Learning Center
Plans & pricing Sign in
Sign Out



  • pg 1
Entrepreneurial Versus Non-Entrepreneurial Models
by John C. Coffee, Jr.

Global Justice Forum Columbia University Law School October 16, 2009


How Should We Organize Aggregate Litigation? One Size Fits All (i.e., What Works in the U.S. Will Also Work in Europe).



Legal Transplants Rarely Take. Rather, Innovations Must Be Consistent With the Local Culture.
U.S.-style class action practices have been and will be resisted in Europe. But innovations are possible that can increase access to justice through class action-like procedures that are congruent with the local legal cultlure. The organization of aggregate litigation is a problem of governance, not just procedural rules.
Slide 2



The View From 40,000 Feet Corporate Governance and Litigation Governance Differ Fundamentally Over Their Rules of Organization Corporate Goverance Has an “Opt-In” rule: to raise capital, an entrepreneur must convince investors to provide capital. The more skeptical that investors are of the entrepreneur, the less they will pay for the offered securities. Hence, the entrepreneur loses when investors have doubts. Litigation Governance (in the U.S.) Has an “Opt-Out” rule: the attorney/entrepreneur bringing the class action is presumed to have the consent of the class members unless they affirmatively “opt out” of the class. Consequence: Because small claimants (particularly those with “negative value” claims) in the traditional class action are rationally apathetic, they will seldom opt in or out. Hence, the entrepreneur has less need to bond with his or her clients, and the attorney/entrepreneur does not bear the same “agency costs” (in Jensen & Meckling’s phrase)
Slide 3

Competition and Accountability 1. In both litigation governance and corporate governance, there is competition for the role of the manager/decision-maker who directs the organization. In corporate governance, competition occurs through proxy fights, takeover battles, board selection procedures, and the market for managers. In U.S.-style litigation governance, competition is limited to contests over the initial selection of the class counsel; these contests generally involve only the insiders, and voting is not used. Under the PSLRA, the choice of class counsel is given to the lead plaintiff – i.e., the party with the largest stake. Consequence: Less accountability exists in litigation governance, and class counsel has de facto monopoly power. But what are the alternatives?
Slide 4



4. 5.

I. The European Context: How Should Aggregate Litigation Be Structured?


In overview, there are essentially three basic approaches to greater accountability:
1. Closer judicial oversight (but courts have much to do already and do not welcome this role). 2. Greater Voice for class members (the PSLRA essentially followed this path in creating the “lead plaintiff” to monitor class counsel). 3. Greater Exit rights.


This essay examines the exit/voice tradeoff in two very different contexts: Europe and the United States. In the former context, I argue that an opt-in class action can be made viable and may actually yield results approximating the opt-out class. In the U.S. context, I argue that exit will outperform voice and become the force that most compels changes in class action and aggregate litigation practices. Let’s begin with a contemporary comparison of actual U.S. and European class action practices.
Slide 5


The Entrepreneurial Model The United States Follows an Entrepreneurial Model for Class Litigation. Its principal elements are: (1) An opt out class, thereby arming plaintiff’s counsel with the ability to represent virtually everyone in the class as counsel defines the class; (2) the contingent fee (typically computed on a percentage of the recovery basis, with a presumptive award of 25% or more of the recovery up to $100 million and declining thereafter);


the “American rule” on fee shifting (each side normally bears its own fees – i.e., not “loser’s pays”);

(4) a right to a jury trial in federal court on common law issues; and (5) the availability of punitive damages. Result: Class Counsel Has Enormous Leverage (But Little Accountability).
Slide 6

The Non-Entrepreneurial Model Europe’s approach to aggregate litigation increasingly accepts the class action as a procedural device, but with the following key differences: 1. 2. An opt-in class, under which each class member must affirmatively consent to be represented; A prohibition on the contingent fee (but some recognition in the U.K. and elsewhere of a less lucrative “conditional fee” that is risk adjusted);

4. 5.

A “loser pay’s rule” under which the loser bears the winner’s reasonable legal expenses;
Little use of civil juries or punitive damages. The representative plaintiff is usually a nonprofit organization which attracts members (and revenues through dues) by demonstrating its commitment to a specified goal or agenda.
Slide 7

A Preliminary Evaluation 1. U.S. commentators have generally opined that opt-in classes will be futile and that class litigation cannot be financed without the contingent fee. But this is more an attitude than an empirical finding. From a policy perspective, opt-in classes present an interesting trade-off:


a. By definition, opt-in classes will include less claims and thus give the attorney/entrepreneur less leverage in negotiations with defendants.
b. But, opt-in classes will have greater cohesion, less conflicts among the interests of class members, and greater control over their counsel. 3. What U.S. commentators have missed is that European opt-in classes are often led by a non-profit organization (such as a consumer association) that already has high reputational capital and may (to a degree) subsidize the action’s costs. From a U.S. perspective, this structure represents a hybrid of U.S. private class actions and U.S. public interest litigation (such as that brought by the NAACP, Sierra Club or ACLU). Slide 8


What is the Potential of the European Opt-In Class Action? A. U.S. Opt Out Rate Approximately .2% (i.e., 2 out of a 1,000) opt out of U.S. class actions. (See Eisenberg and Miller). But does the rate of opt outs predict the rate of opt ins? Answer: No way! B. The Swedish Experience In 2008, Professor Henrik Lindblom has identified nine recent Swedish opt-in class actions, including

1. A 2003 action against a bankrupt airline for ticket refunds, potentially covering 700 passengers with 500 opting in.
2. A 2004 action, led by a non-profit organization, against a life insurance company for an alleged fraudulent conveyance of its assets, with some 15,000 individuals opting in and making a minimum payment of 15 euros to fund litigation costs. 3. A 2004 action, brought by a consumer ombudsman, against a public utility in which 7,000 customers opted in. 4. A 2007 action by residents of a suburb against an airport challenging excessive aviation noise, in which 7,000 out of a potential 20,000 residents opt in. C. The Belgium Experience Some 5,000 out of 11,000 minority shareholders (i.e., 45%) recently opted into an action against a major corporation where the action was led by a well-known consumer organization (Test-Aankoop). Conclusion: Opt in rates will be much higher than opt out rates.

Slide 9

The U.S. Experience With Opt-In Class Actions 1. The Fair Labor Standards Act (“FLSA”) has long authorized a collective action for wage and hour violations that requires each claimant to opt in. 2. A 1996 study by the Federal Judicial Center calculated the opt-in rates on three such collective actions and found them to be 39%, 61% and 73%, respectively. 3. A more recent study of 21 FLSA cases found the average opt-in rate to be either 15.71 or 23.34% (if the two outlying cases were excluded). 4. The highest opt-in rate it found was 89.5%.
Slide 10

Do Opt-Out Cases Truly Do Better?

1. The fallacy in contrasting the assumed universal coverage of the opt out class with lower opt-in participation rates is the assumption that all class members actually receive compensation in the successful opt-out class action. 2. In fact, Professors Cox and Thomas found that less than 30% of the institutional investors in the securities class actions they surveyed actually filed claims after a settlement had been reached.
3. Yet, these were sophisticated claimants with large claims. 4. Hence, because opt out class members must file a claim to receive compensation, the same problem of rational apathy on the part of small claimants again surfaces – now at the back end of the opt out class. In truth, actual participation rates, measured in terms of the percentage of eligible claimants who receive compensation, may be similar. Slide 11

Prospects for the Future

1. The advent of the Internet greatly increased the potential of the opt-in class, as websites can be set up and potential class members contacted by broad email solicitations. 2. The use of a nonprofit “public interest” organization also lessens the impact of “loser pays” fee rules.
3. Litigation funding can be obtained from arm’s length lenders who contract for a portion of the recovery (the Australian third-party funding model) – in effect, a contingent fee that is not paid to lawyers. 4. Because the representative plaintiff would be funded over the long-term by member contributions (as the ACLU or Sierra Club are), its commitment to the litigation’s goal would be less compromised by the possible fee award.
Slide 12

The U.S. Context 1. The opt out class action is well established in the U.S. and the contingent fee funds aggregate litigation effectively. No fundamental change is likely or necessary. 2. But competition remains either lacking or misdirected (i.e., plaintiff’s law firms compete to attract lead plaintiffs, not to maximize the recovery). “Pay to Play” exacerbates these problems. 3. What would encourage competition? A. Voice-based reforms have had limited impact because small claimants remain apathetic and courts will seldom reject proposed settlements. B. Exit-based reforms encourage rival plaintiff’s attorneys to solicit class members to opt out after a proposed settlement is announced. AND THEY WORK! Slide 13

II. THE U.S. CONTEXT A Quick History of the Attorney/Client Relationship in Class Actions 1. Since the modern class action rule (Fed. R. Civ. Pro. 23) was promulgated in 1968, the pattern has been for the attorney to hire the client, not the reverse. In effect, the attorney was more the principal; the client, more the agent. 2. Plaintiffs’ law firms developed relationship with investors who held small investments (100 to 500 shares) in a broad portfolio of companies. Thus, the law firm had a ready, in-house client that it could turn to quickly when it wished to sue (which was important in a prior era when the “race to the courthouse” might decide who became class counsel). A few such plaintiffs (e.g., Harry Lewis or William Weinberger) appeared frequently in class and derivative actions, in some cases more than 500 times. 3. The PSLRA deliberately changed this pattern by assigning presumptive control of the securities class action to the investor with the greatest loss during the class period – usually an institutional investor. The logical premise was that law firms could not dominate (much less in-house) institutional investors. 4. But conflicts still remain because many public pension funds are often run by an elected public official (typically, a state or municipal Controller) who must raise political contributions, and plaintiffs law firms have become major contributors – a system known as “Pay to Play.” Slide 14

What Has Been the Impact of Voice-Based Reform (i.e., PSLRA)?

1. The Short Answer: To date, not much!
2. The Academic Answer: A. Cox and Thomas (2006) find that the ratio of settlement amounts to estimated provable losses in securities class actions has actually declined since the passage of the PSLRA. B. Similarly, they suggest that “the enactment of the PSLRA had no significant impact on settlement size.” C. “Institutional lead plaintiffs have the lowest average and median recovery percentages of any group.”

D. Cox and Thomas further find that institutional investors appear to serve as lead plaintiff in less than one fifth (18%) of the cases since the PSLRA; individuals accounted for 41% of post-PSLRA settlements.
E. Choi and Thompson (2006) find little change in the high concentration level at the securities plaintiff’s bar. Pre-PSLRA, the top five firms accounted for 56.2% of market share; post, PSLRA, 52.2%.

Slide 15

Has Anything Changed in Plaintiff’s Law Firm Behavior Since the PSLRA? Yes! The following changes are clear: 1. Plaintiff’s law firms have developed “repeat relationships” with major institutional investors and appear to invest significant time and money in maintaining those relationships. 2. Plaintiff’s law firms much less frequently serve as a solo lead counsel; the number of solo class counsel fell from 31.9% before the PSLRA to 19.6% immediately afterwards (Choi and Thompson (2006)). 3. Why? There is an “urge to merge” among plaintiff’s law firms in order to aggregate the holdings of their respective lead plaintiffs and thereby form a larger consortium than that of any rival plaintiff team:

Choi and Thompson (2006) Number of Co-Lead Counsel Firms Number of Firms 1 2

Pre-PSLRA 31.9% 35.7% 22.5% 6.0%

Post-PSLRA 19.6% 44.8% 21.9% 9.1%

3 4

6 or more



5. Bottom line: Although the number of plaintiff’s law firms has not necessarily declined since the PSLRA, the number of actual competitors has fallen because of this need to form a team in order to compete. Slide


Does Plaintiff’s Counsel Well Serve the Securities Class Action Class?
1. Over recent years, the ratio of settlement size to investors’ economic losses in securities class actions has ranged between 2% and 3%. Over the last decade, the highest such ratio was 7.2% in 1996 (See Buckberg (NERA)). 2. Although full economic losses may not be recoverable because of problems associated with loss causation, this 2% to 3% recovery ratio seems well below that in other forms of commercial litigation. 3. More importantly, when large institutions opt out of the class and sue individually, they recently appear to have done much better – often by a factor of 20 to 40 times.
Slide 17

The Recent Evidence on Opt Outs in Securities Class Actions

A. The WorldCom Litigation
1. 65 institutions opted out of the class, led by William Lerach (who lost the contest to be named class counsel). This pattern that the loser in the class counsel contest asks its clients to opt out has now become common. 2. The WorldCom class settled for $6.2 billion, but with most of that amount going to debt or bond purchasers and little to equity investors. 3. In 2005, three California funds (CalPERS, CalSTRS, and Los Angeles County Employee Retirement) who opted out of the class settled for $257 million; five New York City pension funds similarly settled their $130 million in claims for $78.9 million (or an approximately 60% recovery rate) and announced publicly that they had received “3x more than they would have recovered” in the class action. B. AOL-Time Warner

1. This class action settled for $2.4 billion in 2006 – a near record.
2. This time, 93 investors followed Mr. Lerach’s suggestion that they opt out and to date they have received at least $795 million as of early 2007.
Slide 18

The AOL Time Warner Differential
Estimated Improvement over Class Recovery 16 – 24 times better than class $9 million (19 times better)

Opt Out Settlement University of California Ohio Pension Funds $246 million $175 million

CalSTRS State of Alaska Funds

$117.7 million
$105 million $50 million (on $60 million claim)

20 times better
20 times better 80%+ recovery was “50 times better than class recovery”
Slide 19

Qwest Class Action 2005 Class Recovery: $400 million Opt Out Recovery: $411 million 1. Alaska Funds settled $89 million in claims for net recovery (after fees) of $19 million, but would have recovery $427,000 under class settlement (a 45:1 ratio). 2. Teachers Retirement System of Texas settled for a net recovery of $61.6 million, but would have received $1.4 million under class action (40:1 ratio). 3. CalSTRS settled for $46.5 million and estimates that this was “30 times more” than class recovery.
Slide 20

Tyco Class Action

1. Again, this was a mega securities class action that settled for $3.2 billion in late 2007.
2. The Court’s order approving the settlement listed some 288 opt outs. 3. The majority of these were mutual funds that opted out with a separate opt out counsel representing each mutual fund family.

4. Thus, plaintiff’s counsel have now built a bridge to mutual funds, which formerly (unlike pension funds) ignored securities class actions and often did not even claim the recovery owed them.
5. But the recovery from opting out is too big to ignore.
Slide 21

Why Do Opt Outs Do Better? 1. They Escape Federal Restraints

A. The PSLRA is Inapplicable in State Court
B. Federal Class Certification Standards Are Irrelevant C. The “Fraud on the Market” Doctrine Is Unnecessary and So Proof of Market Efficiency Becomes Unimportant 2. A More Favorable Forum A. Local Plaintiffs Have a “Home Court” Advantage, Suing in State Court on a Non-Class Basis B. Some “Blue Sky” Statutes Do Not Require Any Showing of Scienter 3. Insolvency Constraints

(The defendant cannot credibly threaten bankruptcy because of an individual or consolidated action, but can in the class action)
4. Agency Costs 1. Greater Loyalty 2. Less Heterogeneity in Preferences 5. Economic and Voting Leverage Large claimants can exploit their economic leverage (CalPERS is more than a plaintiff; it is a permanent participant in the company’s governance structure and must be dealt with) Slide 22

When Will Institutions Stay in the Class? 1. Few opted out of Enron? Why? A. Loyalty to Lerach firm? B. High risk so that few other firms would undertake contingent fee litigation against secondary parties? C. The large class action threatens enormous losses to the defendant and may compel settlement D. The 7.2 billion settlement fund was too good to pass up 2. Disincentives for Opt-Outs A. Higher percentage fees in opt-out litigation because recovery is smaller

B. Higher risk of adverse trial result because defendants can afford to go to trial
C. Loss of Anonymity? 1. Hedge funds may fear depositions about their trading behavior 2. Counterclaims?

3. Where is the breakpoint at which potential opt outs will decide to sue individually? The competition between class counsel and opt out counsel will be for class members with intermediate losses (say, $500,000 to $5 million in market decline).

Slide 23

Implications 1. “Exit” May Work Better than “Voice” A. The Lead Plaintiff is a “Voice”-based reform and has had modest impact to date. B. Opting Out is the paradigmatic “exit” remedy and produces superior results. 2. Will “Opt-Outs” Deplete the Class Recovery and Reduce Per-share Payments to Remaining Class Members? Or Will They Increase the Total Recovery Across Multiple Actions? A. In the mega case, the defendant is still under great pressure to settle even if opt outs reduce the potential damages from $10 billion to $5 billion B. Do “smaller” shareholders deserve a preference on egalitarian grounds? 1. Pension funds (who opt out) actually have the poorest beneficiaries. 2. Maybe smaller shareholders will do better in the long-run because of more active competition. 3. Competition Between Class Counsel and Opt Out Counsel Goes to the Merits and May Produce Greater Value to Investors Than Competition Over the Lead Plaintiff Designation (which tends to produce political contributions and “educational” seminars) Slide 24

What Can Defense Counsel Do to Deter Opt Outs? 1. Advance the date for Class Certification A. Most opt outs occur after a settlement has been struck and appears “cheap.”

B. By advancing the certification date, class members must decide whether to opt out before they know the settlement’s terms. Some will be apathetic.
2. Seek Disproportionate Reductions in the Class Settlement for Institutional Opt Outs (i.e., a possible 5:1 reduction) Query: Is the Settlement “Fair and Adequate” with such a Provision? 3. Seek Legislative Change A. But the right to opt out has constitutional dimensions (e.g., Ortiz).

B. Seek to empower the JPML to keep opt outs before the transferee court (and deny escape to state court – a “CAFA” extension).
Slide 25

What Should Courts Do? If competition is believed desirable, courts should A. Delay the certification decision until a settlement has been reached (or permit an additional opt out period after the settlement is disclosed to the class); B. Reject as unfair settlements that have terms designed to discourage opt outs or penalize the class because of them (i.e., disproportionate reductions of the settlement, or special priorities in bankruptcy). C. Courts should permit “Most Favored Nation” clauses that give the class an increase if opt outs receive more. D. The number, rate and identity of opt outs should be disclosed to the class before the opt out period expires – to better inform their choice. BOTTOM LINE: As in other markets, Competition Works!
Slide 26

To top