS ACTION GOVERNANCE 1 1

1

which settlement is a political act, worth the sacrifice of individualist autonomy
values for the good of the group.

This argument favors almost any settlement because the alternative can

always be painted as very limited or long delayed compensation. Justice Breyer

advocated judging the adequacy of representation by results; if the settlement is

not patently unfair the higher courts should defer to the lower court's view on the

settlement.^'^ By focusing on the fairness ofthe settlement terms overall, courts
shift their attention away from individuals or subgroup conflicts within the class.
But this emphasis on outcomes raises serious questions about process. In the

class action device, who is the decisionmaker most competent to make fairness
determinations? Do we believe that we can evaluate the fairness of a result
without looking to the fairness of the process by which it was reached? In the
alternative, is an unsatisfactory outcome reached by full and fair process
acceptable?

By contrast, the majority opinion in Amchem focused on individual
interests.^^^ How should this differential between individual rights and group
compensation be reconciled? Because this tension will always have to be traded

off to some extent in class action settlements, perhaps courts should look at the
degree to which individual interests have been sacrificed in favor ofthose ofthe

collective. Because the push towards settlement may be especially strong in
difficult and complex cases, we should be concerned when settlement approval
depends on a district judge's idiosyncratic estimation of fairness.^^^

advocate the deferential standard ofreview for all settlements. It is not difficult to argue that every

class action presents unique, intractably difficult problems requiring solution at the expense of

individual fairness. Cf. Richard A. Nagareda, Turningfrom Tort to Administration, 94 MiCH. L.

Rev. 899, 902-03 ( 1 996) (urging the use ofthe "hard look" doctrine in cases where appellate courts

review class action settlements).

2 1 9. Justice Breyer describes himself as "agnostic" about the basic fairness of the Amchem

settlement. 521 U.S. at 639.

220. See id. at 629 (Rule 23, "applied with the interests of absent class members in close

view," cannot carry the load of this type of administrative model.). Although Amchem was a mass

torts, positive claim class action, the Court did not distinguish its reasoning from other types of

class actions, such as the small claims class actions discussed here.

221. As Justice Ginsburg wrote for the majority in Amchem, "the standards set for the

protection of absent class members serve to inhibit appraisals ofthe chancellor's foot kind—class
certifications dependent upon the court's gestalt judgment or overarching impression of the

settlements' fairness." Id. at 621 . While in his dissent Justice Breyer accused the Court of looking

too deeply into the fairness of the settlement, it appears that he himself had also made a

determination as to its fairness in light ofthe state ofasbestos litigation. The factors courts consider

in deciding whether to approve settlements, set forth in City ofDetroit v. Grinnell Corp., 495 F.2d

448 (2d Cir. 1 974) and widely used by circuit courts across the country, set forth concrete standards

ofevaluating settlement fairness. Nevertheless, as Judge Posner's affirmance ofthe approval ofthe

Blood Factor settlement indicates, these factors do not cabinjudges as much as perhaps they should.

See In re Factor VII or IX Concentrate Blood Prods. Litig., 159 F.3d 1016, 1018 (7th Cir. 1998)

(affirming settlement but describing it as "downright weird").



112 INDIANA LAW REVIEW [Vol. 37:65

Within the administrative model there is also a countervailing focus on

process and contractarianism. Justice Breyer's insistence on the deferential

standard of review bespeaks a trust in district court judges, perhaps subscribing

to views expressed by some circuits thatjudges have a fiduciary duty towards the
class. ^^^ To the extent that there is a divergence between the interests of the
judicial system and the interests of the class, however, judicial fiduciary duty

may not be enough. Justice Breyer, for example, expresses trust in the process
of negotiation between the two sides and focuses on the work that plaintiffs'
attorneys put into the settlement in that case.^^-' Unfortunately, the difficulty of

negotiations does not guarantee an absence of conflict between plaintiffs'

attorneys and the class, nor eliminate the structural incentives for self-dealing.^^"*

To be effective, judicial administration requires a free flow of accurate and
complete information to the judge to exercise her expertise. In fact, however,

there are often significant limitations on the information provided to the judge.
^^^

There is a danger, therefore, that the judge will only rely on the parties for

information and be denied any opposing views of facts and law. Although some
judges view their role as that of a fiduciary of the class, others view settlement

as a type of private ordering over which their oversight role is limited. This

variation creates unpredictability, a concern from a governance perspective.

Because it leaves control in the hands of a single actor, the judicial

administration model raises some ofthe same concerns as other expertise models,
that it is an "elitist subversion of democratic equality," infringing on individual

liberty and simultaneously limiting the self-realization ofthe class as a group.
^^^

222. Amchem, 52\ U.S. at 640.

223. Id. at 633.

224. The Amchem settlement structure raised very real concerns that class counsel might sell

out one portion of the class—or the class as a whole—in favor of their individual clients or, worse
yet, in favor ofcurrent money and current fees in exchange for small future awards. There was also

the specter that attorneys could cut off whole classes of claims in the interest of creating global

settlement or simply to settle. The unique problems presented by the settlement addressed in

Amchem are discussed in Susan P. Koniak, Feasting While the Widow Weeps: Georgine v. Amchem

Products, Inc., 80 CORNELL L. REV. 1045 (1995), and Wolfman & Morrison, supra note 5, at 449-
59. For a discussion on the unique aspects of the asbestos litigation, see Deborah R. Hensler, As

Time Goes By: Asbestos Litigation After Amchem and Ortiz, 80 TEX. L. Rev. 1 899 (2002).
225. "Perhaps in no other context do we find courts entering binding decrees with such a

complete lack ofaccess to quality information and so completely dependent on the parties who have

the most to gain from favorable court action." Issacharoff, supra note 7, at 808. This is

compounded by the failure of some courts to permit objector discovery and their expressed desire

not to delve too deeply into settlement. See, e.g.. Mars Steel Corp. v. Continental 111. Nat'l Bank

& Trust Co., 834 F.2d 677, 684 (7th Cir. 1987) ("The temptation to convert a settlement hearing
into a full trial on the merits must be resisted.") (citing Airline Stewards & Stewardesses Ass'n v.
Am. Airlines, Inc., 573 F.2d 960, 963-64 (7th Cir. 1978)). We should be concerned that case law
suggests that discovery rules ought not apply to settlement hearings because the rules "eliminate

the efficiency gained by the settlement itself." Wolfman & Morrison, supra note 5, at 485.
226. Rubenstein, supra note 14, at 1664.



2003] CLASS ACTION GOVERNANCE 1 1

3

It may result in the loss of vigorous discourse because power is concentrated in
the hands of the judge.^^^ Moreover, some have argued that this model raises
separation of powers issues to the extent that the judge usurps the power of the
legislature or executive to create what is essentially a temporary claims

administration body.^^^

On the other hand, a judicial expertise model has its benefits. It centralizes
and unifies decision-making, may lead to better quality and more neutral
decisions, and ensure correct allocation of resources. ^^^ This model also solves

the problems of atomization and the difficulty of creating mechanisms for

participation by giving power to a neutral decisionmaker without a monetary

stake in the outcome of the proceedings.^^^ Still, some practical questions
remain. What kind of duties does such a judge have towards the class? What
kind of flexibility of standards and transfer of power is required to enable the

judge to truly live up to this important role? And would the judge's increased
involvement in settlements compromisejudicial independence? These and other

questions have been raised concerning the work and thought of Judge Jack
Weinstein, who is the most prominent example of the expansive role a judge
might play in settling class action litigation, its pitfalls and possibilities.

Judge Weinstein is perhaps most famous for his involvement in mass tort

cases.^^' His approach has been to expand the litigation to embrace the entire

problem before him through an expansive reading ofprocedural rules. This is in

contrast tojudges who traditionally use procedural rules to exclude parties, limit
the issues and narrow or even dispose of the litigation before them.^^^ Judge

Weinstein 's approach increases the participation of various involved groups in

order to create consensus, while at the same time putting pressure on the parties

through various legal rulings. One tool Judge Weinstein has used in pursuit of
his goal ofglobal settlement was community-wide consultation.^" This practice

227. See id. (discussing drawbacks ofexpertise model in context ofattorney leadership ofcivil

rights movement).

228. As Minow explains: "Functionally, court-supervised settlements that establish systems

for processing individual claims create temporary administrative agencies without proceeding

through the legislative or executive branches." Minow, supra note 217, at 2020. Judicial action

of this type may "trigger action by the other branches, and thereby promote the vision of

overlapping and checking branches of government that lies behind the separation of powers." Id.

at 2023.

229. See Rubenstein, supra note 1 4, at 1 663 (discussing benefits ofexpertise model in context

of attorney leadership of civil rights movement).

230. This is analogous to guardian ad-litem proposals. See, e.g.., Eric D. Green, Advancing

Individual Rights Through Group Justice, 30 U.C. DAVIS L. REV. 791 (1997).

231. ^eegeweA-a/Zy Weinstein, 5Mpra note 185; Peter H. Schuck, AgentOrangeon Trial:

Mass Toxic Disasters in the Courts ( 1 987).

232. Minow, 5Mpr<3 note 217, at 2013.

233. See, e.g.. Hart v. Cmty. Sch. Bd., 383 F. Supp. 699, 756-58 (E.D.N.Y.), supplemented

by 383 F. Supp. 769 (E.D.N.Y. 1974), aff'd by 512 F.2d 37 (2d Cir. 1975) (involving not only

school and parents but also teachers union, religious and other community leaders and experts in



114 INDIANA LAW REVIEW [Vol. 37:65

Is not as rare as it might seem.^^"* Judge Weinstein also used his judicial power
to consolidate litigation, sometimes through unreviewable preliminary orders.

^^^

He relied on a creative interpretation ofthe All Writs Act to transfer cases to his
district, a legal strategy no longer available after Syngenta v. HensonP^ These
creative rulings, although not part of the traditional judicial role, do not depart

from the traditional, dictatorial approach to class action governance.^^^ They are
an example of a "legal world created almost out of whole cloth by an individual
judge.""'

Despite some concerns, scholars and prominent class action activists have
advocated increased judicial oversight as the solution to principal-agent

problems, conflicts of interests and lack of fairness in class actions. These

authors do not view the problem with class actions as a crisis of governance but
as a failure ofjudicial oversight. They recommend a series of changes to Rule
23 to givejudges additional criteria for approving settlements.^^^ Similarly, other

various substantive areas in the resolution of civil rights class action).

234. In the Mexico Money Transfer Litigation, for example, the district judge heard evidence

from leaders in the Mexican-American community as well as elected representatives. See In re

Mexico Money Transfer Litig., 164 F. Supp. 2d 1002 (N.D. III. 2000), discussed supra note 20.

235. Through this mechanism, Judge Weinstein "eliminated" such intractable problems as the

various state law claims of class members in the Agent Orange Litigation. Judge Weinstein

"vaporized the choice-of-law problem" in the Agent Orange case by issuing a preliminary order that

any state court would look to "national consensus law" to determine the manufacturers' liability,

defenses and damages, and therefore the federal court would apply that same national consensus

law to all the class members. This creative provisionary order was insulated from appellate review.

See SCHUCK, supra note 23 1 , at 1 28-3 1

.

236. See Syngenta Crop Prot. Inc. v. Henson, 537 U.S. 28, 34 (2002) (holding that All Writs

Act could not furnish removal jurisdiction and removal of cases was prohibited unless the movant

could show original federal subject matter Jurisdiction, even if petitioners were seeking to remove

a case from state court to prevent the frustration of a federal court order).

237. "Outside the sphere of the adversary process and the rule-bound trial system, the

settlement process permits room for personal persuasion, input, or pressure from the judge."

Minow, supra note 2 1 7, at 2028. See SCHUCK, supra note 23 1 , at 1 58-59 (describing the pressure

Judge Weinstein put to bear on the parties in Agent Orange and the judge's role in dictating the

settlement amount—one lower than what the parties had agreed to in that case). In discussing this
pressure and its propriety, Minow notes "the exposure ofthe sheer will ofthe judge on the amount

ofthe settlement works to remind all observers that law, and judging, inevitably reveal and express

the views of specific, real persons." Minow, supra note 217, at 2029.

238. Minow, 5Mpra note 217, at 2030.

239. For example, Wolfman & Morrison, supra note 5, propose that Rule 23 should "require
the court to reject settlements which provide no compensation for claimants who are giving up

potentially viable claims, unless the court finds that the settlement provides benefits to those

claimants that are comparable to those claims being abrogated." Id. at 498. They recommend that

claims adjudicators should be allowed to take differences in state law and settlement history in

various stales into account, see id. at 499, advocate special scrutiny of non-monetary relief, and

propose a new test for non-monetary settlements (e.g. coupons) requiring the deciding court to



2003] CLASS ACTION GOVERNANCE 1 1

5

scholars advocate guardians ad-litem as a prescription to governance problems,

hoping that the installation of additional monitoring agents will strengthen

judicial oversight.
^'^^ These proposals, while useful, are incomplete solutions

because they only address the role ofthe judge, ignoring the roles ofother actors

in the class action and the extent to which settlements are the results of the

interactions of all these players.

III. A Principled Approach TO Governance

Class action governance is the relationship among various participants in
determining the direction and outcome of the class action lawsuit.^'*'

Fundamental principles of good governance should control the relationships

between the class members, the class counsel, objectors and the judges who
oversee the suit and settlement. The concrete mechanisms of governance—^the
specific rules that control the unfolding ofthese relationships over time—should
be designed to realize these broad principles.

No one analogy of governance fits the class action context perfectly. The
corporate analogy is not entirely appropriate because claims are not fungible; in

other words, there is no market—efficient or otherwise—for class action
settlements that will mitigate the separation between ownership of claims and

their adjudication. The analogy to democratic processes is incomplete because
we are not willing to invest the resources required to obtain genuine direct
participation in small claims class actions. The administrative analogy, while in
many ways the closest to actual class action practice, does not address the
misalignment of interests between class members and their counsel. As
described above, the governance structure that has developed, dictatorship by

class counsel with more or less judicial oversight depending on highly varying

judicial practice, is not satisfactory nor is it required by Rule 23.

Some basic principles are deducible from the provisions of Rule 23 and due
process jurisprudence, in conjunction with established principles of governance

from the political, administrative and corporate contexts. The fundamental
principles proposed here rely on the basic framework of existing due process

jurisprudence and the newly revised Rule 23. Within that framework, these

principles provide a coherent and systematic legal architecture that takes into

account the special problems of class actions articulated in Part I.

Criticisms of previous proposals can also provide some guidelines for
principles for class action governance. Market-based incentives take an overly

determine whether the non-monetary relief "provides all or substantially all of the class members

a realistic opportunity to obtain valuable relief" Id. at 502. Finally, they make substantial

suggestions concerning attorneys' fees. See id. at 503-07.

240. See, e.g. , Koniak & Cohen, supra note 9 (advocating greater use of guardian ad litem to
monitor plaintiffs' attorneys' performance).

24 1 . See ROBERT A.G. MONKS & NELL MiNOW, Corporate Governance 1 (2d ed. 200
1

)

(defining corporate governance as "the relationship among various participants in determining the

direction and performance of corporations").



116 INDIANA LAW REVIEW [Vol. 37:65

simplistic view of the class action. For example, market mechanisms fail to
address other structural factors that go into decision-making, such as the fact that

class actions are a single unique transaction for claimants, and subsequent market

penalties for singular instances of mismanagement are inadequate.^"^^ In
addition, market mechanisms suffer from the persistence of incentives that
encourage attorney collusion and rent seeking without passing on benefits to the

class. Finally, creating a market in class actions would likely have prohibitive
transactions costs. A robust system of governance needs to account for these
structural factors. By contrast, mechanisms relying entirely on popular
participation are too expensive to be worth the candle for small claims.

Therefore, governance cannot be based on participation. Finally, mechanisms
that focus on judicial oversight without attention to attorney incentives,

objectors, and other participants, rely too much on judicial discretion by placing
nearly the entire burden of process on judges, who may not be institutionally
competent because of time constraints, lack of expertise, and lack of desire to

construct a complete governance system on their own.

While courts have put much thought into evaluating substantive fairness of
class actions, the procedures for ensuring fairness are inadequate.^"*^ To fill this
gap, it is necessary to focus on the principles that should dictate the specific

procedural mechanisms employed in class actions. By contrast, corporate
governance regimes emphasize procedural governance guidelines, such as voting

mechanisms, independent boards of directors and disclosure rules, while courts

defer to management on substantive evaluations of corporate business
decisions.^"^"* In the near term, a deferential standard, such as the business

242. See Easterbrook & FiSCHEL, supra note 1 49, at 1 03 (discussing the ineffectiveness of
market mechanisms on singular breaches of fiduciary duty in the corporate context).

243. For example, while courts will look at nine factors to determine the substantive fairness

of a settlement, they will only look at four procedural factors, only one of which address the real

problems in class action governance. See City of Detroit v. Grinnell Corp., 495 F.2d 448 (2d Cir.

1 974) (articulating nine factors for determining whether a settlement is fair). The nine factors are:

( 1
) the complexity, expense and likely duration of the litigation; (2) the reaction of the

class to the settlement; (3) the stage of the proceedings and the amount of discovery

completed; (4) the risks of establishing liability; (5) the risks of establishing damages;

(6) the risks of maintaining the class action through the trial; (7) the ability of the

defendants to withstand a greater judgment; (8) the range of reasonableness of the

settlement fund in light of the best possible recovery; (9) the range of reasonableness

of the settlement fund to a possible recovery in light of all the attendant risks of

litigation. . . .

Id. at 463 (citations omitted); Girsh v. Jepson, 52 1 F.2d 1 53, 1 57 (3d Cir. 1 975) (approving ofnine

Grinnell factors but remanding settlement because objectors were not given full and fair opportunity

to be heard). The four procedural factors are discussed supra note 87 and accompanying text.

244. See. e.g., Unocal Corp. v. Mesa Petroleum Co., 493 A.2d 946, 953-54 (Del. 1985).

The business judgment rule is a "presumption that in making a business decision, the

directors of a corporation acted on an informed basis, in good faith and in the honest

belief that the action taken was in the best interests ofthe company." A hallmark ofthe



2003] CLASS ACTION GOVERNANCE 117

judgment rule, is not appropriate in the class action context. Even if an adequate
governance regime were in place, judges cannot presume that class counsel is

acting in the best interests of the class when the class action is a single
transaction not susceptible to correction by market forces. While governance

mechanisms alone cannot assure fairness, healthy governance when combined
with a fairness inquiry is the most likely avenue to a robust and reliable class

action system.

Procedural rules and standards that are well conceived are more likely to lead
to consistent results, build realistic expectations, and reinforce a culture of

compliance. There are several benefits to a principle-oriented procedural

framework. It is comprehensive because rather than providing limited ad hoc

solutions to isolated problems, procedural mechanisms and rules are designed to

address all the relationships within the system. It also offers flexibility to both

judges and class counsel and therefore encourages creativity in finding the right

procedural mix. Finally, it is responsive to the changing landscape of the class

action and permits the judge and attorneys to add procedural mechanisms where
principles are not being met.

The four fundamental principles ofclass action settlement governance are (i)
maximum disclosure, (ii) an actively adversarial process, (iii) expertise of
decisionmakers, and (iv) independence of decisionmakers from influence and

self-interest. The disclosure principle is integral both to our political structure
and to corporation law. The requirement of an actively adversarial process is
unique to the class action context. The principles ofexpertise and independence
of decisionmakers are analogous to similar requirements of boards of directors

and are familiar from both the political and corporate context. These four

principles encompass most related principles. For example, accountability might

be an obvious choice for a principle of class action governance. A combination
of strict disclosure rules with an actively adversarial process will result in

accountability of class counsel, so an accountability principle is arguably not

analytically distinct from these fundamental principles.

The principled approach to governance requires judicial implementation of
mechanisms to fulfill all four principles. A complete theory and practice of
governance would fulfill these principles at each stage of the litigation. This

proposal is confined to settlement, the most contentious moment in the class
action suit. The mechanisms themselves may be altered, expanded, and/or
contracted, while the principles remain static. Thus, these fundamental

principles might be likened to governance guidelines and codes of best practices

developed in the corporate sphere.^"*^ Firms can use any number of voting

business judgment rule is that a court will not substitute its judgment for that of the

board if the latter's decision can be "attributed to any rational business purpose."

Id. at 954 (citations omitted).

245. See Holly J. Gregory, Overview ofCorporate Governance Guidelines and Codes ofBest

Practice in Developing and Emerging Markets, in MONKS & MiNOW, supra note 241, at 439.
Unlike the governance guidelines discussed by Gregory, meeting the principles articulated herein

would be mandatory though the specific mechanisms for realizing them would not have to be



1 1

8

INDIANA LAW REVIEW [Vol. 37:65

mechanisms, appointment strategies and other procedural mechanisms to meet
governance guidelines. The same is true of this class action governance regime.

A. Strong Disclosure Requirements

The first and perhaps most important principle for class action governance
is the disclosure principle. This principle requires that material information be

disclosed to class members, objectors, and judges. The principle of mandatory
disclosure is necessary for good class action governance because disclosure of
information alters the balance of power, decreases agency costs by enabling
monitoring, and has a sanitizing effect.

Information is power. In the current structure of the class action, control

over information resides almost exclusively with defendants and class counsel.

Class counsel retains expert knowledge about litigation decisions, the substance

and viability of claims, discovery, offers, terms of settlements, attorneys' fees,

and side-deals. Control over this information provides class counsel with

significant leverage over class members and potential objectors. By the manner
in which class counsel releases and presents this information, it can control the
relationship of class members and objectors to the class action. Thus, class
counsel's information is a mode of control. ^"^^ The distribution of information
mirrors and reinforces the existing power structure, that is, the near total control
over the litigation by class counsel with minimal oversight. This prevents

challenges to the existing order because the lack ofinformation available to class

members, and especially objectors and judges, limits the ability of these

important actors to challenge the existing governance structure.^'^^

Mandatory disclosure would alter this imbalance, decentralizing power by
preventing the asymmetry of information that currently characterizes class

actions. The disclosure principle recognizes that the exercise ofpower within the

mandatory. There may be significant benefits to making specific, simple, rule-like mechanisms

mandatory, as suggested by prominent corporations law scholars in a different context. See Bernard

Black & Reinier Kraakman, A Self-Enforcing Model ofCorporate Law, 1 09 Harv. L. Rev. 1911,
1916(1996).

246. See generally MICHEL FOUCAULT, DISCIPLINE AND PUNISH: THE BiRTH OF THE PRISON

( 1 995) (presenting the now widely recognized theory that modern societies exercise power through

the collection and use of information, and use this information and surveillance to internalize norms

and values so that they become normalized). Foucault's theory ofpower relationships as discursive,

that is, that power is created through discourse and the internalization of categories, has been

utilized by scholars of privacy law. See, e.g., Daniel J. Solove, Privacy and Power: Computer

Databases and Metaphoresfor Information Privacy, 53 STAN. L. REV. 1393, 1418 (2001).

247. See Owen M. Fiss, Why the State?, 1 00 Harv. L. Rev. 78 1 , 786-87 ( 1 987) (arguing that

regulation of information, whether by the government or new agencies, limits political debate); see

also Yochai Benkler, Free as the Air to Common Use: First Amendment Constraints on Enclosure

ofthe Public Domain, 74N.Y.U. L. REV. 354, 355 (1999) (arguing that the increase in privatization

ofinformation reinforces the existing power structure and hinders society's information production

and exchange process).



2003] CLASS ACTION GOVERNANCE 1 1

9

class action is relational. Settlements are not produced solely by negotiation

between class counsel and defendants' counsel, but by an interaction of a range

of actors, including the attorneys, the court, opt-outs, and objectors.
^'^^

Disbursing information to all ofthese actors will destabilize the controlling role

of class counsel and empower other actors to contribute their important voices.
Decisions about where information goes and how much will be released will
determine the extent to which power is shifted to other actors. In this regard, the

disclosure principle mandates maximum disclosure.
Second, disclosure decreases agency costs. ^''^ The most important way that

mandatory disclosure reduces agency costs is by enabling informational

intermediaries to monitor class counsel.^^° In the class action, informational

intermediaries may include objectors, who analyze the information provided by
class counsel and determine whether or not intervention and opposition to the

appointment of counsel or to proposed settlements is appropriate; independent

third parties appointed by the court to review settlement proposals or to follow

up on the ultimate distribution ofthe settlement;^^' consumer advocates looking
to prevent abuses;^^^ and other attorneys looking to set up rival class actions.

Mandatory disclosure compensates for the absence ofmarket competition to
measure class actions, and for lack of an ongoing or multi-transactional

relationship between class members and class counsel. Sophisticated scholars

248. This argument is similar to that articulated by scholars who write about intermediary

organizations in the context of administrative and public law. These authors argue that

administrative rules and regulations, as well as public services, are produced by interdependent

networks ofpublic and private partnerships rather than by top-down hierarchies. See, e.g., Matthew

Diller, The Revolution in Welfare Administration: Rules, Discretion, and Entrepreneurial

Government, 75 N.Y.U. L. REV. 1121 (2000) (advocating new forms ofaccountability and arguing

that by moving administrative agencies towards an entrepreneurial, incentive-based model,

administrators have rendered decision-making less visible); Jody Freeman, Extending Public Law

Norms Through Privatization, 116 Harv. L. Rev. 1285 (2003) (critiquing the idea of

"privatization" and arguing that public-private partnerships also extend public values to private

actors); Susan Sturm, Lawyers and the Practice of Workplace Equity, 2002 Wis. L. REV. 277

(arguing in favor of a problem solving approach that harnesses and recognizes the role of

intermediaries in creating norms in the workplace, rather than a top-down code of conduct

approach).

249. See William M. Sage, Regulating Through Information: Disclosure Laws andAmerican

Health Care, 99 COLUM. L. REV. 1701, 1747-57 (1999) (explaining the uses of mandatory

disclosure to reduce agency costs and deter agency failure ex ante).

250. On the use ofinformational intermediaries generally, see Easterbr00K& FiSCHEL, supra

note 149, at 292-93. Examples of informational intermediaries in the corporate sphere include

underwriters who price stock and auditors who evaluate firms' books.

251. This is similar to instances where class counsel does not take their fee until after the

actual compensation has been paid out. In those cases counsel's fee is a percentage of the actual

settlement amount, not of a projected amount.

252. Public Citizen Litigation Group and Trial Lawyers for Public Justice are examples ofsuch

groups.



120 INDIANA LAW REVIEW [Vol. 37:65

in the corporate context have argued that disclosure is not empirically proven to

prevent fraud or reduce agency costs.^" Certainly, mandatory disclosure rules

did not prevent the recent corporate graft perpetrated by the managers of
companies such as Enron,^^'* WorldCom,^^^ Tyco,^^^ and Qwest.^^^ And because
disclosure requires collecting and disseminating information, it is costly.

Even without empirical evidence as to the necessity of disclosure in the
context of public corporations, the current disclosure regime for class actions

seems merely to create more opportunities for self-dealing. The class members
are essentially trapped in the class action. Unlike the management of a
corporation, class counsel need not be concerned about attracting additional

claimants through good disclosure practices or by proving their reliability. Class

counsel need not devise a settlement that treats the class fairly, but merely a

settlement not so egregious as to alert the judiciary or activists to its flaws. Class

counsel need only be concerned with attrition so significant that it kills a

settlement. Such a boycott is unlikely, and even if it were not, judges are

generally inclined to approve settlements despite significant numbers ofopt outs

and objectors.^^^ Thus, strict disclosure requirements compensate forthe absence

of the continuing, long-term relationship between class members and class
counsel by requiring disclosure of information that an individual client would be
able to obtain more easily.^^^

Third, mandatory disclosure works as an external discipline to increase

253. See Easterbrook & FlSCHEL, supra note 149, at 287-97 (arguing that market
mechanisms are just as or more likely to create a verification regime to attract capital). But see

Merritt B. Fox, Required Disclosure and Corporate Governance, 62 Law&Contem. Probs. 1 13

(1999) (arguing that mandatory disclosure is necessary to educate shareholder-voters and assist

shareholders to enforce management's fiduciary duties); Louis Lowenstein, Financial Transparency

and Corporate Governance: You Manage What You Measure, 96 COLUM. L. REV. 1335 (1996)

(arguing that mandatory disclosure makes management seek out information it might not otherwise

obtain and therefore improve its management strategies).

254. See Paul Beckett et al., Power Outage: How Energy Traders Turned Bonanza Into an
Epic Bust—Unleashed by Deregulation, Industry Greed and Deceit Undid the Nascent Market,
Wall St. J., Dec. 3 1 , 2002, at A 1

.

255. See Rebecca Blumenstein & Susan Pulliam, Leading the News: WorldCom Fraud Was
Widespread—Ebbers, Many Executives Conspired to Falsify Results in Late 1990s, Probes Find,
Wall St. J., June 10, 2003, at A3.

256. See David Armstrong, Tyco to Restate Financial Results, Wall St. J., June 1 7, 2003, at

A2.

257. See Dennis K. Berman & Deborah Solomon, Ex-Executives Are Indicted in Qwest Probe,
Wall St. J., Feb. 26, 2003, at Bl

.

258. Settlements with significant opt outs are often approved. See supra note 73 and

accompanying text.

259. See Lowenstein, supra note 253, at 1344-45 (arguing analogously that disclosure

compensates for the absence of long term, knowledgeable shareholders able to sit on corporate

boards to represent other shareholders).



2003] CLASS ACTION GOVERNANCE 1 2

1

internal discipline.^^^ Class counsel may behave differently when the prospect
of transparency looms over them, and thus results may be improved ex ante. If
sunlight is the best disinfectant,^^' then the sanitizing effect of mandatory

disclosure may work from the inside as well as out—not just catching
irresponsible behavior but encouraging better behavior to make disclosure less
painful. The other side ofthis coin is that the absence ofstandardized disclosure
requirements sends a message that hiding and manipulating information is

acceptable or at least free of consequences.

Like sunshine laws in the political sphere, disclosure in the class action

context builds trust in the settlement process by exposing it to public view. Such

trust-building is especially beneficial to the class action bar which has received

substantial criticism.^" Unquestionably, disclosure requirements create

additional monitoring costs for courts as well as costs for class counsel who must
compile materials. There is a trade-off between the costs of complying with

disclosure rules and the losses that classes and the public would suffer in the
absence ofdisclosure. Whether disclosure rules will in fact produce social gains

in excess of their cost is a matter for empirical study.^^^ Even if disclosure does
not create greater efficiencies from a market perspective, it is nonetheless a

valuable counterpoint to the very serious criticisms of manipulation and lack of

accountability that plague class action litigation.

To some extent, every rule regime will encourage potential violators to seek
the edges of the rules or to manipulate the rules in order to obtain the best

individual result. Moreover, the possibility exists that mandatory disclosure may
have unintended negative consequences. For example, a vigorous disclosure

regime may chill objectors who have little funding and for whom the expense of
disclosure may be too high. Or disclosure may end settlement negotiations too
early or give unfair advantage or information to defendants. There are also third-

party problems with mandatory disclosure, particularly at the negotiation stage

of the class action. Class counsel should not be required to release information

that would hurt its ability to negotiate the best settlement for the class. Arguably,
it is difficult to determine in advance what information should be privileged.^^"^

For disclosure to be effective, it must be institutionalized properly. To be
meaningful, disclosure must be comprehensive, comprehensible, and must

happen at the right time in the decision-making process. Because ofthe potential

for manipulation and the need to protect class counsel's ability to negotiate, the

260. For an interesting discussion of this idea in the corporate context, see id. at 1358.

261. Louis D. Brandeis, Other People's Money and How Banks Use It 92 (1914)
("Sunlight is said to be the best of disinfectants; electric light the most efficient police man.").

262. See supra notes 9, 3 1 (describing criticisms of plaintiffs side class action attorneys).

263. If class counsel and defendants are required to release information regarding settlement

outcomes after approval, it may be possible to do the empirical work to determine which procedural

rules, if any, are making a difference in outcomes.

264. See Mars Steel Corp. v. Cont'l 111. Nat'l Bank & Trust Co., 834 F.2d 677, 684 (7th Cir.
1987) (Posner, J.) ("Discovery of settlement negotiations in ongoing litigation is unusual because

it would give a party information about an opponent's strategy.").



122 INDIANA LAW REVIEW [Vol. 37:65

question of what information should be disclosed is as important as whether
information is disclosed at all. A narrow disclosure standard might include only
information material to the decision ofclass members andjudges concerning the
substantive fairness of the settlement.^^^ A broad standard would require
disclosure ofall information related to the settlement and not subject to privilege,

including side-settlements and agreements not currently subject to discovery,

without any requirement ofajudicial finding ofmateriality.^^^ A third alternative
would provide a two step process: a mandatory rule requiring the disclosure of
certain specific information, in conjunction with a more liberal discovery rule
that would permit discovery of information related to the settlement but not
covered by privilege. Any ofthese alternatives would be superior to the current
regime, but the most beneficial would be the most open standard, requiring
disclosure of all information not subject to privilege.

Comprehensibilityofdisclosure depends on the audience for the information.

Disclosures in notices sent out to the class directly would differ from disclosures
to the court, objectors, or the public. In evaluating whether the disclosure

principle is met, courts must consider audience, substance, and form, i.e., both

the information disclosed and the manner in which it is disclosed. The disclosure
principle cannot be met when attorneys provide information in ways that are
difficult for the recipient to access or understand.

In notices to class members, there is a tension between completeness and

comprehensibility. In complex settlements, notices should err on the side of
comprehensibility.^^^ It is the court's obligation to make notices understandable
for the average citizen. The court should determine whether or not the notices
are easily understood, both by reviewing them and by testing the notices on a

representative group of class members prior to the actual distribution of the
notices.

In disclosures to objectors' attorneys and sophisticated third parties, the

court's oversight will relate to accessibility. Pleadings, motions, and orders are

difficult for objectors to access because the courts where they are filed are often

far away and copying court documents is expensive.^^^ A solution to these access

265. This might echo the standard in the area of securities regulation, requiring disclosure of

all information where there is a "substantial likelihood that the disclosure ofthe omitted fact would

have been viewed by the reasonable investor as having significantly altered the 'total mix' of

information made available." Basic Inc. v. Levinson, 485 U.S. 224, 231-32 (1988) (quoting TSC

Indus. Inc. v. Northway, Inc., 426 U.S. 438, 449 (1976)).

266. See supra text accompanying notes 85-86 (describing barriers to objector participation).

267. The notice could provide a two to three page summary of the settlement and make

additional information available to interested claimants. Information can also be provided to

claimants in a question and answer format which may be more accessible. Disclosures in the

notices sent out to class members should be in plain English and in sufficiently large font for the

average person to read. Many notices in class actions are so densely written as to be unintelligible

even to a well-educated attorney. In addition, these notices are often in ten point font or less and

are difficult to read.

268. In my experience, copying filed documents costs from $.25 to $.50 per page.



2003] CLASS ACTION GOVERNANCE 123

problems would be to require class counsel to post all notices, motions, opinions,

orders, and other filings on a public Internet site.^^^ This way, all information

disclosed to the court and necessary for objections would be easily and cheaply

accessible to objectors' counsel and interested claimants.

Disclosures should be required both before and after settlement approval.

The minimum information necessary for disclosure prior to settlement approval
includes (i) the total amount of attorneys' fees sought and method of calculation

for those fees,^^^ (ii) the attorneys' actual expenditure ofhours and costs, (iii) any

agreements between class counsel concerning labor and fees, (iv) predicted

administrative costs for distribution of the fund, (v) the total amount of the
settlement fund, (vi) the claimant response rate for similar class actions, (vii) the

mean recovery per claimant or the predicted value of any coupon settlement on
a secondary market, (viii) the content of any agreement concerning the

distribution of excess settlement funds, and (ix) the content of any side

agreements between class counsel or defendants and objectors or attorneys filing

other class actions arising out of the same conduct by defendant, including the

specific amount of any payments. This information should be disclosed prior to

the filing deadline for any objections and prior to the Rule 23(e) hearing.

After settlement approval, class counsel and defendants should disclose (i)

the total amount of attorneys' fees sought and received, (ii) the total amount of

actual administration costs (such as the costs ofcutting checks), (iii) the average

amount of recovery per claimant who responded, (iii) the number of claimants
responding to settlement, (iv) the total settlement fund actually paid to class

members, (v) the number of opt outs, and (vi) if applicable, the actual value of

any coupon settlement on a secondary market. The revised federal rules help
somewhat in this regard by requiring courts to approve awards ofattorneys' fees
only after submission of a motion and requiring courts to make factual findings
and conclusions of law regarding that motion.^^^

Post settlement reports, made under penalty of perjury, will provide policy
makers,judges, and objectors a means for evaluating and comparing settlements.
It also gives both sides an incentive to make sure that their rosy predictions at the

269. Telephone Interview with Alan Morrison, Director, Public Citizen Litigation Group,

Washington, D.C. (July 15, 2003).

270. See Staton v. Boeing Co., 313 F.3d 447 (9th Cir. 2002) (holding that notice of amount

of attorneys' fees is required), opinion withdrawn and superseded on denial ofreh 'g by 327 F.3d

938, 963 n. 1 5 (9th Cir. 2003) (determining to "scrutinize the attorneys' fees provisions with special

care . . . [because] class notice did not break out the amount of attorneys' fees provided for in the

settlement agreement . . . ."); Goldenberg v. Marriott PLP Corp., 33 F. Supp. 2d 434, 441 (D. Md.

1998) ("Notice ofthe potential extent ofattorneys fee awards is deemed essential because it allows

class members to determine the possible influence of the fees on the settlement and to make

informed decisions about their right to challenge the fee award.").

27 1

.

See FED. R. Crv. P. 23(h). This brings class action fee issues closer to the rules governing

civil rights fee-shifting cases, where courts decide what amount of attorneys' fees are reasonable

pursuant to motion. See 42 U.S.C. § 1988 (2000).



124 INDIANA LAW REVIEW [Vol. 37:65

fairness hearing are realistic.^^^ The Hmited empirical evidence available on this
question indicates that self-reporting correlates with superior outcomes.^^^ Some
class action firms voluntarily provide clauses in stipulations of settlement

requiring that some or all of the attorneys' fees be paid after the class has been
paid and that the fees reflect the total payout.^^"* But such disclosure cannot

depend on the occasional voluntary act of progressive class counsel.
Objectors play a pivotal role as informational intermediaries. A robust

disclosure principle would provide objectors and potential objectors an

opportunity to obtain information about all aspects of settlement, including side

deals between the participants in the litigation and third parties. Currently, to

obtain discovery, objectors must show a reason to believe that the side deal
affected the settlement.^^^ Objectors are unlikely to meet this burden, and given

the monopoly class counsel has over the class action and the misalignment of
interests, side settlements are inherently suspect. Likewise, objectors who make
side deals to drop objections must be required to publicly disclose the terms of

all such settlements and to have them approved by the court.

The involvement ofthese intermediaries raises "superagency" problems, that

272. See Staton, 321 F3d Sit 95Sn.\2.

Even when there is no direct proof of explicit collusion, there is always the

possibility in class action settlements that the defendant, class counsel, and class

representatives will all pursue their own interests at the expense of the class. For that

reason, the absence of direct proof of collusion does not reduce the need for careful

review of the fairness ofthe settlement, particularly those aspects of the settlement that

could constitute inducements to the participants in the negotiation to forego pursuit of

class interests.

Id.

273. For example, in Pm«>^v. Great Western Bank,^o.CW 95'2\\0 {CD. Ca\. 1997),acase

study described in depth in the 1999 RAND study, class counsel provided a final report showing
the actual settlement paid out, which was nearly identical to what had been predicted. See Hensler

ET AL., supra note 2, at 1 75-84. By comparison, in Roberts v. Bausch & Lomb, Inc., also discussed
in Hensler et al., supra note 2, at 145-73, the total settlement was presented to the court as

approximately $68 million depending on how many class members claimed rewards. The attorneys

were paid $8 million based on that representation, calculating a fee of a little less than fifteen

percent. The parties were not required to report the ultimate payout to the court, but based on SEC

filings RAND researchers deduced that the defendant never allocated more than $37.7 million to
pay out all expenses relating to the litigation, including attorneys' fees and administration costs.

Thus, there was likely a substantial overpayment to class counsel.

274. See, e.g.. Submission ofProposed Stipulation of Settlement, Tyson v. City ofNew York,

97 Civ. 3762 (JSM) (S.D.N.Y., Jan. 5, 2000) (requiring that fees to class counsel be Judicially

approved and not paid out until administration ofclass action was completed) (on file with author).

275. See Duhaime v. John Hancock Mut. Life Ins. Co., 183 F.3d 1 (1st Cir. 1999). In

Duhaime, the First Circuit held that objectors could not have access to the substance of a side-

settlement between objecting class members who had appealed settlement approval. Id. at 4. Non-

appealing objectors were denied discovery concerning this side settlement because the court found

that the side settlement did not "affect" the main settlement. Id.



2003] CLASS ACTION GOVERNANCE 125

is, who will monitor the intermediaries?^^^ The natural answer is judges. Class
counsel and defendants complain that many objectors are fee-seeking
mercenaries attempting to hold settlements hostage. Objectors' counsel should

be held to the same standard as class counsel, but they should not be held to a
higher standard that will chill objections.^^^ Objector intervention will likely

increase when objectors' counsel are encouraged by monetary incentives such as
the award of attorneys' fees or buy-outs to preserve global peace. Nevertheless,
the fear that some attorneys will take advantage of the system, whether they are
serving as class counsel or as objectors' counsel, should not prevent courts from

encouraging objectors to air the limits or negative aspects of settlement.

B. An Actively Adversarial Process

The adversarial principle requires that the formulation and evaluation of
settlements be reached through a vigorous, active adversarial process. This

principle recognizes two factors in class action practice. First, active class
member participation is impossible and more than likely undesirable in Rule
23(b)(3) class actions. Second, in the absence of client monitoring and

involvement, a vigorous adversarial process is necessary to expose inadequate

settlements and produce optimal outcomes. Accordingly, rather than focusing

on direct participation or class member activism, this principle takes a broader
view and redirects our lens to defendants, class counsel, the court, objectors, and

the extent to which their relationships can be structured to ensure that the

concerns of absent class members are aired and addressed.
The requirement that settlements be reached through arms length

negotiation,^^^ is a poor solution to concerns that settlements are not currently

reached through a sufficiently adversarial process that would expose all the

competing interests and attempt to resolve them in the most equitable manner
possible.^^^ Even with arms length negotiation, class counsel's interests may

276. See Samuel Issacharoff& Daniel R. Ortiz, Governing Through Intermediaries, 85 Va.
L.REV. 1627(1999).

277. See supra note 100 and accompanying text (discussing chilling effect of sanctions on

objectors).

278. See supra text accompanying note 87 (discussing requirements for procedural fairness

in class action settlements: arms length negotiations, sufficient discovery, proponents were

experienced, and limited objections).

279. As Judge Easterbrook so eloquently put it.

Representative plaintiffs and their lawyers may be imperfect agents of the other class

members—may even put one over on the court, in a staged performance. The lawyers
support the settlement to get fees; the defendants support it to evade liability; the court

can't vindicate the class's rights because the friendly presentation means that it lacks

essential information.

Kamilewicz v. Bank of Boston Corp., 100 F.3d 1348, 1352 (7th Cir. 1996) (Easterbrook, J.,

dissenting from denial of rehearing en banc), quoted with approval in Amchem Prods., Inc. v.

Windsor, 521 U.S. 591, 621(1997).



126 INDIANA LAW REVIEW [Vol. 37:65

impede her ability to serve the class. Furthermore, once they reach settlement,

defendants and class counsel share an interest in getting the settlement approved.

Commentators have also lamented that many of the problems in settlement are
the result of a lack of adversarial process, which in turn is the result of lack of

client monitoring by class counsel.^^^

Democracy theorists have pointed to participation and deliberation as ofone
of the salient characteristics of a democratic society.^^' But direct and active

class member participation is impossible because in consumer class actions
participation is too expensive in relation to the interests at stake. For instance,

democratic process might drain potential recovery to the extent that insufficient

resources would be left to attract class counsel.^^^ Furthermore, because the

stakes for individual class members are low, the external investment in
participation must be all the more significant to yield results. There are good
utilitarian reasons for not investing substantial resources in pursuit of a direct

democracy in class action governance.^^^ As a society we do not want to invest
the amount of resources necessary to render the class action a fully deliberative
and participatory process in the way that we are willing to invest in our political
process.

This does not mean, however, that there is no room for deliberation in the
class action context. Nor does it mean that the only means of participation and
communication in class actions need be passive, such as opt-out mechanisms.^^"^

Because oftheir deterrent and compensatory functions, smal 1 claims class actions

280. See Issacharoff, supra note 60, at 348 (arguing that even though the Supreme Court

warned against the evils of settling without reference to testing ofclaims in the adversarial process

in Amchem, that is just what settlement classes are).

281. See Michelman, supra note 158, at 1503 (discussing the importance of political

participation as a positive human good).

282. See generally Coffee, Understanding the Plaintiffs Attorney, supra note 24, at 725

(discussing arguments in favor ofsubstantial attorneys fees). For example, in addition to their time

and other expenditures, class attorneys are required to pay for notice and conformity with other due

process requirements. See Eisen v. Carlisle & Jacqueline, 4 1 7 U.S. 1 56, 1 77 ( 1 974) (requiring that
plaintiffs pay cost of notice). This rule creates an incentive towards settlement-only certifications,

because once a settlement has been reached, defendants will often pay the costs ofnotice and other

administrative costs.

283. In other contexts, there may be normative arguments in favor of expensive participation

regimes. One could imagine a regime where civil rights class actions would require the creation

of a truly deliberative process because of the special connection between civil rights laws and our

larger democratic processes. In that case, too, fee shifting statutes require that the violator pay

attorneys fees, and thus the additional expenditure on democratic process would not come at the

expense of attorney incentives to bring suits. Because those are not the types of class actions

discussed here, I leave the question of whether civil rights is a special case to another day.

284. See supra Part I.C (arguing that the opt out solution is ineffective because the small

claims class action is essentially a non-competitive space during the critical process of settlement

approval and noting that while opt outs may be more valuable during certification, class members

are hampered by lack of information and the incentive towards reverse auctions).



2003] CLASS ACTION GOVERNANCE 127

are a public good. Engaging in a deliberative process in the courtroom is still the

best method to probe issues the judge may consider in approving or rejecting
settlements. As we saw earlier, decisionmakers will often only address those
problems they have studied and measured.^^^ An adversarial process will
improve deliberation by expanding the depth and breadth of the issues brought

before the court. Such a process may have the added benefit of helping to form
class member preferences in favor of reasonable settlements.^^^

The size ofsettlement funds and manner in which these funds are distributed
should be the subject of deliberation and discussion. Under the current regime,
the class representative is an 'authorized' representative, meaning one to whom
the court gives the power to authorize actions on behalf of the class and little
more.^^^ Courts do not look too deeply into the self-interest of class

representatives to determine, for example, the effect of additional payments on

the class representatives.^^^ Nor do courts inquire whether class representatives
were consulted in the formulation of the settlement.

Fairness hearings provide a perfect avenue for deliberation, but have

historically tended to be insubstantial and pro-forma. The doctrines encouraging
settlement and limiting review reinforce this reality. Some courts have expressly
limited settlement hearings to prevent what they feared would become a full-
blown trial on the merits, and fear that increased support ofobjectors will dojust
that.^^^ Such courts take too narrow a view ofthe fairness hearing.^^° Rule 23(e)

hearings are intended to protect class members. Class members have only
limited means of communicating through (or with) counsel and class counsel

have their own set of interests that are not aligned with those of class members.
During settlement, class counsel and defendants share an interest in getting the

settlement approved, regardless of its objective fairness. In order to expose

unfairness. Rule 23(e) hearings should be vigorous and adversarial, involving a

variety ofagents, including class counsel, objectors' counsel, class members and
third parties.^^'

285. See Lowenstein, supra note 253, at 1335.

286. See Bartels, supra note 89, at 304 (positing that "actual outcome of political processes

must be determined not by preferences alone but also by the structure of political institutions that

channel preferences in particular ways").

287. See supra Part H.B.l (discussing the concept of representation in the class context).

288. While such payments may be necessary to encourage class members to participate as

representatives in small claims class actions, it also means that their interests are not aligned with

those of the class.

289. "The temptation to convert a settlement hearing into a full trial on the merits must be

resisted." Mars Steel Corp. v. Cont'l 111. Nat'l Bank & Trust Co., 834 F.2d 677, 684 (7th Cir. 1987)
(Posner, J.).

290. See Wolfmjm & Morrison, supra note 5, at 488 (describing a half-hour long fairness
hearing).

29 1 . For example, in Hoffman v. BancBoston Mortgage Corp., No. 91 -1 880 (Ala. Cir. Ct. Jan.

24, 1 994), a particularly egregious settlement which resulted in plaintiffs paying more in attorneys'

fees than they received in the settlement, there was only one objector. That case is described in



128 INDIANA LAW REVIEW [Vol. 37:65

The fairness hearing should be a substantive presentation of several
viewpoints on the settlement. In this sense, the fairness hearing will be a kind of

trial on the merits of the settlement, although not a trial on the underlying merits
of the case.^^^ Reconceiving the settlement hearing as a bench trial, or even a

jury trial, is the working metaphor for the kind of adjudicatory procedure we
should be looking for.-^^^

For a Rule 23(e) hearing to be adversarial, it first requires adversaries, who
may be unprompted objectors represented by counsel or objectors solicited by the
court.

^^'^
In the absence of self-motivated objectors represented by competent

counsel, the adversarial principle requires that the court appoint a third party to

act as a "devil's advocate" for the class, such as a guardian ad litem.^"^^ Such an
appointed "objector" should bring her independent evaluation of the substance

of the settlement, as well as its procedure, including the extent to which that

procedure has met the four principles discussed here. She should also evaluate

opt outs and determine what motivated any decision to opt out. A third-person
evaluation ofa settlement from claimant's point ofview should expose problems

glossed over in the presentation of the settlement by self-interested counsel.
^^^

Another avenue for introducing intermediaries in the absence of vigorous

objectors is to appoint advocacy groups such as Public Citizen Litigation Group

Koniak & Cohen, supra note 9, at 1058-74.
292. The current view, as discussed supra at 86, is that the Rule 23(e) hearing not resemble

atrial at all.

293

.

See Koniak & Cohen, supra note 9, at 11 29-30 (proposing settlements be heard byjuries).
294. See, e.g., Bell Atl. Corp. v. Bolger, 2 F.3d 1304 (3d Cir. 1993).

In assessing settlements ofrepresentative actions, judges no longer have the full benefit

of the adversarial process. ... In seeking court approval of their settlement proposal,

plaintiffs' attorneys' and defendants' interests coalesce and mutual interest may result

in mutual indulgence. The parties can be expected to spotlight the proposal's strengths

and slight its defects. In such circumstances, objectors play an important role by giving

courts access to information on the settlement's merits.

Id. at 1310 (citations omitted).

295. See, e.g., Alon Klement, Who Should Guard the Guardians: A New Approach to

Monitoring Class Action Lawyers, 21 Rev. LiTlG. 25 (2002) (proposing private monitors who

would be paid out of settlement fund). But see Koniak & Cohen, supra note 9, at 1 1 1 1 (discussing
problems faced by guardians ad litem and stating that "we can report without attribution, for

whatever it may be worth, that the guardians we have talked to understand their job is to approve

the deal that the settling parties have constructed, after suggesting a few minor changes, not to

recommend that the settlement be chucked"). The concern about guardians and other objectors

having their own self-interest is discussed below.

296. The value ofsuch independent analysis can be best illustrated in Koniak & Cohen, supra
note 9, in which these scholars provide a thorough and shocking analysis ofa misleading settlement

in which class members who received nothing were nevertheless required to pay substantial

attorneys' fees which were deducted from their escrow accounts without their direct consent. One

hopes that such a thorough presentation would have affected the court's decision in that case.



2003] CLASS ACTION GOVERNANCE 129

or Trial Lawyers for Public Justice as class guardians. ^^^ Not-for-profit groups

are more likely to provide an independent evaluation of settlements and therefore
to contribute to the adversarial nature ofthe process. Objectors' counsel seeking

fees, or guardians chosen by the defendants and class counsel, by contrast, are

more likely to be subject to influences that could prevent them from opposing an
inadequate settlement. If they are independent, state attorneys general or

consumer advocacy groups might also serve this intermediary role.^^^

In most settlements, objectors' counsel (including not-for-profits) need to be

encouraged by the payment of fees, just as class counsel does. How much ofthis
type of encouragement should courts provide? It is clear that some incentive is
necessary as objectors are a critical part of the settlement process. Courts have

held that any objectors who contribute materially to the proceeding may obtain
a fee."^^ Some courts have interpreted a "material contribution" to include a
contribution to the adversarial nature of the proceedings and the provision of

representation for a significant group of objectors. ^°° Other courts have adopted

a much narrower approach, limiting the recovery of objectors' counsel's fee to
where they "produce an improvement in the settlement worth more than the fee
they are seeking; otherwise they have rendered no benefit to the class. "^°' On an

297. These groups are described in Hensler et al., supra note 2, at 89-90. The positive

involvement of Trial Lawyers for Public Justice in one settlement is described in id. at 201-04.

298. See Koniak & Cohen, supra note 9, at 1 083 (discussing proposal to mandate alerting state
attorneys general of all class action settlements).

299. See. e.g., Petrovic v. Amoco Oil Co., 200 F.3d 1140, n56{8thCir. 1999) ('To recover

fees from a common fund, attorneys must demonstrate that their services were of some benefit to

the fund or enhanced the adversarial process."); Gottlieb v. Barr>', 43 F.3d 474, 490-91 (10th Cir.

1994) (awarding fees to attorneys for objectors whose work resulted in '"a reduction of certain fee

and expense awards, and thereby benefited the class"); Fisher v. Procter & Gamble Mfg. Co., 613
F.2d 527, 547 (5th Cir. 1980) ("In opposing the consent decree in a case in which the plaintiffs

ultimately prevailed at trial, the objectors benefited their class and helped to vindicate the important

public rights protected by Title VII."); Lindy Bros. Builders, Inc. v. Am. Radiator, 540 F.2d 102,

1 1 2 (3d Cir. 1 976) (attorneys fees awarded from a common fund depends on whether the attorneys'

"specific services benefited the fund whether they tended to create, increase, protect or preserve the

fund*'); White v. Auerbach, 500 F.2d 822 (2d Cir. 1974).

Accordingly, it is well settled that objectors have a valuable and important role to

perform in preventing collusive or otherwise unfavorable settlements, and that, as the

district court recognized, they are entitled to an allowance as compensation for

attorneys' fees and expenses where a proper showing has been made that the settlement

was improved as a result of their efforts.

Id. at 828.

300. See Elliott v. Sperr>- Rand Corp., 680 F.2d 1225, 1227 (8th Cir. 1982) (per curiam)

(granting award ofattorneys' fees to representatives ofobjectors who were formerly representatives,

and ruling that amount of fees must come from fee set-aside in settlement not from general fund).

301. Reynolds v. Beneficial Naf 1 Bank, 288 F.3d 277. 288 (7th Cir. 2002) (Posner, J.); see

also Class Plaintiffs v. Jaffe & Schlesinger, P.A., 19 F.3d 1306, 1308-09 (9th Cir. 1994) (per
curiam) (affirming denial of attorneys' fees for work of attorneys in state court class action settled



130 INDIANA LAW REVIEW [Vol. 37:65

abstract level, the latter approach is more narrowly aimed at defeating
irresponsible objectors. Just as fees for class counsel should be linked to the

benefits they produce for the class, fees for objectors' counsel should be linked

to the benefits they produce.^^^ But a measure of objectors' counsel's fees that
is linked too closely with a specific monetary increase in the settlement fund may
encourage objectors to withdraw valid objections in order to settle with class

counsel, especially when those objections are sufficiently serious to scuttle rather
than increase a settlement. Accordingly, the adversarial principle requires a

broad view of compensation for objectors' counsel. Under this view, attorneys
who make a material contribution to the adversarial nature ofthe proceedings and
assist in the realization of the other principles of governance should be
compensated even if they did not produce a monetary improvement in the
settlement.

Although opt outs, as currently exercised do not contribute to the adversarial

process, some procedural adjustments may improve this situation. First, the
timing of opt outs should be altered to permit opt outs after the 23(e) hearing

rather than prior to the hearing as currently practiced. This would encourage
objections by permitting potential opt outs to state their objections and influence

the process prior to the adoption of a settlement. It would also allow individual

claimants to take advantage of the information adduced prior to and during the

fairness hearing in choosing whether or not to participate in the settlement.^^^

Second, opt outs might be made a collective right rather than an individual
right.^^'* This would permit state attorneys general or other authorized persons
to represent sub-groups for whom the settlement is unfair and may prevent some
collateral attacks against settlements.^^^ Since opt outs do not permit claimants

to realize their autonomy, the only way they will constitute an expression of
choice is if they are collectivized. Permitting collective opt outs is not so

different from authorizing separate representation of sub-groups or the initial

appointment of class counsel.^^^ In both cases, the legitimacy of representation

in federal court class action because the work was attenuated and stating that "[w]e know of no

authority which mandates an award of fees to attorneys not formally representing the class, whose

activities in representing others incidentally benefit the class").

302 . See generally Coffee, Understanding the Plaintiffs ' Attorney, supra note 24 (arguing that

fee awards in class actions should be a percentage of recovery rather than the lodestar approach).

303. See, e.g.. In re Brand Name Prescription Drugs Antitrust Litig., 1 15 F.3d 456 (7th Cir.

1 997) (barring opt outs from appealing approval of settlement). It seems likely that the opt outs in

that case were attempting to use the leverage of individual suits to extract rent from class counsel

or defendants.

304. See supra note 68 (discussing opt out as an individual right).

305. See State v. Homeside Lending, Inc., 826 A.2d 997 (Vt. 2003) (holding that class action

settlement did not preclude suit by Vermont residents because Alabama Court lacked personal

jurisdiction due to defects in due process protections).

306. See Ortiz v. Fibreboard Corp., 527 U.S. 815, 856 (1999) (requiring sub-classification

where diverse groups within class have conflicts of interests); see also FED. R. Crv. P. Rule 23(g)

(requiring court appointment of class counsel).



2003] CLASS ACTION GOVERNANCE 131

is in question and the class action risks fragmentation.

A requirement that class counsel communicate with and build consensus
among class members, as has been suggested by some commentators,^^^ is
insufficient to comply with the adversarial principle because class counsel is

likely to learn from class members only what class counsel wants to know.
Representatives might be included in the process by permitting objectors to

cross-examine appointed representatives about their understanding of the

settlement. In the alternative, an independent "jury" ofrepresentative claimants

could be presented with the settlement to determine whether or not it is fair.^°^

Such a jury could be made up of a random selection or a representative cross-
section of class members and receive a payment commensurate with the work
they do (rather than for approving the terms of a settlement and perhaps the

award of attorneys fees). The same problems that plague jury trials would be a
problem here, particularly the fear that the claimantjury would lack the requisite

expertise or ability to master the issues at stake as well as a judge with

substantial expertise in class action litigation and settlement. A jury may be as
well placed as a judge to determine whether a settlement is substantively fair

once they understand it. Furthermore, it may be more in keeping with our
concept of settlements as a type of private ordering to allow a jury of claimants

to make this determination. Even if we believe the court is better situated to
make fairness determinations, a jury of claimants may be given a consultative
role, including the ability to ask questions and express concerns to the judge.-^^^

C. Expertise ofDecisionmakers andAgents

Quality of leadership is a well-recognized principle ofcorporate governance

that applies equally in the class action context.^'^ Quality may mean different
things for different decisionmakers. The three participants whose expertise is
most important are class counsel, judges and objectors' counsel. With respect

307. See Rubenstein, supra note 14, at 1670-71 (proposing requirement of community

consultation); Shapiro, supra note 21, at 959 (proposing representative group ofclaimants to curb

excesses of class counsel).

308. These representatives should not be persons chosen by class counsel in order to preserve

their impartiality.

309. Cf. Mariano-Florentino Cuellar, Rethinking Public Engagement in the Administrative

State 70-74 (2003) (unpublished manuscript on file with author) (proposing negotiated rulemaking

in which persons affected by a proposed rule are given an opportunity to be educated about it and

have input with legal effect).

3 1 0. For example, Business Week declared "Director Quality" one of five good governance

principles that investors should look for. See Louis Lavelle, Special Report: The Best And Worst

Boards: How The Corporate Scandals Are SparkingA Revolution In Governance, BUSINESSWEEK,
October 7, 2002 at 104 (defining "Director Quality" as follows: "Boards should include at least one

independent director with experience in the company's core business and one who is the CEO of
an equivalent-size company. Fully employed directors should sit on no more than four boards,

retirees no more than seven. Each director should attend at least 75% of all meetings.").



132 INDIANA LAW REVIEW [Vol. 37:65

to class counsel, quality has traditionally meant experience in litigating similar

claims.^" This principle is the one that has been most consistently addressed in
the present class action regime, and is expressed in the newly adopted Rule

23(g), which requires the court to evaluate the experience and quality of counsel

prior to appointment.

Quality of decision-making as applied to judges is a more difficult concept
to discuss, because to raise this issue seems to impugn the intelligence, expertise,
work ethic and level of commitment of judges. Already embedded into the
structure of class action practice are mechanisms that recognize the importance
ofexperience in complex litigation. For example. Multi-district Litigation panels
are an attempt to consolidate decision-making with judges who have special
expertise in class actions.^'^ Appellate deference to district courtjudges' fairness

determinations likewise recognizes the importance of expertise, in that case the

district court's case-specific expertise.^'^ Nevertheless, there are some indicia
that case-specific expertise is not so important. For example. Multi-district

Litigation panel judges do not retain cases referred to them for trial,^''* and
appellate courts do not always defer to the case-specific expertise of district

courts.^
'^

Finally, objectors' counsel's history and experience, as well as their financial

dealings with class counsel and defendants, should also be subject to judicial

scrutiny. Standards for evaluating objector's counsel need not be different than

those governing class counsel. Objectors represented by experienced counsel

will likely be given more attention than objectors without representation.^'^ This
may mean that objectors lacking representation will need to have a guardian
appointed on their behalf Experienced not-for-profit objectors, such as Public

Citizen Litigation Group, Trial Lawyers for Public Justice, or states' attorneys

general should be encouraged to fill this role because of their limited financial

interest and their history of responsible involvement in egregious cases.

311. The trend in Rule 23 jurisprudence is towards increasingly specialized substantive areas

of class action practice, so that a consumer class action is sufficiently different from a mass tort

class action to require separate expertise.

312. 5ee 28 U.S.C. § 1407(a) (2003).

313. See supra notes 1 06, 1 08, 2 1 8 (discussing standard of review).

314. See Lexecon, Inc. v. Milberg, Weiss, Bershad, Hynes & Lerach, 523 U.S. 26 (1998)
(holding that district court judge hearing case pursuant to multi-district litigation statute has no

authority to transfer case to himself for trial). There has been significant criticism of this decision

and Congress has considered bills to reverse this result. See, e.g., HR 1756, 106th Cong. 1st Sess.
(1999).

315. Arguably, mAmchem Products, Inc. v. ^F/W^or the Supreme Court made an independent

determination of the fairness of that settlement. 521 U.S. 591 (1997). See discussion

accompanying supra note 218.

316. See In re Westinghouse Sec. Litig., 219 F. Supp.2d 657 (W.D. Penn. 2002) (denying

attorney's fees award to pro se objector who benefited class).



2003] CLASS ACTION GOVERNANCE 1 33

D. Independence ofDecisionmakersfrom Influence and Self-Interest

The final principle for class action governance is the independence principle.
Independent decisionmakers are decisionmakers who are not likely to gain
individually from the decisions they make, so that they are free from incentives

towards self-dealing transactions. That class counsel are not independent

because oftheir financial interest in the outcome ofclass action settlements is the

staple ofmuch ofthe literature on class actions.^'^ As described in Part II, much
of this literature focuses on the agent-principal problem and on the incentives

influencing the behavior ofthe attorney as decisionmaker.^ '^ Fewer have focused
on the role of the judge as decisionmaker.-"^ Both kinds of decisionmakers are

necessary to any class action governance regime and class members are forced
to rely on their expertise. So too, state attorneys general, objectors' counsel and

guardians ad litem are not always independent of self-interest. The integrity of
all these decisionmakers is particularly of concern where, as in the class action

context, they are not selected through an arms-length bargaining process or by

popular vote, but by some more suspect mechanism: eitherjudicial appointment
or self-appointment.^^^

3 1 7. Some scholars characterize the attorney self-interest in settlement as an incentive based

problem which is not the same as collusion or self-dealing. See Coffee, Understanding the

Plaintiffs ' Attorney, supra note 24. Other scholars see this problem as resulting in rampant

breaches of fiduciary duty. See Koniak & Cohen, supra note 9. This difference is a matter of
degree rather than of kind.

318. See, e.g.. Coffee, Accountability, supra note 3 (proposing a competing proxy-type

solution to give attorneys incentive to provide better representation); Coffee, Understanding the

Plaintiffs' Attorney, supra note 24 (advocating that attorneys' fees in class actions be a percentage

ofthe fund rather than loadstart to create an incentive to maximize compensation); Macey & Miller,
supra note 7 (advocating a regime of auctions for class counsel to encourage better quality

representation); Fisch, supra note 120 (arguing in favor of increased reliance on stronger lead

plaintiffs to control attorney excesses); Rhode, supra note 1 1 1 (arguing for greater ethical

obligations for attorneys representing class to seek out class opinions); Rubenstein, supra note 14

(arguing, in part, in favor of an expertise model allowing attorneys to make litigation decisions);

Susan P. Koniak & George M. Cohen, In Hell There Will Be Lawyers Without Clients or Law, 30
HOFSTRA L. Rev. 129 (2001) (discussing the difficulties of the lawyer-client relationship in the

class action context and the possibilities for abuse); Koniak & Cohen, supra note 9, at 1 103-19
(arguing for the encouragement of subsequent suits to deter attorney misconduct and collusion in

crafting settlements).

3 1 9. Two notable examples of articles that take this issue head on are Koniak & Cohen, supra
note 9, at 1 105, 1 123-28 (describing the incentives judges have to approve collusive or inadequate

settlements); Minow, supra note 217 (discussing the tensions created by judicial involvement in

settlement making).

320. Proposed Rule 23(g) would require judicial appointment of class counsel. It is not clear,

however, whether this change in the rule would make much of a difference from current practice.

Koniak and Cohen make a very convincing argument that class action attorneys' use of side deals

to remove competing attorneys from the playing field and retain control over the class action



134 INDIANA LAW REVIEW [Vol. 37:65

With regards to class counsel, meeting the principle ofindependent decision-
making requires two types of mechanisms, internal and external. The first type
concerns internal incentive mechanisms, such as the compensation scheme for
class counsel. Internal mechanisms might include an initial auction for services
combined with compensation that should rise proportionately to actual claimant
recovery, and judicial oversight of the fairness of that compensation in light of

the ultimate outcome ofthe case.^^' The problem with pure percentage payments
is that they have an arbitrary quality to them. What percentage will sufficiently
encourage valid small claims class actions without giving attorneys a windfall out

of class members' pockets? Judicial review of attorney compensation or the
attorneys' fee motion should be accompanied by an evaluation either by objectors
or other intermediaries. ^^^ This would mimic the corporate use of independent
auditors to review filings. Generally, incentive mechanisms based on attorney
compensation should be linked to actual benefits conferred rather than predicted

outcomes and should bear some relationship to the effort expended by class
counsel.^^^ The linking offees to the actual amount paid out in settlement, rather
than counsel's predictions, should be mandatory.^^"*

Undeniably, fraud or misleading representations can be made even with the
protection of 'independent' audits and reporting requirements.^^^ This type of

malfeasance requires external control mechanisms. Corporate governance

scholars have suggested that strong legal remedies with simple bright line rules

will produce the best compliance results in an economy with limited resources
for enforcement.^^^ Such deterrents might include the encouragement of

constitutes a potential antitrust violation. See Koniak & Cohen, supra note 9, at 1200-06. As
discussed in Part II, such auctions likely will not solve other problems associated with class

representation.

321

.

Rule 23(h) now requires judicial approval of "reasonable" attorneys' fees pursuant to

motion, much like civil rights practice.

322. This is to avoid situations in which class counsel bases its fee figure on a percentage of

the total compensation fund but only a fraction of that fund is eventually paid out, leaving counsel

with a windfall. See discussion supra accompanying notes 273-74. For an excellent description

of the sophistication of these kinds of abuses, see Koniak & Cohen, supra note 9, at 1083-84.
323. The attorneys' fees in most of the egregious cases of poor settlements seem to be

calculated based on predictions of benefit for the group rather than actual benefits conferred. See

supra note 273 (discussing the Bausch & Lomb litigation); see also Koniak & Cohen, supra note
9, at 1058-74 (describing Hoffman v. BancBoston Mortgage Corp., No. 91-1 880 (Ala. Cir. Ct. Jan.

24, 1994), in which the attorneys' fee calculation was based on moneys claimants already had and

not on any benefits conferred by class counsel).

324. See, e.g., Tyson class action, discussed supra note 274.

325. The role of the Arthur Anderson accounting firm in the most recent corporate scandals

is an excellent example of this. See Kurt Eichenwald, Enron 's Many Strands: the Investigation.

Anderson Charged with Obstruction in Enron Inquiry, N.Y. TIMES, Mar. 15, 2002, at Al.

326. See Bernard Black & Reinier Kraakman, A Self-Enforcing Model ofCorporate Law, 1 09
Harv. L. Rev. 191 1, 1916 (1996) (arguing in favor of bright line rules that are easy to follow

accompanied by strong legal remedies on paper to compensate for low probability that these



2003] CLASS ACTION GOVERNANCE 135

malpractice suits to deter class action attorneys who take advantage oftheir client
by abusing attorneys' fee allocations, or the use of disciplinary proceedings

against them.^^^ Scholars critical of the plaintiffs-side class action bar have

pointed out that the cost of litigation under the current regime ofattorney control

make pursuing such suits very difficult for wronged plaintiffs,^^^ and have noted
that disciplinary regimes can be ineffectual.^^^ This is especially so where

sophisticated wrongdoing is unlikely to be captured by bright-line rules. In the

context ofallegations ofattorney misconduct, there is a trade-offbetween bright-

line rules and standards. While rules are easily administered, they are likely to

be underinclusive, targeting egregious misconduct but missing more
sophisticated misconduct. Standards can encompass more sophisticated

misconduct, but are difficult to prosecute and may overdeter attorneys from
bringing suits. "° A bright line rule with strong legal remedies for breach is
probably the most efficient choice. Even if such external mechanisms only deter
the most blatant and egregious abuses, they nevertheless should not be ignored.

Independent and objective decision-making is also an issue for evaluating the

work ofjudges in class action litigation. The court's interest may be at odds with
the interests of the class in two ways. First, judges' overloaded dockets create
an incentive to approve settlements to dispose of the complex litigation before

them.^^' Second, judges who take an active role in crafting settlements may be
influenced by their own contribution in overseeing the fairness hearing.^^^ These
issues have generally been glossed over injudicial review of settlements, but

powerful sanctions will actually be applied).

327. 5"^^ Koniak & Cohen, 5w/?rflf note 9, at 1103. It is not clear whether the use ofsuch strong
deterrents against objectors is wise given the hurdles that objectors must already overcome. See

supra note 100 (discussing heavy sanctions against objectors).

328. See Koniak & Cohen, supra note 9, at 1 106-09.
329. See Lester Brickman, Contingency Fee Abuses, Ethical Mandates, and the Disciplinary

System: The Case Against Case-By-Case Enforcement, 53 WASH. & LEE L. Rev. 1339 (1996)
(providing survey evidence to support claim that case-by-case disciplinary enforcement of

contingency fee ethics rules does not work and arguing for systematic reform).

330. At the same time, we should not forget that class actions themselves are intended to serve

a valuable compensation and deterrent function and that overly harsh penalties can result in an

inefficient chilling of one of the only avenues for consumer redress of wrongs. Because we as a

society have decided to address most consumer (as well as tort and other) wrongs through the

mechanism of group litigation, the elimination or limitation on group litigation in the absence of

legislatively imposed penalties or regulatory schemes is likely to result in no enforcement or

deterrence of these individually small but collectively substantial wrongs.

331. See Koniak & Cohen, supra note 3 1 8, at 1 5 1 (discussing the possibility that judges seek
to approve class action settlements in order to clear their dockets).

332. As Minow points out, judges both declare the law and resolve disputes, and these roles

are not always coextensive. Wherejudges are in their role ofresolving disputes, "confidence in the

judge becomes the central point of contention." Minow, supra note 2 1 7, at 2027. This is because

the "settlement process permits room for personal persuasion, input, or pressure by the judge." Id.

at 2028.



136 INDIANA LAW REVIEW [Vol. 37:65

should not be.

Expertise and independence are linked to the extent that the most
experienced judges are also those most deeply involved in the structuring of

settlements. Such judges may have difficulty stepping back at the 23(e) hearing
and making an independent assessment ofthe settlement. Although critics have
expressed concern overjudges overstepping their bounds, preventingjudges from

being involved in settlement structuring would be too great a detriment to the
governance of class actions. Class members would likely be worse off without
judicial expertise in dispute resolution, especially with respect to attorneys' fees

questions in which there is a structural incentive for class counsel to engage in

self-dealing. The newly adopted changes to Rule 23, now requiring fee motions
in class action cases, signal the rule-makers' intentions to expand judicial

involvement and oversight."^

One method of ensuring judicial independence is to separate the judicial
function of overseeing litigation from that of settlement approval. Under such a
regime, the judge who oversees the negotiation, and is likely to try the case,
would not be the same judge conducting the 23(e) hearing.""* The 23(e) judge
would be less subject to the influences of either the prospect of having to try a
complex case or her own intervention in resolving the dispute. The appeal ofthis
solution would depend in part on the cost of gaining case-specific knowledge.

Another method is to require that all settlement negotiations be handled by

a magistrate judge, rather than making use of a magistrate discretionary. This

solution would have limited effect because it would not remove the incentive for

the Rule 23(e)judge to remove the case from his docket by approving settlement.

Requiring a separate hearing before a new 23(e) judge would counteract the role
of the judge in developing settlements and remove the judicial incentive to

dispense with complex cases through inadequate settlements. At the same time,

such a solution would permit the same judge who influenced the dynamics of
settlement, in part because ofthe parties' predictions ofwhat the judge would do

at trial, to continue involvement in the case if settlement is not approved."^

There are significant problems with the idea ofa Rule 23(e) judge, however.

First, to adopt such a scheme is to admit that judges are driven by incentives and

influenced by their own deep involvement in cases. It may be perceived as an
affront to the ideal ofjudicial objectivity and independence. Second, a two-judge

solution may also be inefficient because ofthe costs and time for the 23(e)judge
to familiarize herself with a complex case. It is difficult to tell whether the costs

333. See FED. R. Civ. P. 23(h) discussed supra note 271

.

334. Minow proposed a similar experiment, asking "[wjould it be possible to ready a different

judge to proceed with trial so thejudge presiding over the settlement process could distinguish that

role from the task ofjudging?" Minow, supra note 21 7, at 2029. This is different from the current

practice of permitting a Multi-District Litigation judge to oversee a case until trial in that it

separates the judges' negotiation and settlement approval functions.

335. This would resolve a problem, pointed to by Minow, that a twojudge solution may prove

"destructive to the dynamics of settlement (if it depends in part on predictions of what actually

would happen at trial)." Id.



2003] CLASS ACTION GOVERNANCE 137

of familiarizing a 23(e) judge would exceed the benefits of more objective and
independent review of settlements and the freedom it would give the presiding

judge to influence settlement discussions.

This tension between judicial dispute resolution and judicial independence

is not resolved by the availability of an appeal. The deferential standard for
reviewing settlement approval prevents a critical review ofsettlements that might

compel judges to exercise more care ex ante to reject unfair settlements. ^^^ There

are twojustifications courts provide for this deferential standard ofreview: first,

that the trial judge is most familiar with the litigants, their strategies, positions

and proofs, and second, the public policy presumption in favor of settlements.^^^

The deferential standard may also be justified by the analogy of the class action
to a "quasi-administrative proceeding.""^

The first justification flips the criticism articulated above. Instead ofseeing
familiarity with the case as a form of bias, the prevailing doctrine of judicial

review of class actions views familiarity purely from an efficiency perspective.

Even taken at face value, however, this justification loses force where the judge

below does not take the fairness hearing seriously."^ Active involvement and

familiarity with the case come at the cost of impartiality and independence. The
second justification, a presumption in favor ofsettlements, requires a logical leap

in the class action context. The intuition behind the law favoring settlements

must be a contractual one, that private ordering is more efficient, saves judicial
and public resources, and resolves conflicts to the satisfaction ofboth contracting

parties.^'^^ But in the class action context, because ofthe inherent agent-principal

problems, the settlement does not represent a contract between the class and

defendant but between the class counsel and defendant. For this reason, courts

336. "[EJvaluation of [a] proposed settlement in this type of litigation . . . requires an amalgam

of delicate balancing, gross approximations and rough justice," and the trial court's ruling on the

adequacy of a proposed compromise is given great deference. Detroit v. Grinnell Corp., 356 F.

Supp. 1380, 1385 (S.D.N.Y. \912\affdinpartandrev'dinpartA95^2(\AAZ{2dC\x. 1974).

337. See, e.g., Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 630 (1997) (Breyer, J.,

dissenting in part) ("The law gives broad leeway to district courts in making class certification

decisions, and their judgments are to be reviewed by the court of appeals only for abuse of

discretion."); Califano v. Yamasaki, 442 U.S. 682 (1979) (applying abuse of discretion standard);

Petrovic v. Amoco Oil Co., 200 F.3d 1 140, 1 148 (8th Cir. 1999) (applying abuse of discretion

standard and citing Grunin v. Int'l House of Pancakes, 513 F.2d 114, 123 (8th Cir.), cert, denied,

423 U.S. 864 (1975)); Little Rock Sch. Dist. v. Pulaski County Special Sch. Dist. No. 1, 921 F.2d

1371, 1388 (8th Cir. 1990); In re Equity Funding Corp. of Am. Sec. Litig., 603 F.2d 1353, 1362

(9th Cir. 1979) ("A district court's approval of a class action settlement and plan of allocation of

proceeds among plaintiff classes is to be reviewed under the abuse of discretion standard.").

338. See Phillips Petroleum v. Shutts, 472 U.S. 797, 808 (1985) (describing the class action

as a "quasi-administrative proceeding" from class members' point of view); Richard Nagareda,

Turningfrom Tort to Administration, 94 MiCH. L. Rev. 899 (1996) (advocating application of the

"hard look" doctrine to mass tort class actions).

339. 5ee5Mpra note 13 (discussing limitations of fairness hearings).

340. But see Owen M. Fiss, Against Settlement, 93 YALE L. J. 1073 (1984).



138 INDIANA LAW REVIEW [Vol. 37:65

have held that district court judges have a fiduciary duty to the class and courts
have developed a number of procedural and substantive factors to review prior
to approving a settlement.^"*'

An alternative is a de novo standard of review for class action settlement
approval. Under this proposal, the problems surrounding class actions are
viewed as so intractable and damaging to the perception of our justice system
that a higher standard of review is warranted. It seems likely, however, that the
same disincentives to reject settlements that apply to district courtjudges will be
applicable at the appellate level as well. Appellate judges are no less likely to
want to reduce their docket, especially if they are inundated with class action
settlement appeals. Nor are appellatejudges going to be less reticent than district
court judges to re-open a complex matter that has been resolved.

A third option may be an intermediate standard of review.^"*^ This approach
would vary the deference of the appellate courts' standard of review depending
on the strength of the protections afforded the class. It would require de novo
review of the process and protections afforded class members. The measure by
which these protections would be judged should be the governance principles
articulated in this section—disclosure, adversarial process, expertise and
independent leadership—and correlating mechanisms to fulfill each of them.
Thereafter, appellate courts would review de novo the district court's substantive
fairness determinations in cases where governance is otherwise weak. Examples

include cases without objectors, where the fee award was protected from an
adversarial process because of a 'clear sailing' provision in the settlement

agreement,^"*^ or where no separate Rule 23(e) judge reviewed the settlement

prior to approval. Especially in complex and difficult cases, the standard of

review should reflect the level ofprotection received by the class members. The
better the mechanisms protecting the class in the first instance, the more
deferential the standard ofreview should be. The underlying assumption of this
proposal is that due process requires that the four principles of governance be

met robustly in order to bind absent class members.^"*"*

341

.

See, e.g.. City ofDetroit v. Grinnell Corp., 495 F.2d 448, 463 (2d Cir. 1974) (listing nine

substantive factors and three procedural factors).

342. The dissent in Amchem indicated that the majority was indeed applying an intermediate

standard of review and that is the reason why the court found the settlement to be patently unfair.

See Amchem Prods., Inc. v. Windsor, 521 U.S. 591 , 630-3 1 (1 997) (Breyer, J., dissenting) ("These

difficulties flow from the majority's review of what are highly fact-based, complex, and difficult

matters, matters that are inappropriate for initial review before this Court."). Id. at 630.

343. A "clear sailing" provision is one in which the defendant promises not to contest class
counsel's fee request. Such a provision, because it prohibits defendant from objecting to exorbitant

fees, exacerbates the conflict of interest between class counsel and the class and further erodes the

adversarial aspects of the fairness hearing.

344. This would mean that the right to collateral attack would depend on whether the four

principles ofgovernance were adequately reviewed, because collateral attack on ajudgment binding

class members is available where there has been a violation of due process. For enlightening

discussions on this topic, see Monaghan, supra note 6 and Kahan & Silberman, supra note 47.



2003] CLASS ACTION GOVERNANCE 1 39

Like other actors in the class context, the independence of objectors and

other intervenors from external influence—by defendants or class
counsel—should also be measured. While objectors should be encouraged, they
must also be monitored and subjected to the same governance requirements as

class counsel. This is because the very leverage that permits objectors to

destabilize settlements and obtain gains for the class also allows them to draw

scarce resources from the class members with the threat of holding hard-fought
settlements hostage. The independence of community groups and states'
attorneys general may be at issue, particularly in settlements that provide for cy
pres funding for those groups or give away benefits to persons outside the class.
State attorneys general or community representatives, for example, may be
inclined to support settlements that give away money to their constituencies even
ifthey do not compensate the consumers harmed.^"*^ Independence oiguardians

adlitem may be questioned where they are appointed by the defendants and class
counsel, or where they are paid out of the settlement fund or only paid if the

settlement is approved. Courts should scrutinize the interests of intervenors in

settlements that do not provide cash payments to claimants particularly carefully.

Side agreements between objectors and class counsel at any stage ofthe litigation

should be reviewed with close attention by courts.^"^^

Control of objectors' counsel's fees is another means to assure

independence.^'*^ Objectors' counsel have a perverse incentive to get a settlement

approved (albeit with their proposed changes) so that they can get paid. If they

succeed in scuttling a settlement, they will ordinarily not be paid unless they

manage to strike a new deal with the defendants. Thus the least mercenary
objectors, such as public interest organizations, are least likely to be paid for

their efforts. Some might argue that this incentive is not perverse but in fact
beneficial, because it pushes objectors to try to work within a hard-fought
settlement and save the costs ofrenegotiation. Nevertheless, this incentive isjust

as likely to allow defendants and class counsel to get away with too much. This
problem might be solved by requiring defendants and class counsel to post a

bond. If the court rejects the settlement as unfair to class members, objectors'

counsel who succeeded would be paid a reasonable amount out of that bond.
This would put some incentive in the hands ofdefendants not to go forward with
settlements that they know to be patently unfair.^'*^

345. See, e.g., //ire Toys "R" Us Antitrust Litig., 191 F.R.D. 347(E.D.N.Y. 2000) (attorneys

general supported settlement provided for contribution of toys worth $36.6 million to states).

346. This would require a departure from the current practice and the revised rules, which

currently only require the reporting of the existence of such side-settlements. See FED. R. Crv. P.

23(2) (2).

347. This criticism applies equally well to guardians ad litem.

348. Without a doubt, the defense bar would oppose such a requirement. The burden of the

bond could be put on class counsel, but this would simply reinforce the antagonism between

objectors' and plaintiffs' counsel and succeed in erasing the role of the defendant in forming the

settlement. Policy makers should not forget that defendants also have a role in producing

inadequate settlements.



140 INDIANA LAW REVIEW [Vol. 37:65

A truly adversarial process is impossible without stringent court oversight
over the independence of decisionmakers and intermediaries from self-interest

and influence. This will depend in some part on the court's discretion, but the
inquiry into the interests of these players in conjunction with the institution of

specific procedural safeguards should eliminate some ofthe "chancellor's foot"
problems created by unfettered judicial discretion.

Conclusion

The problems of class action abuses are a result of a failure of the existing
governance regime. In place of that flawed regime, this Article proposes a

system of governance for small claims damages class actions in the settlement

process based on the four fundamental principles of disclosure, an actively

adversarial process, agent expertise, and independence of decision-making.

These principles form a comprehensive approach to class action settlement

governance, requiring a number of mechanisms that should be implemented in
all small claims damages class action settlements. By thinking of the class as a
group in need of governance, and taking a comprehensive rather than piecemeal

approach to determining what that governance should be, we can begin to have
an understanding of the nature of the class as an organization, to improve the

functioning of that organization, and to create a fair process more likely to yield

fair substantive results.