25 Rev. Litig. 497

Dissent from Recommendation to Set Fees Ex Post

Charles M. Silver

This dissent argues that judges should mimic the market (1) by setting fees when class actions begin (ex ante) instead of when they end (ex post) and (2) by using evidence drawn from markets to determine how large fee awards should be. The view that judges can set fees accurately only ex post, because only then can time expended be known, reflects a misunderstanding of the contingent fee and wrongly elevates lawyers’ efforts over the results they achieve. This dissent ends with a prediction that the ex ante approach, which has gained a foothold in securities fraud class actions, will eventually displace the tradition of setting fees ex post.


   The Task Force recommends that judges presiding over class actions set fees when litigation ends ("ex post") rather than when it begins ("ex ante"). Although this recommendation tracks the traditional practice, it denigrates the emerging trend of setting fees ex ante and it may extend the life of a bad policy. To see that ex post fee setting is wrongheaded, one need only realize that nearly all clients and lawyers set compensation terms ex ante when bargaining face to face, as all members of the Task Force and all witnesses who testified before it concede. The market's signal is unambiguous: terms governing fees and costs are best established when litigation begins. By encouraging judges to lay out compensation terms at or near the outset of class litigation, the Task Force would have made a commonsense effort to bring judicial regulation into conformity with private practice.    The tradition of setting fees ex post is responsible for much that is wrong with the modern class action. It discourages plaintiffs' attorneys from investing in litigation as heavily as absent class members would want by making the returns on investments uncertain. It encourages sellouts by giving defendants leverage over plaintiffs' attorneys, who must bargain with defendants over fee terms when negotiating settlements. It violates the Due Process Clause by creating conflicts between class members and their representatives, and by denying absent claimants information that would help them decide whether to stay in or opt out of litigation classes. These deficiencies are obvious and well known. n1    Given the shortcomings of the ex post approach, why did most Task Force members endorse it? Their principal reason was the belief that judges can set fees accurately only after the results of litigation are known. Before critiquing this assertion, I note that the Task Force received no evidence supporting it. Some judges have set fees ex ante in class actions, and sophisticated institutional investors have done so many times in class actions brought under the Private Securities Litigation Reform Act. In theory, one could compare cases in which fees were set ex ante with cases in which fees were set ex post and ask whether class members fared better in the former or the latter. The Task Force received no testimony making this comparison. The belief that judges set fees more accurately ex post than ex ante is simply an article of faith.    There are also obvious reasons to expect inaccuracy when fees are set ex post. First, one purpose of the contingent fee is to compensate lawyers for taking risks. To assure that fees and risks correspond, it would seem better to set fees when litigation risks are palpable than when a case has settled and these risks have disappeared. Judge Frank Easterbrook made this point in the Synthroid class action. "Only ex ante can bargaining occur in the shadow of the litigation's uncertainty; only ex ante can the costs and benefits of particular systems and risk multipliers be assessed intelligently." n2 Second, ex post fee setting is undermined by the hindsight bias, which causes judges (and other human beings) to misestimate ex ante odds. Although the matter has never been quantified, the hindsight bias probably causes judges to miss the mark considerably when setting fees ex post. n3    Task Force members also tied the concern about accuracy of ex ante fee setting to the absence of reliable benchmarks. They asked: How can a judge setting a fee at the start of a class action know how large the fee should be? How can anyone know until a case is over and the number of lawyer hours is known? These questions convey a striking misunderstanding of the contingent fee. Clients including sophisticated clients with sizeable financial stakes have set contingent fees at the start of litigation millions of times. They have always done so without knowing how much time litigation would actually require, this information being unavailable until litigation ends. The only conclusion one can draw is that the members of the Task Force who endorse the ex post approach wrongly think information about lawyer hours actually expended is essential. They do not see that the contingent fee rewards lawyers for bearing risks and generating results, not for time actually expended.    The ex ante approach uses a simple benchmark: the percentage or range of percentages prevailing in the private market in similar contingent fee representations. Use of this benchmark follows naturally from the object of the fee setting process, which is to approximate the fees claimants would have agreed to pay had they retained counsel directly. Judge Posner made it clear that this is the objective: "It is not the function of judges in fee litigation to determine the equivalent of the medieval just price. It is to determine what the lawyer would receive if he were selling his services in the market rather than being paid by court order." n4    Judge Easterbrook expressly endorsed the use of private market fees as benchmarks in Synthroid. Responding to the charge that one "cannot know precisely what fees common fund plaintiffs ... would agree to," n5 he observed that "a court can learn about similar bargains. That is at least a starting point ... . Any method other than looking to prevailing market rates assures random and potentially perverse results." n6 I agree. Judges presiding over class actions should set fees ex ante, and they should use percentages negotiated in the private market in comparable cases as guides.    Some Task Force members responded to the proposal to use fees paid in comparable cases as benchmarks by denying that cases comparable to class actions exist. I reject this assertion categorically. There are thousands of comparable cases. Plaintiffs have formed voluntary groups in mass accident cases, pollution cases, defective product cases, securities fraud cases, and cases of other kinds. Associations, including homeowners' associations, interest groups, unions, partnerships, and corporations have sued on behalf of their members or owners thousands of times. Governmental entities, including states, cities, and counties have hired private attorneys to vindicate the interests of taxpayers and depositors. Bankruptcy trustees have sued to collect money for creditors. All of these lawsuits are examples of aggregate litigation, and in all of them lawyers' fees have been set ex ante via negotiations.    The problem is not a dearth of comparable cases. The problem is that too little reliable data on fees paid in comparable cases has been collected. n7 If judges were to start using comps and setting fees ex ante, this would quickly change. The health care sector teaches a relevant lesson. Historically, neither hospitals nor doctors gathered data on health care quality. They did not measure, for example, the frequency with which surgery patients contract post operative infections, the percentage of heart attack patients who receive recommended treatments on being admitted with chest pain, or the risk adjusted survival rates for patients who receive cardiac bypass operations. Health care providers began to measure these variables and others when employers made doing so a condition for treating patients covered by their benefit plans, when states started issuing heath care report cards, and when Medicare starting giving providers bonuses for delivering superior care. A few years ago, we could not tell the better health care providers from the inferior ones because we had no data. Today, we may have more data on health care quality than we can use because data collection is financially advantageous.    "It is tough making predictions, especially about the future," Yogi Berra once observed. Even so, I will go out on a limb. I predict that in ten years or less, the ex ante approach will gain a solid following and the ex post approach will start to wane. Securities class actions led by sophisticated investors will push the trend ahead. Already, these investors bargain over fees with class counsel up front, and judges are rightly reluctant to adjust the terms of their deals on the back end. Sophisticated investors with large financial stakes have no incentive to agree to fee terms that are unduly generous or overly stingy. The ease of handling fees in these cases will show judges that ex ante fee setting works well, and will prompt them to experiment with it in other cases. Eventually, judges and lawyers will develop a preference for the ex ante approach and will regard ex post fee setting as both exceptional and suspect.
Past Issues Subscriptions Submissions Editorial Board Journal Staff
    - TROL Home
- Advisory Board
- Abstracts
- Sponsors
- News