131 F.R.D. 688 | N.D. Cal. | 1990
ORDER
The “all too familiar path of large [class action] securities cases,” such as those now before the court, is one of “lugubrious” pleadings contests, “massive” discovery and settlement, on the eve or just after the start of trial. In re Activision Securities Litigation, 723 F.Supp. 1373 (N.D.Cal. 1989). At the end of this path are plaintiffs’ applications for attorney fees, which led Judge Patel of this court to lament:
“It is at this point in these and other common fund cases that the court is abandoned by the adversary system and left to the plaintiffs' unilateral application and the judge’s own good conscience. Rarely do the settling defendants, who have created the pool of money from which the attorneys’ fees are awarded, offer any counterpoint; rarely do members of the class come forward with any response or opposition to the fees sought. There are no amici curiae who volunteer their advice.”
Id. at 1374.
For reasons well articulated by Judge Patel, the lodestar approach,
While the protracted nature of most common fund actions is not solely the result of court-adopted attorney compensation schemes, the lodestar and its variants create incentives for wasteful litigiousness by both sides. Remedies initially sought are delayed, while unnecessary costs are imposed on the parties and — to the extent that other pressing matters are crowded out of the judicial system — society.
Because Judge Patel ultimately utilized the lodestar approach in Activision, notwi£hstanding an eloquent and lengthy condemnation of it, Activision was merely a shot across the bow of the legal profession — a call for future courts to rely on new methods of determining attorney compensation in common fund securities litigation.
I. SELECTION OF LEAD COUNSEL FOR THE CLASS.
On March 27, 1990, Oracle Systems Corporation announced what plaintiffs allege were disappointing earnings for the quarter ending February 28, 1990. The next day, the price of Oracle’s stock dropped thirty-one percent on heavy volume. The day after that, four of these actions were filed on behalf of various classes of Oracle shareholders, together with a derivative action on behalf of the corporation itself. The following day, five more proposed class actions and another derivative action were filed. Within a week, fourteen separate class action complaints had been filed by more than 25 of the leading plaintiffs’ class action law firms in the country. Four more class action complaints straggled in the following week. Motions to certify a class and to appoint lead counsel were filed.
Meanwhile, out of what they claimed was “the immediate necessity for an organizational structure to coordinate and lead a unified prosecution,” on April 2, Berger & Montague and Milberg, Weiss, Bershad, Specthrie & Lerach (two of the firms filing class actions) “set in motion the process by which all plaintiffs’ counsel would come together at a meeting to discuss and vote upon the appropriate leadership for the class actions.” Deck of Sherrie R. Savett filed April 20, 1990, ¶ 3. A letter setting up a meeting was telecopied to the other plaintiffs’ law firms. Id. ¶ 4. The lawyers calling the meeting, which was eventually set for April 12, urged that “it is important that everyone attend so that a consensus on organization can be reached.” Id. Exh. 4. Some of the plaintiffs’ lawyers, principally those from the David B. Gold firm and Kaufman, Malchman, Kaufman & Kirby, declined the invitation, evidently sensing that they would be outvoted on any decisions made at the meeting. At the April 12 meeting, 15 lawyers from 15 separate firms executed a sign-up sheet and voted on an organization of the litigation and a leadership structure of two co-lead counsel. Id. ¶¶ 9-12. After nomination and voting, the Berger and Milberg firms were elected co-lead counsel. Minutes of these proceedings were prepared and signed. Id. Exh. 6.
At a hearing on May 4, 1990, the Berger and Milberg firms sought the court’s ratification of their election at the April 12 meeting as co-lead counsel. The Gold and Kaufman firms opposed this, and sought designation of themselves as co-lead counsel. The two camps of plaintiffs’ lawyers squared off, sending volleys of disparagement at each other. On the one hand, the Berger and Milberg firms, led by Ms. Savett, lavishly praised their own virtues and excoriated Mr. Gold as a “chronic dissident” and “obnoxiously old-fashioned,” among many other things. Mr. Gold, for his part, lauded his own accomplishments and accused the Berger/Milberg group of “meritless gamesmanship” and “Mayor Daley electoral processes.” With an air of gallantry, Stdart Savett, counsel in one of the derivative actions and Ms. Savett’s husband, defended his wife and accused Mr. Gold of classic “chutzpah.” Mr. Gold darkly hinted at an undercovers arrangement between the Savetts to the detriment of the class and derivative plaintiffs.
Putting aside , this sport, the court told the contestants to compete for lead counsel designation on the basis of budgets for the litigation to be submitted in ten days. On
The Berger-Gold joint proposal called for trial “to commence in approximately fifteen (15) motions [sic],” estimated costs prior to trial of between $460,000 and $780,000, with an additional $400,000 to $600,000 necessary in the event of a trial, estimated damages in the “tens, if not hundreds of millions of dollars” and urged the adoption of the “benchmark" contingent fee percentage of 30 percent suggested by Judge Patel in Activision. The Berger-Gold proposal promised a “no-holds barred prosecution of this case for as long as it takes, no matter what it costs,” but in the context of a “willingness to settle the case at any time.” Whatever the meaning, much less the value, of these vague assurances, the Berger-Gold proposal is inadequate because it clearly evinces an unwillingness on the part of plaintiffs’ counsel to compete for lead counsel designation. In light of the discussion that follows, therefore, the Berger-Gold proposal is rejected.
II. THE COURT AS FIDUCIARY IN COMMON FUND CLASS ACTIONS.
The complaints allege causes of action based on §§ 10(b), 20, and 29(a) of the Securities Exchange Act of 1934, fraud, and negligent misrepresentation. There is no express basis in the Exchange Act or in common law for an award of attorney fees in the actions at bar. However, because these actions may create a “fund-in-court,” and because courts have determined that counsel whose efforts lead to the creation of such a fund should be rewarded, it is necessary for the court to consider the subject of attorney compensation. See Vincent v. Hughes Air West, 557 F.2d 759, 768-772 (9th Cir.1977).
In contrast to situations in which attorney fees are recoverable by statute from one of the parties and where fee claims may be attacked and defended in an adversary proceeding, courts in common fund class actions are, as Judge Patel noted, abandoned by the adversary system. Yet the court bears fiduciary responsibilities to the class. Under Fed.R.Civ.P. 23(d), the court may make appropriate orders “for the protection of the members of the class.” In light of the undeniable non-involvement of the plaintiffs themselves on the matter of class counsel fees or indeed any other aspect of the litigation,
Hindsight will invariably alter the perceived fairness of class counsel’s compensation arrangements. For example, if counsel engages in only a limited amount of discovery and motion practice, a judge’s perception of how much compensation is due plaintiffs’ lawyers is likely to be diminished — even if a large and early settlement is produced. Conversely, lengthy discovery and numerous motions will always seem deserving of significant recompense, even if the settlement is smaller and long in coming. Yet the former outcome is clearly of. greater value to the class.
Moreover, it is inherently illogical for lawyers to undertake litigation on the basis of the risks and rewards they perceive at the beginning, yet be compensated on the basis of the risks and rewards the court perceives at the end of the litigation.
In addition to its effect on the conduct of the litigation once initiated, this uncertainty surrounding attorney compensation also
In any case, the ex post approach is unnecessary, at least in the present litigation. Here, more than 25 law firms, most of them fully qualified to undertake this litigation, have filed proposed class actions. Since it is in the nature of class actions that no single class member has a significant enough stake in the outcome to justify oversight of the litigation, it is inconceivable that any “shopping around” among these law firms to arrange the best terms available will take place. That the task of protecting the class has instead fallen to the court does not mean that the selection and compensation of class counsel are not best determined by a freely competitive process.
III. MARKET FOR LEGAL SERVICES: BASIC METHODS FOR COMPENSATING LAWYERS.
Lawyers, no less than sellers of goods and other services, prefer if possible to compete on some basis other than price. Historically, lawyers have differentiated their services from their competitors through a variety of non-price criteria: prestigious degrees and associations, impressive offices, orotund rhetoric about their successes, identification with what is hoped will be perceived as the “public interest” and, in a few instances, buffoonery.
To some degree, the relative absence of price competition in the market for lawyers’ services reflects the difficulties experienced by litigants, on the one hand, in measuring the value of those services and by lawyers themselves, on the other hand, in measuring the value of the inputs necessary to produce those services. The uncertainty of litigation outcomes poses the greatest difficulty. Institutional litigants such as insurance companies and product liability defendants engage in some price shopping for lawyers, and low-cost legal climes and group legal plans provide another counterpoint to the generalization about the absence of price competition, but it remains largely true.
The hourly fee “creates an incentive to run up hours, to do too much work in relation to the stakes of the case,” ibid., a point made daily in the federal courts. However, when a sophisticated client (typically, but not always, a large corporate litigant) can easily monitor the lawyer’s performance, and the stakes are high, the hourly fee is a reasonably satisfactory method of compensation.
A contingent fee may be a fixed sum or, more commonly, a percentage of any recovery.
As explained by Judge Posner in Chesny v. Marek, 720 F.2d 474 (7th Cir.1984), rev’d on other grounds, 473 U.S. 1, 105 S.Ct. 3012, 87 L.Ed.2d 1, however, the contingent fee arrangement does not entirely eliminate the conflict of interest between a lawyer and his client.
IV. SUITABILITY OF CONTINGENT COMPENSATION FOR CLASS ACTION COUNSEL.
For these reasons, and as illustrated by Judge Patel's opinion in Activision, there is growing recognition in the courts that the percentage contingent fee is a suitable method for compensating counsel in a common fund class action. See, e.g., Paul, Johnson, Alston & Hunt v. Graulty, 886 F.2d 268 (9th Cir.1989). Indeed, when compared to the murky criteria of the lodestar approach, contingent fee compensation is vastly superior.
In Activision, Judge Patel surveyed the attorney fee awards in common fund cases (most of them purporting to apply the lodestar approach), and concluded that while the awards ranged from 11 to 46 percent, 30 percent must be the fair and reasonable percentage because it was the most frequent.
An arbitrary and inflexible percentage contingency is, whatever its virtues, still arbitrary, and measures only one dimension: the judge’s own assessment of what constitutes a fair fee. The widely varying percentages awarded (the range in Warner Communications was 20 to 50 percent) indicates that agreement on appropriateness of a particular percentage is unlikely to exist. Even if there were such agreement, however, it is in the nature of judges and lawyers to find facts or circumstances that warrant upward or downward adjustments from the standard. Indeed, the Ninth Circuit has, ominously, foreshadowed this characteristic in reviewing an award of fees to class action counsel:
We note with approval that one court has concluded that the “bench mark” percentage for the fee award should be 25 percent. [Citation omitted]. That percentage can then be adjusted upward or downward to account for any unusual circumstances involved in this case.
Paul, Johnson, Alston & Hunt v. Graulty, 886 F.2d at 272.
What progress have the courts, class members or their counsel made if the lodestar is replaced by a “benchmark percentage” and the Johnson-Kerr factors by adjustments upward or downward depending on “circumstances” that can be evaluated only after the completion of the litigation?
It is no answer to the demand for a fixed percentage fee that litigation is uncertain and unanticipated events or “unusual circumstances” might develop which could justify ex post upward or downward adjustments. Life is full of uncertainties far greater than any encountered in litigation. Yet individuals and firms enter into long-term arrangements because the benefits of settled terms outweigh the risks of positive or negative windfalls. In the matter of attorney fees, the benefits of competitive bidding include, among other things, a fee determined prior to litigation and an increased ability on the part of counsel, the parties and the court during litigation to forecast one (perhaps the most) important variable for purposes of strategy, settlement and the like. Successful plaintiffs’ securities lawyers are able to evaluate the value of a case accurately and bring about its prompt resolution. It is precisely because of such lawyers’ knowledge of risk analysis and the time value of money that competitive bidding is appropriate in cases such as those now before the court. It is hardly unfair to hold such lawyers to any bargain they strike short of circumstances amounting to commercial frustration.
Compared to the plaintiffs’ lawyers themselves, and despite the Johnson court’s reference to a trial judge’s “knowledge, experience and expertise” and “past experience as a lawyer and his observation from the bench of lawyers at work,” 488 F.2d at 717-718, judges are poorly suited to determine reasonable fees. Many judges come to the bench with no, or have only distant, experience in private practice; few, if any, come from the ranks of plaintiffs’ lawyers who specialize in high-stakes commercial litigation. Almost never in sitting on the bench does a judge know the hourly rate of the lawyers in court, how much of
All judges can do by way of fixing a percentage contingency, therefore, is to set it at some level that seems “fair.” If the consequences of a judge’s missing the mark and occasionally under- or over-compensating the lawyers were limited to the case before that judge, the randomness of judicial fixing of compensation might be tolerable. But the consequences do not stop there. The effect of uncertainty relating to compensation has been considered in Part II above. Three additional consequences deserve mention.
First, all things being equal, a contingent fee set at a fixed percentage by judicial fiat or by statute or regulation assumes a certain time value of money and thus virtually ensures that delay in the litigation commensurate with the implied interest rate is not penalized. Those cases that are not settled or otherwise resolved as quickly as they would be if the contingent fee were fixed at a lower percentage do not adversely affect the lawyers’ compensation. On the other hand, cases which would be brought if the lawyers were allowed greater compensation are not brought at all if the contingent fee is fixed at an arbitrary and inappropriately low percentage. As one consultant noted:
The essence of establishing fees in an individual case is the effect this will have on incentives in future, roughly comparable cases. Estimation of this effect necessarily involves not simply measuring what has already happened (direct costs) but also trying to estimate the ways in which the fees arrived at in the present case will motivate legal activities in future cases (incentive effects).
In re Folding Carton Antitrust Litigation, 84 F.R.D. at 275-276.20
Second, judicial acceptance of an invariable percentage contingency, like any other scheme of fixing prices, tends to redistribute income in a manner adverse to class members. In cases which would be brought for less than whatever “benchmark” percentage is chosen, be it Judge Patel’s 30 percent or the Warner Communications court’s 25 percent, lawyers receive more in fees than they would have bargained for, to the detriment of the class. Cases not worth bringing for the benchmark would never be brought, even though some plaintiffs may have strong claims. This redistribution favors class defendants. For a court simply to fix a percentage contingent fee without seeking competitive alternatives is thus inconsistent with the court’s fiduciary duty to the class.
Third, and possibly most important, if courts set contingent fees at an arbitrary percentage (thereby eliminating the only quantifiable factor in the selection process), there will be no principled basis upon which to select lead class counsel. Lawyers will apply with self-serving descriptions of their claimed experience and qualifications, and, as they have done here, with a rhetorical sparring match and criticism of other applicants. This is perhaps the greatest vice in the standardized percentage contingency suggested in Activision. It is far better to make the selection of counsel based on the benefits to the class implicit in a firm’s price quote rather than on qualitative
Furthermore, the efficiencies to be gained from the appointment of a single firm as lead class counsel should necessarily result in bids lower than the Activision “benchmark,” a standard so readily accepted by the consortium of plaintiffs’ counsel here when the court first suggested a competitive process.
Accordingly, the court’s selection of lead class counsel will be made on the basis of competitive bids submitted pursuant to the process set forth below.
V. ORDER.
IT IS HEREBY ORDERED THAT each law firm wishing to compete for the position of lead class counsel shall, on or before August 24, 1990, submit an in camera application to the court (1) establishing its qualifications to serve as lead counsel and (2) specifying the percentage of any recovery such firm will charge as fees and costs in the event that a recovery for the class is achieved.
The court envisions that material relating to a firm’s qualifications will consist of detailed descriptions of the role such firm played in each class action it has brought or assisted in bringing and the contribution such firm made to the welfare of the class plaintiffs.
Each firm submitting an application shall certify to the court that its compensation proposal was prepared independently and that no part thereof was revealed to any other bidder prior to filing with the court. Applicants are not to confer in any manner with other firms during the preparation of bids.
Upon receipt of all bids, the court will determine whether supplemental information is necessary.
. The lodestar approach appears to have originated in Lindy Bros. Builders, Inc. of Phila. v. American Radiator & Standard Sanitary Corp., 487 F.2d 161 (3d Cir.1973). The Ninth Circuit took its lead from Johnson v. Georgia Highway Express, Inc., 488 F.2d 714 (5th Cir.1974), which did not use the term "lodestar" but applied a conceptually similar approach. See Kerr v. Screen Extras Guild, Inc., 526 F.2d 67 (9th Cir. 1975). In Moore v. Jas. H. Matthews & Co., 682 F.2d 830 (9th Cir.1982), the Ninth Circuit explicitly endorsed a blend of pure lodestar and Johnson-Kerr. The distinctions between these formulations are without material difference since they all call for the same thing: fixing hourly compensation that is "reasonable" in light of various factors that "often seem to take on the character of so much Mumbo Jumbo,” In re Union Carbide Consumer Products Business Securities Litigation, 724 F.Supp. 160, 170 (S.D.N. Y.1989). This Order uses the term "lodestar” to refer to all manifestations of the hourly fee approaches.
. Counsel for all parties should not ignore the judicial dissatisfaction with the lodestar so apparent in this district and elsewhere. The undersigned can speak only for himself, but it is clear many judges are searching for a different way. For reasons soon to be explained, the undersigned adopts an approach different from Judge Patel’s suggestion in Activision, but joins in her condemnation of the lodestar.
. These lawyers, many of whom claim expertise in the anti-trust laws and to all of whom those laws apply, see Goldfarb v. Virginia State Bar, 421 U.S. 773, 788, 95 S.Ct. 2004, 2013, 44 L.Ed.2d 572 (1975), have displayed a rather cavalier indifference to at least the spirit of the anti-trust laws.
. An exception appears to have occurred in that zenith of the lodestar, In re Fine Paper Antitrust Litigation, 751 F.2d 562 (3d Cir.1984). But the involvement of plaintiffs simply underscores the inherent incongruity of the lodestar approach. Plainly, it is unfair when class counsel (whose efforts, usually against the vigorous opposition of defendants, have created a fund) are required to fight with the class in order to get paid.
. One distinguished jurist went so far as to warn of the “grave danger that the bench and bar will be brought into disrepute" absent a lodestar approach. Cherner v. Transitron Electronic Corp., 221 F.Supp. 55, 61 (D.Mass.1963) (Wyzanski, J.). But the interests of the class should be foremost, not concern for the reputation of the bench and bar. Experience over the last quarter century, in any event, demonstrates that the lodestar is a source of, not savior from, such disrepute. Court Ordered Attorney Fees, 108 F.R.D. 237 (3d Cir. Task Force 1985).
At bottom, public disapproval of class actions arises from the perception that lawyers profit undeservedly from them. Class actions would not be possible without the use of scarce public judicial resources, and lawyers use such resources free of charge. Charging lawyers the value of those resources would obviously alleviate the risk that lawyers might obtain excess profit from class actions.
. A substantial settlement with high legal fees and expenses might well be better for the class than a long-delayed and smaller settlement, a larger percentage of which goes to plaintiffs. The critical factors in evaluating a settlement are the timing of settlement opportunities and amounts left “on the table.” A court will almost never have reliable information on these factors.
. Not the least of the differences between the risk/reward assessments of court and counsel is the former’s failure, typically, to compound any interest on the fee award and thus to take into account the time value of money. See In re Folding Carton Antitrust Litigation, 84 F.R.D. 245, 275 (N.D. Ill.1979).
. See, e.g., the assessments in In re Fine Paper Antitrust Litigation, 98 F.R.D. 48, 110 (E.D.Pa. 1983); Rothfarb v. Hambrecht, 649 F.Supp. 183, 189-197 (N.D.Cal.1986); In re Capitol Underwriters, Inc. Securities Litigation, 519 F.Supp. 92, 105 (N.D.Cal.1981). The idea that a retrospective award of attorney fees — in judicial parlance, "weighing the issues in light of a fully developed record” — protects the class against overreaching lawyers assumes that lawyers who prosecute class actions are unaffected by the uncertainty surrounding their compensation which stems from an ex post determination.
. Because late resolution and wasteful efforts tend to serve the economic interests of defendants’ lawyers, who in cases such as these tend to be compensated by the hour, the defense bar does not provide an effective check.
. There is another reason to disfavor retrospective fee determination, wholly apart from its potential impact on ultimate results. If the court were acting as a private fiduciary, the law would require it to obtain the best price the market would yield for the services of the class’ lawyers. The only way to create a competitive market for legal representation of the class is to require the lawyers to price their services in advance of the litigation rather than afterward.
. This is an inference drawn from the Law of Large Numbers. See W. Feller, An Introduction to Probability Theory and its Applications, Vol I 152-153 (3d ed.1968). Litigation outcomes are the product of many individual determinations (e.g., parties, lawyers, judges, juries, insurers, among others) and thus tend toward the probable outcome, id. at 152, while the fee determination under the lodestar is to a large degree the product of one person’s discretion. The latter is very much more unpredictable.
. In seeking lead counsel status, plaintiffs’ lawyers are pursuing a potentially valuable franchise. While the need to prosecute the claims of the class collectively rather than individually may create a so-called “natural monopoly,” this does not imply that a court’s regulation of the franchise should not make use of the mechanisms of a competitive market. H. Demsetz, Why Regulate Utilities?, 11 J. L. & Econ. 55 (1968).
. For an interesting perspective on the legal profession, see R. Abel, American Lawyers (1989) (especially pp. 40-211).
. The absence of price competition and efforts to exclude non-lawyers from providing certain kinds of legal services (e.g., estate planning, tax advice) may contribute to the public’s low opinion of lawyers. See R. Abel, supra n. 13, at 163-164.
. Occasionally, the lawyer and client agree to more than one contingency (e.g., settlement, trial, appeal) with differing percentages, fixed amounts or some combination for each contingency.
. For example, if a defendant offers a settlement of $100,000 and a lawyer whose contingent fee is 30 per cent of the recovery is sure that by going to trial he can obtain a judgment for $120,000 or more, but at a cost of $8,000 in time and expenses, the lawyer’s interest is to accept the settlement while the client's interest is to reject it. The client is better off with a trial (70% of $120,000 = $84,000, while 70% of $100,-000 = $70,000) whereas the lawyer is not (the extra $8,000 in costs exceeds the lawyer’s marginal gain from trial of $6,000 or 30% of $20,-000). Chesny, 720 F.2d at 477. This is usually referred to as the "sell-out settlement" and is simply a manifestation of the principal/agent problem.
. The lodestar approach to a large degree reflects the influence of a most unlikely source for invigorating economic competition within the legal profession: the American Bar Association. See Johnson v. Georgia Highway Express, Inc., 488 F.2d 714, 719 (5th Cir.1974).
. Another district judge performed a survey of reported decisions similar to that of Judge Patel and came up with justification for a fee under 25%. In re Warner Communications Securities Litigation, 618 F.Supp. 735, 750 (S.D.N.Y.1985).
. In fact, the possibility of upward or downward adjustments protects not class members but their lawyers, who are far better able to advance their own interests in the litigation and are much more practiced than class members at doing so.
. By inclination, judges áre wont to look for the "fair, just and reasonable” level of attorney compensation. However, what might by conventional principles be regarded as unconscionably high attorney compensation might very well have value: such awards attract lawyers to bring cases that would be neglected if only "reasonable” fees were always rewarded. Assuming that a competitive market for legal services is encouraged, appropriate levels of attorney compensation will obtain for all claims in due course.
. Selection of lead derivative counsel will be postponed until such time as it appears that the derivative actions will proceed and that separate counsel is desirable or necessary.
. An applicant may specify alternate contingent events and the corresponding percentages to be charged. If so, the applicant shall also provide an estimate of the amount of recovery at each contingent event and the basis for that estimate.