In re California Micro Devices Securities Litigation

168 F.R.D. 257, 1996 WL 431736

• Court: District Court, N.D. California
• Decided: February 2, 1996
• Docket: No. C-94-2817-VRW
• Status: Published

Opinion

ORDER

WALKER, District Judge.

The first year and some months in the life of this securities class action have been spent in pursuit of a “settlement” which is now before the court for provisional approval.

At the outset of this litigation, the court advised, in an order dated October 19, 1994, that it contemplated an early case management conference to select through competitive bidding one of the numerous plaintiff law firms to serve as class counsel. Led by Lieff Cabraser Heimann & Bernstein (“LCH & B”), the plaintiff lawyers requested postponements of the conference and used these delays for “[n]early round-the-clock” negotiations, Reporter’s Transcript, January 19, 1995, at 14, with lawyers from Howard Rice Nemerovski Canady Robertson Falk & Rabin, representing defendant California Micro Devices (“CAMD”). Although these negotiations did not at first produce a settlement, the LCH & B and CAMD lawyers warned that failure to settle the litigation would lead to CAMD’s bankruptcy. Id at 13-14; 27.

In spite of the danger that such an early resolution might be “a premature, even a collusive,” one, January 13, 1995, Order at 2, the court nonetheless attempted to discharge its responsibilities under FRCP 23(c) and (e) as well as heed counsel’s dire warnings. The court did so by seeking competitive bids to represent the class while permitting settlement discussions to go forward. In March 1995, only two of the seventeen law firms claiming to represent a class submitted proposals and only LCH & B’s looked serious. The lawyers thereafter produced the settlement. It consisted of only a small portion of cash and new securities of uncertain value with resulting dilution of CAMD shareholders’ stake in the corporation.

Concerned about such a settlement and the apparent lack of competitiveness of the bids, the court declined to appoint LCH & B as class counsel or provisionally to approve the settlement but, instead, afforded LCH & B and CAMD the opportunity to demonstrate that the proposed settlement and the fees proposed in LCH & B’s bid enjoyed affirmative support of the purported class, as opposed to mere tacit toleration, and to seek approval of the settlement and the bid on that basis. See August 4, 1995, Order. In accordance with FRCP 28(d)(2) (“for protection of the members of the class or otherwise for the fair conduct of the action, [the court may require] that notice be given in such a manner as the court may direct to some or all of the members of any step in the action * * * or of the opportunity of members to signify whether they consider the representation fair and adequate, to intervene and present claims or defenses, or otherwise to come into the action”), the court suggested a survey be made of significant class members. Id at 16-17.

The results of this survey have convinced the court that this litigation should not proceed as a class action with any of the named plaintiffs offered by the plaintiff law firms as class representatives and that a truly independent class representative must be appointed to engage class counsel and to conclude any settlement or direct the litigation. Such a plaintiff is available to the class in the Colorado Public Employees Retirement Fund (“COLPERA”). The court therefore DENIES the motions of LCH & B and CAMD for preliminary approval of the settlement, declines to appoint LCH & B class counsel and finds that none of the named plaintiffs will fairly and adequately represent the proposed class. The court further CERTIFIES plaintiff Colorado Public Employees’ Retirement Association as class representative. FRCP 23(a).

I

Permitting class counsel who are not effectively monitored to prosecute a class action is the functional equivalent of allowing that counsel to serve as both class representative and class' attorney. Such a situation directly implicates the danger of collusion between plaintiff and defense counsel recognized in FRCP 23, which assigns to the courts both broad responsibility and broad power to monitor the conduct of class actions to ensure their essential fairness. See especially FRCP 23(a), (e). For their part, courts have recognized this danger of collusion and have accordingly applied their Rule 23 mandate aggressively; court supervision of class actions has over time created a rich jurisprudence which contains several per se rules whether a given class representative adequately represents the class.

One of the most fundamental of these rules addresses the situation discussed above: an attorney may not serve both as class representative and as class counsel. See, for example, Susman v. Lincoln American Corp., 561 F.2d 86, 90-92 (7th Cir.1977) (collecting cases). This rule has direct application in this case. When class counsel are not effectively monitored by the class representative, the result is indistinguishable from the situation in which an attorney serves as both class counsel and class representative. As will be discussed in parts II and III below, the contents of the settlement currently before the court, and the manner in which it was achieved, convince the court that LCH & B has been prosecuting this action with little or no monitoring from the putative class representatives. LCH & B has thus effectively appointed itself class representative and violated the above rule.

A

A critical factor in determining whether a settlement is worthy of court approval under FRCP 23(e) is whether the class has been “fairly and adequately” represented by the class representative during settlement negotiations. See, for example, In re General Motors Corporation Pick-Up Truck Fuel Tank Products Liability Litigation, 55 F.3d 768, 784 (3d Cir.), cert. denied sub. nom. General Motors v. French, — U.S. -, 116 S.Ct 88, 133 L.Ed.2d 45 (1995) (noting, when considering adequacy of proposed settlement, that “[t]he protection of absentees’ due process rights depends in part on the extent the named plaintiffs are adequately interested to monitor the attorneys * * * and also on the extent that the class representatives have interests that are sufficiently aligned with the absentees to assure that the monitoring serves the interests of the class as a whole.”); In re Corrugated Container Antitrust Litigation, 643 F.2d 195, 208-10 (5th Cir.1981) (analyzing whether class representative fairly and adequately represented class during settlement negotiations); Weinberger v. Kendrick, 698 F.2d 61, 73 (2d Cir.1982), cert. denied sub. nom. Coyne v. Weinberger, 464 U.S. 818, 104 S.Ct. 77, 78 L.Ed.2d 89 (1983) (same).

Particular scrutiny is required under FRCP 23(e) when the proposed settlement has been negotiated prior to class certification, as was the case here. Judge Posner explained in Mars Steel Corp. v. Continental Illinois National Bank & Trust, 834 F.2d 677, 681-82 (7th Cir.1987):

[W]hen class certification is deferred, a more careful scrutiny of the fairness of the settlement is required. * * * The fairness of a settlement of a legal dispute is like the adequacy of the consideration supporting a contractual promise: a matter best left to negotiation between the parties. A settlement is a contract, and normally the test for the fairness of a contract is strictly procedural: were the parties competent adults duly apprised of the basic facts relating to their transaction? The problem in the class-action setting, and the reason that judicial approval of the settlement of such an action is required, * * * is that the negotiator on the plaintiffs’ side, that is, the lawyer for the class, is potentially an unreliable agent of the principals. * * * Ordinarily the named plaintiffs are nominees, indeed pawns, of the lawyer, and ordinarily the unnamed class members have individually too little stake to spend time monitoring the lawyer—and their only coordination is through him. The danger of collusive settlements * * * makes it imperative that the district judge conduct a careful inquiry into the fairness of a settlement to the class members before allowing it to go into effect and extinguish, by the operation of res judicata, the claims of class members who do not opt out of the settlement.

(internal citations omitted).

Parties who bypass the certification process deny the court the opportunity to determine whether the putative class representative is capable of “fairly and adequately” representing the class which usually arises at the class certification hearing. See In re GM Fuel Tank Litigation, 55 F.3d at 787-88. (“Ordinarily, a court relies on class status, particularly the adequacy of representation required to maintain it, to infer that the settlement was the product of arm’s length negotiations. * * * Where the court has not yet certified a class or named its representative or counsel, this assumption is questionable.”). The problems unique to pre-certification settlements thus implicate the adequacy of the class’ representation. Id at 788; see also In re Baldwin-United Corp., 105 F.R.D. 475, 480 (S.D.N.Y.1984) (“Arguments in opposition to settlement classes have merit when they are addressed to the problem of inadequate representation or possible collusion among the named plaintiffs and some or all defendants.”); Mars Steel, 834 F.2d at 680; Malchman v. Davis, 706 F.2d 426, 433 (2d Cir.1983); Weinberger, 698 F.2d at 73.

' Before the court may even look at the terms of the proposed settlement to determine if those terms are fair and reasonable, then, it must first determine whether the putative class representatives “fairly and adequately protected] the interests of the class” during settlement negotiations. FRCP 23(a)(4); see also General Telephone Company of Southwest v. Falcon, 457 U.S. 147, 160, 102 S.Ct. 2364, 2372, 72 L.Ed.2d 740 (1982) (“actual, not presumed, compliance with Rule 23(a) remains, however, indispensable.”). The Rule 23(e) and 23(a)(4) inquiries are not entirely distinct, however; as will be seen in section III, infra, the terms of a settlement may be so defective that they themselves provide compelling evidence that the class representative did not “fairly and adequately” protect the interests of the class during settlement negotiations.

B

Since the settlement currently before the court was reached before class certification, Rule 23(e) requires the court to examine closely whether the putative class representatives have “fairly and adequately” monitored the conduct of LCH & B and the other plaintiff attorneys in this case.¹ During settlement negotiations the putative class representatives’ primary duty was to ensure that LCH & B did not sacrifice the interests of the class in order to maximize its own recovery from the instant litigation. The seriousness of the risk that LCH & B would do so is reflected in the long line of cases which have prevented attorneys from serving as both class representatives and class counsel. See, for example, Shroder v. Suburban Coastal Corporation, 729 F.2d 1371, 1375 (11th Cir.1984); Lowenschuss v. Bluhdorn, 613 F.2d 18, 20 (2d Cir.), cert denied, 449 U.S. 840, 101 S.Ct. 117, 66 L.Ed.2d 46 (1980); Zylstra v. Safeway Stores, Inc., 578 F.2d 102, 104 (5th Cir.1978); Susman v. Lincoln American Corporation, 561 F.2d 86 (7th Cir.1977); Turoff v. May Co., 531 F.2d 1357, 1360 (6th Cir.1976). The risk recognized by these courts and inherent in having an attorney wear the hats of both class representative and class counsel is obvious: ordinarily, class counsel stands to gain much more in fees from a settlement than any individual class member stands to gain from the settlement itself. If one attorney played both roles, he would be sorely tempted to sacrifice the interests of his fellow class members in favor of maximizing his own fees.

This danger of class counsel self-dealing is not, of course, limited to the situation in which an attorney serves as both class representative and class counsel; rather, it is present in every situation where class counsel is allowed to prosecute an action and negotiate settlement terms without meaningful oversight by the class representative. In response to this danger, courts have developed monitoring techniques designed to make themselves “surrogate clients” capable of ensuring that class action litigation does not become a vehicle which serves class counsel before all others. One such technique is careful and searching implementation of the certification and settlement approval mechanisms found in FRCP 23 itself. See, for example, Mars Steel, 834 F.2d 677. Another such technique is the competitive bidding process employed in In re Oracle Securities Litigation, 131 F.R.D. 688, 132 F.R.D. 538 (N.D.Cal.1990) and In re Wells Fargo Securities Litigation, 156 F.R.D. 223 (N.D.Cal.1994), which enables the court to monitor fee arrangements between class counsel and class members.

The competitive bidding process by itself is inadequate to cope with the dangers inherent in a pre-certification settlement, however, because it does not prevent settlement class counsel from obtaining an edge in the bidding process by compromising the interests of absent class members and being the first to settle with defendant. This appears to have happened here; LCH & B spearheaded the settlement negotiations which led to the proposed settlement and LCH & B was the only plaintiff firm to present a serious bid in competitive bidding. LCH & B’s pursuit of an early settlement in this case thus gave it an edge in the bidding and frustrated the court’s efforts to use such bidding to ensure that the class was “fairly and adequately represented.”

C

It was primarily in order to compensate for the manner in which the pre-certification settlement compromised the competitive bidding process in this case that the court in its August 4,1995, Order directed LCH & B and CAMD to show that the proposed settlement had the affirmative support of the class. As will be explained in section II, the showing made by the parties has failed to convince the court of this fact and, indeed, has led the court to question whether court monitoring of class actions can ever serve as an adequate substitute for an informed and engaged class representative. As will be explained in section III, however, the court need not face the consequences of such a conclusion in this case; while the court finds that the named plaintiffs do not qualify as class representatives under FRCP 23(a)(4) because they have shown no ability or inclination to monitor LCH & B, the court does find that an adequate class representative has come forward.

II

In its August 4, 1995, Order, the court required LCH & B and CAMD to show that the proposed settlement had the affirmative support of the class members. The court had hoped that by making such a showing, LCH & B and CAMD could assure the court that the proposed settlement was non-collusive, and thus that the class representative had “fairly and adequately” represented the class during the settlement negotiations.

The showing made by CAMD and LCH & B in support of the proposed settlement revolves around the analysis conducted by LCH & B’s expert, Mark S. Kangas of Princeton Venture Research. Mr. Kangas first attempted to show support for the settlement by identifying 44 institutional investors who on June 30, 1994—the closing date of the quarter immediately preceding the first disclosure of CAMD’s improper bookkeeping on August 3, 1995—owned approximately 55 percent of CAMD’s outstanding common stock. Surveys were sent to these 44 firms, the vast majority of whom indicated support for both the settlement and the proposed attorney fees. As a result of this support, Mr. Kangas claims that “[ijnstitutions owning 38.4 percent of the total shares outstanding as of June 30, 1994 responded favorably to the proposed Settlement, and 35.1 percent responded favorably to the suggested attorney’s fees.” Because “[ijnstitutions with large holdings in the June 30,1994 quarter are likely to represent a significant amount of the claims in this proposed Settlement,” Mr. Kangas concludes that the court can infer approval of the proposed settlement from the settlement class as a whole from the approval of these 44 firms.

Before discussing in detail the limitations of LCH & B’s showing, the court must first make an important distinction concerning the definition of “class member,” a distinction which lies at the very heart of the limitations of that showing. The putative class in this case is defined as consisting of all persons who purchased CAMD securities between September 7, 1993, and January 9, 1995. In order to recover from CAMD under Rule 10b-5, putative class members must show that they suffered out-of-pocket damages as a result of CAMD’s alleged misrepresentations. See Blackie v. Barrack, 524 F.2d 891, 908-09 (9th Cir.1975), cert. denied, 429 U.S. 816, 97 S.Ct. 57, 50 L.Ed.2d 75 (1976); Green v. Occidental Petroleum Corp, 541 F.2d 1335, 1341-46 (9th Cir.1976) (Sneed, concurring); Wool v. Tandem Computers Inc, 818 F.2d 1433, 1436-38 (9th Cir.1987). The out-of-pocket rule “fixes recoverable damages as the difference between the purchase price and the value of the stock at the date of purchase.” Wool, 818 F.2d at 1437 (internal quotations omitted). For in and out traders, both the time of purchase and the time of sale are considered when determining whether out-of-pocket damages were suffered: “[f]irst, the spread between market price and value of the stock at the time of purchase is determined. When the purchaser sells the stock, the spread is again measured. If the spread has diminished, the purchaser would recover the differences.” Id at 1437-38, n 4.

In order to show that they had suffered damages and thus are entitled to participate in the proposed settlement, then, the putative class members who are in and out traders must show that the spread between the price of CAMD stock and the true value of CAMD stock was smaller at the time of sale than it was at the time of purchase. All of the putative class members whose spread did not shrink between the time of purchase and the time of sale suffered no harm due to CAMD’s misrepresentations and thus have no claim on any settlement offered by CAMD; they are not, in other words, claims class members.

LCH & B has ignored the distinction between class members and claims class members; indeed, LCH & B has failed to present the court any information at all regarding the true value of CAMD stock during the class period. While the court in its August 4, 1995, Order directed the parties to show support for the proposed settlement among “class members” and did not expressly draw the distinction between class members and claims class members, the distinction is a vital one and should have occurred to anyone attempting to put forth a showing of class support for the settlement; since only claims class members would be prejudiced by a poor settlement, only their support for the settlement should be accorded significant weight. By making its showing of class support for the settlement without observing this distinction between class members and claims class members, LCH & B undercut its own showing.

Another problem with LCH & B’s analysis becomes apparent after a moment’s reflection. One manner in which LCH & B apparently attempted to identify potential class members was by looking at holdings on June 30,1994. Because LCH & B surveyed investors that owned approximately 55 percent of CAMD common stock outstanding on that day, LCH & B claims that the surveyed group represents a “significant” amount of the potential claims. This analysis, however, ignores all potential claims class members who did not yet hold stock on June 30, 1994. Since CAMD’s stock price was rising throughout the month of July 1994, this is a potentially large number of claims. The court thus is not convinced that support of investors who held a majority of shares on June 30,1994, is synonymous with support of a majority of claims class members.

Apparently aware of these problems, Mr. Kangas attempted to sharpen his analysis by estimating the number of shares owned by the 44 surveyed investors on August 3, 1994, the day before the first disclosure. These estimates were achieved by comparing share ownership on June 30, 1994, and September 30, 1994, and pro-rating any change in ownership over the intervening three months according to reported trading volume. Mr. Kangas claims that this analysis reveals that either 3.6 or 5.9 million (depending on the date on which the analysis is ended) of the shares owned by the 44 surveyed firms were purchased during the settlement period and held on August 3, 1994. LCH & B claims that this in turn shows that the surveyed investors held 60 percent of the shares purchased during the first part of the settlement period and held on August 3,1994.

The court, frankly, fails to follow LCH & B’s logic, and is baffled how the 60 percent figure was produced. Which estimate of share ownership, 3.6 million or 5.9 million, was used as a numerator and why? LCH & B and Mr. Kangas do not explain.

The court has another, more fundamental objection to LCH & B’s method of analysis. Mr. Kangas proposes to identify share ownership on August 3, 1994, by looking at ownership on June 30 and September 30 of that year and pro-rating any change in ownership over the intervening three months by reported share volume. For example, Fred Alger Asset Management (“Alger”) is reported as holding 1,104,700 shares on June 30, 1994, and 1,194,600 shares on September 30, 1994. Alger’s share ownership on August 3, 1994, would thus be calculated by pro-rating the 89,900 share increase between quarters across the third quarter according to trading volume and then looking at what figure this pro-rating produced for August 3.

The problem inherent in this method is the rather large assumption that trading on any given day was uniform among share holders. This method thus ignores the possibility that any given surveyed investor in fact liquidated a large chunk of the stock held on June 30, 1994, before August 3 and then bought the stock held on September 30, 1994, after August 3, when the price was relatively low. In other words, Mr. Kangas’ analysis ignores the possibility that a surveyed investor successfully sold high and bought low and is in fact not a claims class member at all.

The problems identified above inhere not so much in the fact that LCH & B chose the wrong method of estimating claims class membership, but rather in the fact that LCH & B chose to identify claims class membership by estimation in the first place. The court simply is not satisfied that Mr. Kangas’ very rough-and-ready identification of the surveyed institutional investors as a large chunk of the claims class is a satisfactory alternative to factually-based identification of such class members. One might expect the survey forms circulated among the 44 institutional investors to shed some light on this factual question. The survey forms used by LCH & B disappoint this expectation.

Because LCH & B has made no effort to survey only those investors who suffered a loss as a result of CAMD’s alleged fraud, there is no way to tell from the submitted survey forms which of the responding investors are actually claims class members. A majority of the surveyed institutions were in- and-out investors during the proposed class period. The survey form which provides the basis for LCH & B’s current claims of support is quite brief: it asks only how many shares of CAMD stock were purchased in the proposed class period of September 7, 1993, through January 9, 1995, how many of those shares were later sold and when they were sold. Because LCH & B has not attempted to learn more accurately when stock was purchased or to determine the value line for CAMD stock during the class period, it is possible that an in and out investor whom LCH & B is currently assuming is a class member did not have its spread shrink between the time of sale and the time of purchase and thus suffered no harm as a result of CAMD’s misrepresentations.

Turning again to the example of Fred Alger Asset Management, it purchased 125,600 shares of stock during the class period, and sold all of that stock at “various” times, according to its survey response. The court currently cannot determine whether the spread between the price of these shares and their value shrank or grew between the time of purchase and the time of sale, and thus whether Alger even has a 10b-5 cause of action against CAMD. Yet Alger is counted by LCH & B as an approving class member.

Since LCH & B and CAMD have made no effort to show that Alger or the other approving investors are claims class members, the court concludes that their approval of the settlement is entitled to little weight when determining whether the proposed settlement enjoys the affirmative support of the class.

If the methods employed by LCH & B were the only methods available for identifying claims class members, the court might accept LCH & B’s current showing as adequate. The court is confident, however, that other methods do exist through which LCH & B could more accurately have identified the claims class members. To name just one example, 15 USC § 78m(d) requires a person whose stock purchases have given him ownership of more than 5 percent of a class of stock to make a number of disclosures to the SEC within ten days of that purchase. Scrutiny of these 13(d) reports, which would give a more accurate picture of the trading activities of at least some of the investors which are purported class members, could have helped define more precisely which of the surveyed investors are claims class members. Alternatively, LCH & B could have simply asked the surveyed investors for a breakdown of their day-to-day trading activity during the class period, a request which would have to be made at some point in any event in order to determine eligibility for allocation of the settlement proceeds. Finally, LCH & B could have attempted at least to estimate the true value of CAMD stock during the class period.

LCH & B and CAMD have done none of the above. Given the delays which have already occurred in this ease, the court is not inclined to give them another opportunity to do so. The court therefore concludes that LCH & B and CAMD have failed to show that the proposed settlement enjoys the “affirmative support, as opposed to the silent toleration,” of a majority of class members.

B

Even if the court were convinced that the showing put forth by LCH & B and CAMD represented support from a significant number of claims class members, many important omissions in the polling letter would prevent the court from according much significance to this showing.

The court did not dictate to the parties how they had to conduct the polling in this case; rather, the court allowed the settling parties to make their ease for the settlement in the way they saw fit. The result of this decision was a polling letter which inadequately describes the current state of this action. These inadequacies are explained in detail below.

As several class members have pointed out, the polling letter is less than candid in its disclosure of the current financial status of CAMD and the lawsuits it is facing. See generally COLPERA Application to Appear at 2; Letter of Professor Joseph A. Grundfest on behalf of class members Bank of America National Trust and Savings Association and Wells Fargo Institutional Trust Company and other institutional investors² (“Institutions Letter”) at 7.

First, the polling letter claims that “[s]ignificantly, the Outside Directors were not named as defendants in any pending lawsuit,” when those directors are in fact named as defendants in a derivative suit currently pending in Santa Clara County superior court. LCH & B has since claimed that “any pending lawsuit” was intended to refer only to the suits comprising this class action and has asserted that even if the language in the polling letter was misleading, “it was probably understood in context.” The court does not share LCH & B’s sanguine assumption that a reader of the polling letter would understand “any pending lawsuit” really to mean “any pending federal lawsuit.”

Second, the polling letter fails to disclose the existence of the above-mentioned derivative action, brought against the outside and inside directors in Santa Clara County superior court (Perry v. Desaigoudar, No. CV749829). In response to this objection, LCH & B blithely asserts that this failure to disclose is harmless because the derivative action is meritless and certain to be dismissed. Such an assessment is properly made by class members, not by putative class counsel without the input of class members.

Third, the polling letter fails to include information regarding CAMD’s stock sale to and strategic partnership with Hitachi Metals Corporation during the class period. Nor does the polling letter mention the subsequent “readjustment” of that partnership after the disclosure that triggered this litigation and settlement of Hitachi’s claims. An investor weighing the value of the proposed settlement would be very interested in the terms of the Hitachi settlement; even though Hitachi is not a potential class member, the settlement reflects at least partially the value which CAMD places on these claims.

Criticism of the polling letter’s failure to disclose the Hitachi settlement is not blunted by LCH & B’s various responses to it. These include the arguments that (1) Hitachi cannot be a class member in any event, since it purchased its shares directly from CAMD and not on the open market; (2) even if Hitachi were a class member, it received a worse deal than the class members will; and (3) the information concerning the Hitachi settlement was a matter of public record and anyone interested in the terms of the settlement could discover them with the click of a mouse. See LCH & B’s Reply Memorandum in Support of Partial Settlement at 5-6.

LCH & B’s first argument has already been addressed. Its second argument rests on the assumption that since the class members will receive more than Hitachi, they would not be interested in the terms of the Hitachi settlement. This assumption reduces to the assumption that class members are interested only in maximizing recovery. As will be discussed below, there is much reason to question this assumption when the class member in question is an investor who is a repeat player in the securities market. Such an investor might find the very superiority of the proposed settlement to the Hitachi settlement troubling, since it suggests that the company’s overriding priority is to settle quickly the current actions rather than ensuring that those who actually engaged in the wrongdoing are forced to pay its wages,

LCH & B’s final argument goes right to the heart of the problems which bedevil investor monitoring of settlements. LCH & B asserts that the Hitachi settlement need not have been disclosed in the polling letter because that information was publicly available. This assertion assumes that the polled investors had an incentive to seek out information which would allow them to analyze critically the merits of the proposed settlement. Most investors know, however, that proposed settlements are routinely approved in securities class actions. See Institutions Letter at 2 (“once plaintiff and defense counsel agree to settle a securities class action, there is typically no one before the court with an incentive to challenge the merits of the settlement or of the underlying claim.”); Elliot J. Weiss and John S. Beckerman, Let the Money Do the Monitoring: How Institutional Investors Can Reduce Agency Costs in Securities Class Actions, 104 Yale LJ 2053, 2104 (1995) (“Objecting to settlements also has been a low-percentage proposition.”). Given the likelihood that the proposed settlement would be approved regardless of objection, most investors would likely conclude that the independent investigation suggested by LCH & B would be a waste of time and resources. The only rational course for the polled investors was thus to support the settlement in the hopes of quick court approval, since such approval would maximize the time value of the settlement monies. Other courts have recognized that class members are subject to this sort of pressure. See, for example, In re GM Fuel Tank Litigation, 55 F.3d at 789 (when settlement is negotiated prior to class certification, “even if [putative class members] have enough information to conclude the settlement is insufficient and unsatisfactory, * * * the mere presentation of the settlement notice with the class notice may pressure even skeptical class members to accept the settlement out of the belief that, unless they are willing to litigate their claims individually—often economically infeasible— they really have no choice.”); Mars Steel, 834 F.2d at 680-81 (“where notice of the class action is * * * sent simultaneously with the notice of settlement itself, the class members are presented with what looks like a fait accompli.”). The fact that few of the polled investors bothered to fill in even the two “comments” lines which were provided in the polling form supports the court’s suspicion that the polled investors felt that they were presented with a done deal.

The fourth and final way in which the polling letter is misleading is in the way it summons the specter of CAMD’s impending bankruptcy if the proposed settlement is not promptly accepted. The letter also refers darkly to “confirmatory evidence” of this bankruptcy shared by CAMD with LCH & B. The court is especially troubled by this use of unverifiable yet supposedly confirmed threats of bankruptcy. Because it is so patently in the best interests of both LCH & B and CAMD for the court and the shareholders to believe that the only choice is between settlement and bankruptcy, it is difficult for the court to put much credence in LCH & B’s claim of “confirmatory evidence.” Yet the court is in no position to evaluate the threat of bankruptcy for itself as long as the parties choose not to share their “confirmatory evidence” with the court. Nor are the polled investors capable of such an evaluation as long as LCH & B refuses to share this “confirmatory evidence” with them. See COLPERA Application to Appear at 5 (class member rebuffed in its request to review the “confirmatory evidence” referred to in the polling letter). LCH & B has thus frustrated independent analysis of the risk of CAMD’s bankruptcy even as it used that risk of bankruptcy as an argument in support of the proposed settlement.

In sum, even if LCH & B and CAMD had shown that a majority of claims class members approved of the proposed settlement, the court would still have to disregard this showing due to several critical omissions in the polling letter. The court therefore concludes that LCH & B and CAMD have failed to show that the proposed settlement has the affirmative approval, as opposed to mere passive toleration, of the putative class members. The motion of LCH & B and CAMD for approval of the proposed settlement is DENIED.

In addition to eliciting the showing discussed above, the court’s August 4, 1995, order had another, felicitous effect: it prompted several class members and indirectly affected parties to send the court cogent observations and criticisms of both the proposed settlement and the manner in which that settlement was reached.³ This sort of input from actual class members and indirectly affected parties who have hired independent counsel, as opposed to the figurehead plaintiffs who are essentially hired by class action lawyers, is very unusual in securities class actions.

These critical responses have convinced the court of two points. First, the proposed settlement contains several troubling features which suggests that it was negotiated in quasi-collusive circumstances; these features necessitate its rejection by the court on grounds altogether distinct from LCH & B’s and CAMD’s failure to marshal convincing affirmative support from the class. Second, the best way to ensure that class counsel is adequately monitored in this ease is to certify as class representative an independent and informed institutional investor. These two points will be taken up next.

Ill

The failure of the LCH & B and CAMD convincingly to show that the proposed settlement enjoys genuine, informed support among claims class members means that the court cannot rely on such support as evidence that the putative class representatives “fairly and adequately” represented the class during the settlement negotiations. Because the court cannot draw this conclusion from the polling results, it must now analyze for itself whether the putative class representatives have “fairly and adequately” represented the interests of the class in this litigation.

“Fair and adequate” representation requires that the class representative be both willing and able to monitor closely the conduct of class counsel during settlement negotiations. The great risk of settlements such as the one currently before the court is that the settlement terms will serve more the interests of class counsel than the needs of class members. This risk arises because “[sjhareholders with well-diversified portfolios or small holdings lack the incentive and information to police settlements—the cost of policing typically outweigh any pro rata benefits to the shareholder.” Bell Atlantic Corp v. Bolger, 2 F.3d 1304, 1309 (3d Cir.1993); see also Greenfield v. Villager Industries, Inc, 483 F.2d 824, 832 n. 9 (3d Cir.1973) (“Experience teaches that it is counsel for the class representative and not the named parties, who direct and manage these actions. Every experienced federal judge knows that any statement[] to the contrary is sheer sophistry.”); Saylor v. Lindsley, 456 F.2d 896, 900 (2d Cir.1972) (Friendly) (“There can be no blinking at the fact that the interests of the plaintiff in a stockholder’s derivative suit and of his attorney are by ho means congruent”).

As will be elaborated below, the proposed settlement has so many troubling aspects that the court is forced to conclude that it was negotiated under quasi-collusive circumstances and that the putative class representatives, by failing to monitor adequately LCH & B and by allowing the proposed settlement to be put before the court, have shown that they are unwilling or unable “fairly and adequately” to represent the interests of the class in this action, as is required by FRCP 23(a)(4).

The first aspect of the proposed settlement which suggests that it was negotiated with insufficient oversight by the putative class representatives is the fact that it contains a relatively small cash component even while CAMD maintains a large cash reserve. As COLPERA has pointed out, CAMD’s Form 10-Q for the period ending June 30, 1995, indicates that its current assets exceed its current liabilities by some $20 million, “thus suggesting that [CAMD] has substantially more than $1 million available to deal with the claims at issue [.] To the extent that the proponents of the settlement may have uncritically accepted the assertion that [CAMD] has no more than $1 million available to settle the instant claims, this case may illustrate the harm to stockholders of a rush to settle putative class action claims before class counsel has been designated.” COLPERA Application to Appear at 2; see also COLPERA Reply Memorandum at 2 (“The statement in the [polling letter] that [CAMD] has only $1 million of ‘available unrestricted liquid assets’ does not rest on any actual restriction, but only on supine acceptance of the unsupported and implausible assertion that Hitachi Metals, Ltd.’s purchase of a general equity interest in the company is not available to pay the company’s debts.”).

The court agrees with COLPERA’s criticism of this aspect of the settlement. LCH & B has apparently failed to assess critically CAMD’s claim that virtually all of its current assets are unavailable for distribution to the class members. Given this apparent lack of critical analysis by LCH & B, the court also agrees with COLPERA’s suggestion that LCH & B was not acting as a zealous advocate of the class members when it negotiated the settlement. Rather, LCH & B apparently was concerned primarily with securing the role of class counsel and thus “offer[ed] favorable terms as a means of securing [that] role.” COLPERA Application to Appear at 2. This aspect of the settlement suggests that greater client monitoring of class counsel than has yet taken place is needed to ensure that the interests of the class are fairly and adequately represented.

The second aspect of the proposed settlement which suggests that it has been negotiated with insufficient oversight by the putative class representatives is the fact that the settlement relieves the outside directors of liability without requiring them to contribute any consideration to the settlement. Again, COLPERA has led the charge in criticizing this aspect of the settlement, noting that “[i]t is in the interests of stockholders that, before stockholder claims are asserted or settled, there be careful and reflective review by would-be class counsel of possible claims against directors because the assertion of such claims, when they are warranted, is likely to be an incentive to effective director ■ performance, and the non-assertion of such claims, when they are not warranted, encourages able men and women to be willing to serve as directors.” COLPERA Application to Appear at 4.

Neither CAMD nor LCH & B have ever explained why the outside directors should be exonerated without contributing to class recovery. The court thus suspects that this aspect of the settlement is one of the concessions which COLPERA identified in the passage quoted in section III A 1 as bargained away by LCH & B in order.to induce CAMD to reach a quick, as opposed to optimal, settlement. Even if this suspicion is not ultimately borne out, the mere fact that suspicion is warranted convinces the court that an independent and informed class representative is required to oversee the actions of class counsel.

The third aspect of the settlement which suggests that it was negotiated with insufficient oversight from the putative class representatives is the fact that its terms appear to be based on LCH & B’s acceptance of CAMD’s claims of imminent bankruptcy should the proposed settlement not be approved. It is now apparent that CAMD is not as financially imperiled as it appeared to be in the two or three months immediately following the filing of this litigation; its stock price, despite the occasional setback common among technology stocks, rose steadily throughout the summer of 1995, and disclosures to the court in the last few months have revealed that the company has cash reserves in excess of $23 million as a result of a strategic partnership entered into with Hitachi. LCH & B and CAMD have nonetheless continued to argue that CAMD is in fragile financial health and that this fragility is a powerful argument in favor of approving the settlement. See, for example, polling letter at 3.

The possibility that CAMD may be in bad financial shape is not really the point; that LCH & B has no incentive to argue or even believe otherwise is. As noted earlier, the court is in no position to verify for itself CAMD’s claims of impending bankruptcy, and LCH & B and CAMD have been less than forthcoming in sharing the “confirmatory evidence.” LCH & B’s incentive to accept uncritically CAMD’s claims of bankruptcy thus creates a serious risk that the current settlement was negotiated in a quasi-collusive manner and that LCH & B thereafter embraced these claims as another argument in support of the proposed settlement. An informed and independent class representative is needed to monitor class counsel and prevent class counsel from being overly eager to accept CAMD’s claims of impending financial doom.

The fourth aspect of this settlement which leads the court to suspect that its approval may not be in the best interests of the class is the proposed levels of attorney fees generated by the lackluster bidding which occurred in this ease.⁴ COLPERA, Bank of America and Wells Fargo have argued that the proposed 20.5% of recovery fee to which LCH & B would be entitled under its bid is excessive in light of the relatively small amount of resources which LCH & B has been required to invest in its putative representation of the class. See COLPERA Application to Appear at 5 (“The proposed 20.5% attorney fee is excessive in view of * * * the fact that [CAMD] has effectively admitted liability by, inter alia, filing an action against its former chief executive in which it asserts claims similar to those asserted against the company herein”); Institutions Letter at 9-10 (“the proposed settlement would * * * reward plaintiffs’ counsel with approximately $2.8 million in fees for the resolution of an action against defendants who are asserted to be blameless. It would generate $2.8 million in fees primarily through the dilution of current shareholders’ interests for the benefit of counsel who have conducted no formal discovery, who have engaged in no motions practice directed at the merits of the claim, and who have behaved in a manner that has been criticized by the Court.”).

LCH & B’s response to the above criticisms is that the results of competitive bidding should be respected and its winning bid of 20.5% is therefore beyond challenge. While the court might be persuaded by this argument in a case in which the successful bidder did not engage in unauthorized settlement negotiations and in which there was genuine and robust competition among the bidding firms, it finds LCH & B’s rejoinder unpersuasive in light of the checkered history of LCH & B’s involvement in this lawsuit in general and in the competitive bidding process in particular. The court agrees with COLPERA, Bank of America and Wells Fargo that LCH & B’s 20.5% fee appears excessive in this case. This is yet another reason why an independent and informed class representative is needed to ensure that the class is fairly and adequately represented.

The final aspect of the settlement which leads the court to suspect that it was negotiated with insufficient oversight by the putative class representatives is the fact that the settlement orders a substantial recovery to be paid from the company’s assets before actions are even instituted against individual defendants whom the company appears to claim are guilty of wrongdoing. Several class members have noted the impropriety of this settlement: CAMD, which all parties seem to agree is largely blameless, would be the sole source of funding the settlement, while its former officers, whom all parties seem to agree are largely to blame for the misstatements to the securities markets, remain untouched until unspecified later action. Professor Grundfest’s letter submitted on behalf of class members Bank of America and Wells Fargo bears quoting at length:

The largest part of the settlement described in the Polling Letter arises from [CAMD’s] agreement to issue 1.5 million shares of its common stock. * * *

This provision in the Settlement Agreement is problematic. The issuance of these 1.5 million shares does not affect the real economic value of [CAMD] as an ongoing enterprise. Instead, this agreement to increase the number of outstanding [CAMD] shares by approximately 17 percent will simply dilute the equity interests of current [CAMD] shareholders by approximately 17 percent, and shift that percentage of equity interest in [CAMD’s] shares to members of the proposed class who are current or former shareholders of [CAMD].

In a worst case situation, current shareholders who acquired their interests during the class period may discover that the proposed settlement dilutes their current holding by 17 percent only to compensate them by precisely the same amount—putting aside attorneys’ fees—as members of the plaintiff class. But attorneys’ fees cannot be put aside, and with respect to such shareholders the proposed settlement will charge a fee of 20.5 percent for shifting economic value from one pocket to the other.

Such a settlement reasonably leaves shareholders with mixed feelings. While a member of the class may today benefit from a settlement, the same investor may tomorrow find herself diluted in another lawsuit when she is not a member of the plaintiff class. In the aggregate, shareholders might rationally prefer to reject all such settlements, and thereby save the attorneys’ fees generated by the transfer of assets from one pocket to another. However, once such a settlement is offered, the only rational shareholder response may be to accept the proposed settlement because shareholders facing an opportunity to collect a real and present settlement cannot bind shareholders facing analogous settlements in cases yet to be filed to reject those proposed settlements. The settlement proposal facing shareholders in this proceeding may thus create a form of “prisoners dilemma” in which the rational noncooperative response is to accept the settlement, whereas the individually and socially superior cooperative result would be to reject it.⁵

Institutions Letter at 8-9 (emphasis added).

There are two groups of plaintiffs in this case who may be expected to view the proposed settlement “in the aggregate.” The first group consists of those class members who are current CAMD shareholders. As the court has identified in previous eases, these “equity” or “retention” class members have interests which vary greatly from those of “in and out” class members, who no longer own CAMD stock. At least some in and out class members may be expected to be truly indifferent to the source of the settlement funds; since they have no continuing interest in CAMD, a dollar of settlement constitutes a net one dollar increase in their wealth. For equity class members, by contrast, the source of payment has a direct effect on the actual value of the settlement: one dollar of recovery from an outside source reflects a net increase in wealth of one dollar, while one dollar of recovery from CAMD reflects a significantly lower net recovery as the funds used to pay the latter settlement will deplete CAMD’s assets, dilute its stock and reduce the value of the class member’s current CAMD holdings. Indeed, at some point a settlement paid by CAMD will become a net loss for equity class members, even if they participate fully in the settlement, since a considerable portion of the settlement proceeds will be paid to class counsel, to say nothing of the fees paid to defense counsel. Since settlement payments made by CAMD are to equity class members little more than the shifting of wealth from their right pocket to their left, and since these class members were to be charged a twenty percent fee by class counsel for this “service,” there is ample reason to suspect that equity class mem-

bers have a keen interest in the source of the settlement funds. See generally In re Seagate Technology II Securities Litigation, 843 F.Supp. 1341, 1362-64 (N.D.Cal.1994).

The second group interested in where the settlement payments come from are major institutional investors, both those who remain equity shareholders and those who were in and out traders in CAMD. The defining characteristic of this group is that they are repeat players in the securities market and thus in securities fraud lawsuits. Repeated exposure to these lawsuits gives such institutional investors a perspective quite different from that held by an investor who owns little or no stock other than holdings in one issue and is unlikely to participate in the securities markets outside of such holdings. When an institutional investor’s transactions are viewed in the aggregate, the institutional or repeat investor can be seen as an equity shareholder in the markets as a whole, and accordingly can be expected to value a settlement for its effect on the larger market. Given this longer-term perspective, it seems likely that an institutional investor’s first priority will not be to maximize recovery in any given case, since the recovery maximized in one case may simply be lost in a settlement paid out by a company in which the institution owns shares in another case.⁶

A rational investor would be willing to incur the net loss which results from repeated exposure to securities lawsuits only if they produce a benefit. The only benefit generated by such lawsuits other than the actual money recoveries is deterrence of future fraudulent behavior. A rational institutional investor would thus be interested in pursuing securities class actions only when they generate a genuine deterrent effect. A settlement such as the one currently proposed, in which the company foots the initial bill and the actual wrongdoers are asked to pay later, if at all, fails to do so. Institutional members of the class would likely seek a much different settlement if given the opportunity to negotiate one themselves.

The objecting class members have said as much in their responses to the polling letter. COLPERA has criticized the settlement for its lack of deterrent effect, complaining that “[t]he proposed settlement sends the wrong message respecting director behavior because it (a) does not call for any payment by inside directors, whom the company apparently regards as culpable, and (b) does not appear to be based upon any careful review of the possible bases for claims against the outside directors.” COLPERA Application to Appear at 4. Similarly, James N. Roethe, Director of Litigation for Bank of America, criticized the proposed settlement for “improperly ignor[ing] any responsibility of individual defendants who appear to be most culpable, and thereby ignores the potential of obtaining settlement funds on behalf of such defendants.” Roethe Letter at l.⁷

LCH & B, in contrast, has no stake in the integrity of the securities markets and has no interest in ensuring that the negotiated settlement serves an appropriate deterrent function. Nor can the plaintiffs named in the complaints and currently put before the court as purported representatives be expected to have much interest in ensuring that this action has an appropriate deterrent effect. Those plaintiffs appear to be small traders with little or no abiding interest in the integrity of the securities markets as a whole.⁸ This latter fact is not surprising; as commentators have noted, plaintiff lawyers often put forth “figurehead” plaintiffs as representatives in securities class actions in order to allow them quickly to present the court with a settlement. See generally Weiss and Beekerman, 104 Yale LJ at 2060-61 (describing how plaintiff attorneys maintain a list of potential plaintiffs in order to be able to file securities class actions quickly after disclosures). Given the relatively small amount of trading in which the named plaintiffs apparently engage, and given that those plaintiffs have yet to make an appearance in this case but through their counsel, even when the court has clearly manifested its desire to hear their input and even when other class members such as COLPERA, Bank of America and Wells Fargo have made appearances independent of purported class counsel, the court concludes that the named plaintiffs have not and will not exert pressure on LCH & B to conduct this action in a manner which would maximize its deterrent effect.⁹

The court is not convinced by LCH & B’s response to the concerns elaborated above. This response is threefold. First, LCH & B claims that Professor Grundfest’s analysis submitted on behalf of Bank of America, Wells Fargo and other institutional investors is faulty because it ignores the fact that the shares to be issued in the proposed settlement were to be accompanied by a “contingent value right” which ensured that the share price would never fall below $8. Second, LCH & B claims that the institutional investors’ analysis is faulty because it fails to account for the likelihood that the value of CAMD securities will increase with time. Finally, LCH & B argues that the institutions’ analysis is irrelevant, because it applies only to class members who retain CAMD stock and such class members are a minority. LCH & B Reply Memorandum at 3, n 3.

LCH & B’s first two criticisms of the institutions’ analysis completely miss the point of that analysis, and the fact that the bulk of LCH & B’s rebuttal to the institutions’ analysis is so tone-deaf to the tenor of their argument only underscores the extent to which LCH & B was out of touch with the interest of class members when negotiating the proposed settlement. LCH & B’s first criticism misses the mark because it fails to acknowledge that the dilutive effect of the current settlement is troubling not because it means that the current dollar figure ascribed to that settlement is somehow illusory, but rather because it means that the value of those shares will not represent a net gain for retention class members. LCH & B’s second criticism is misguided for the same reason; no matter how much CAMD’s stock appreciates in the future, the fact remains that the value of the settlement shares will by necessity be subtracted from the value of the shares outstanding at the time of the settler ment. The settlement will not increase the net wealth which retention shareholders derive from their interest in CAMD; it will only cause those shareholders to derive that wealth from two stock certificates instead of one.

LCH & B’s final response to the institutions’ analysis, while responsive to the concerns it raises, is no more persuasive. It is by no means clear to the court that retention shareholders are the de minimis group of class members which LCH & B claims they are. Moreover, even if retention shareholders are a small group within the class, LCH & B has not explained why their interests should be sacrificed to those of the majority. LCH & B’s response also fails to account for the fact that institutional investors who are in and out traders will nonetheless be troubled by the dilutive effect of the proposed settlement for the reasons discussed previously.

In sum, LCH & B has failed to answer the criticisms of the proposed settlement submitted by the institutional investors. In fact, LCH & B’s attempt to do so has only further convinced the court that LCH & B did not negotiate the terms of the proposed settlement with the best interests of the class in mind.

The court concludes that the proposed settlement was reached primarily through LCH & B’s willingness to grant CAMD excessive concessions in order to present the court with a quick settlement, on the one hand, and through CAMD’s willingness to burden its shareholders with a settlement of claims for which its former officers are more properly liable, on the other. Because the putative class representatives allowed this settlement to be presented to the court for approval, and because the court has not heard from any of the putative class representatives during the lengthy proceedings in this case, the court must also conclude that LCH & B is operating with little or no monitoring from the putative class representatives and has, therefore, effectively made itself the class representative. Such a situation is expressly forbidden by the long line of cases, adverted to in section I, which hold that an attorney cannot simultaneously act as class representative and class counsel. See, for example, Shroder v. Suburban Coastal Corporation, 729 F.2d 1371, 1375 (11th Cir.1984); Lowenschuss v. Bluhdorn, 613 F.2d 18, 20 (2d Cir.1980); Zylstra v. Safeway Stores, Inc., 578 F.2d 102, 104 (5th Cir.1978); Susman v. Lincoln American Corporation, 561 F.2d 86 (7th Cir.1977); Turoff v. May Co., 531 F.2d 1357, 1360 (6th Cir.1976). Because the putative class representatives have allowed LCH & B to assume control of this action, the court is forced to conclude that the putative class representatives do not satisfy the requirement of FRCP 23(a)(4) that they be capable of “fairly and adequately” representing the class. See Social Services Union, Local 535 v. County of Santa Clara, 609 F.2d 944 (9th Cir.1979).

Having concluded that putative class representatives do not satisfy FRCP 23(a)(4), the court must now consider whether it is possible to identify a class member who does satisfy FRCP 23(a)(4). One of the unintended benefits of the polling procedure has been to inform the court’s decision regarding the type of plaintiff which is best suited to perform this role. This case demonstrates that litigation such as this must be monitored by an informed and independent plaintiff and simply cannot be left for the lawyers to manage. The court concludes that when putative class counsel are not monitored by an independent and informed client and when that counsel has taken significant action in the ease without court oversight or approval, the only adequate class representative under FRCP 23(a) is a class member who is well-informed about the action and independent of its counsel. In this case, COLPERA meets these requirements and has expressed a willingness to assume the responsibilities of class plaintiff.

IV

In this action, input from actual class members—as opposed to input from plaintiffs’ law firms which have hired figurehead clients and defendants’ counsel directed by corporate managers—has proven invaluable in assessing the true value of a settlement. It is no coincidence that this input has come from institutional investors; there are many reasons to suspect that such investors will be willing and able to monitoring attorney conduct in securities class actions much more rigorously than either figurehead plaintiffs or courts can do so. Institutional investors have financial interests in the outcome of securities class actions which dwarf the interests of individual plaintiffs, and with this increased financial interest comes an increased incentive to monitor class counsel’s conduct of the action. Institutional investors, moreover, are much better situated to conduct such monitoring, both because they have greater resources and because, as repeat players in the securities class actions, they are experienced in the issues which these actions inevitably raise.

Institutional investors are also in a superi- or position to either the court or the individual investors to determine whether or to what extent a company’s claims of imminent bankruptcy or other financial catastrophe should be given credence. Institutional investors are experienced both in general business matters and in securities liability exposure, and thus are well equipped to determine whether a company’s solvency is truly threatened by litigation or whether such a claim is merely a form of puffery employed by the defendant to induce the plaintiff class to trade time for money.

Finally, institutional investors such as COLPERA have fiduciary responsibility to the very investors whom the securities class action is designed primarily to help. The ability of the class action device to give redress to investors with losses too small to justify individual litigation has long been one of the main justifications for the use of class actions in Rule 10b-5 cases. See, for example, In re Seagate II, 843 F.Supp. at 1350; Herbert Newberg and Alba Conte, 4 New-berg on Class Actions § 22.10 (3d ed 1992). Institutional investors such as COLPERA, which is responsible for investing the money of state workers, represent precisely this type of interest.

Institutional investment vehicles, notably pension and mutual funds, have enabled millions of relatively small investors and savers to participate in the securities markets. Since an institutional investor, representing such interests, will likely use securities class actions to preserve the integrity of the securities markets rather than maximize damages claims and attorney fees, it makes sense for courts to look first to such parties in seeking to find an adequate representative plaintiff for a securities class action.

For all of the above reasons, the court concludes that the named class representatives have not adequately represented the purported class, which should instead be represented by COLPERA. COLPERA is a putative class member¹⁰ which has manifested its interest and willingness to represent the class by hiring its own counsel and seeking to intervene in this action. The court concludes that COLPERA is both willing and able to select class counsel and to monitor the performance of that counsel, thus ensuring that any future settlements are designed primarily to benefit the class members, not the attorneys. The court therefore CERTIFIES COLPERA as class representative to represent a class of claimant purchasers of CAMD securities. FRCP 23(a).

COLPERA may choose class counsel and proceed with the litigation as it sees fit. After conferring with all other parties, COLPERA should seek a status conference with the court when appropriate.

V

In sum, the court

1. DENIES the motions of LCH & B and CAMD for preliminary approval of the proposed settlement;

2. DENIES the request of LCH & B to be appointed class counsel; and

3. CERTIFIES the Colorado Public Employee Retirement Fund as class representative. FRCP 23(a).

IT IS SO ORDERED.

1 . The determination whether a class representative satisfies the requirements of FRCP 23(a)(4) is primarily one of fact, McGowan v. Faulkner Concrete Pipe Company, 659 F.2d 554, 559 (5th Cir.1981), and is committed to the sound discretion of the district court. Id..; Social Services Union, Local 535 v. County of Santa Clara, 609 F.2d 944, 947 (9th Cir.1979).

2 . Bank of America bought 100,000 CAMD shares during the class period, while Wells Fargo purchased an unspecified number. Other institutional investors represented by Professor Grundfest did not purchase CAMD shares, but regularly participate in the securities markets; they are: California Public Employees' Retirement System; Capital Group Companies, Inc. (mutual funds); College Retirement Equities Fund; General Motors Investment Management Corporation; Frank Russell Trust Company and Stanford Management Company. In aggregate, the entities represented by Professor Grundfest appear to manage well over $600 billion in investor accounts.

3 . The following potential class members have contacted the court with concerns over the proposed settlement: Bank of America, Wells Fargo, COLPERA and Mark Scott, President of the Volume Investor. In addition to the institutional investors who are not potential class members also represented by Professqr Grundfest, see n. 2, supra, the State of Wisconsin Investment Board, representing the pension accounts for employees of that state, also submitted a letter supporting the polling of class members regarding settlement and competitive selection of counsel, but criticizing LCH & B’s polling letter in this case. The court has also received inquiries and less substantive submissions from other purchasers of CAMD stock.

4 . The court has elsewhere commented on the troubling manner in which this bidding was carried out and will not repeat those comments here.

5 . Indeed, the fact that the polled investors are facing just such a “prisoner's dilemma” may explain why so many of them responded positively to the polling letter. Given the transaction costs of objecting to this settlement, and given the fact that the purportedly approving class members may well be asked to fund settlements in other cases in which the settlements are not subjected to the cuiTent scrutiny, many of the polled investors may have felt that the most rational course was quickly to approve this settlement and thus attempt at least to break even in the securities litigation in which they are involved.

6 . This analysis is supported by some of the submissions which the court has received from institutional investors in response to the polling letter. See, for example, October 18, 1995, letter to the court from James N. Roethe, Director of Litigation for Bank of America (“As both an issuer of stock and an investor, we are concerned about these types of settlements that redistribute wealth from one group of shareholders—those who presently own shares—to another group which purchased shares within the class period.”).

7 . Indeed, the former chief accounting officer of CAMD, Ronald Romito, recently settled a civil action brought by the SEC arising out of the very misstatements which are the basis of this litigation. See SEC v. Romito, C-95-20857-RMW. Yet the proposed settlement requires CAMD to pay for the effects of those misstatements before the class even attempts to win a recovery from CAMD’s former officers, including Mr. Romito.

8 . None of the named plaintiffs in this action purchased significant amounts of CAMD stock. The purported class representatives represented by LCH & B and named in C-94-2817 and C-94-3982, Michael Feder and Edward H. Mitch-um, Jr, purchased 300 and 5,000 shares, respectively, of CAMD stock during the proposed class period. The Michael Berman Trust, Michael Berman, trustee, named in C-94—2830, purchased 500 shares in the proposed class period. Lee Drasnin, named in C-94—3642, purchased 2,500 shares in the proposed class period. Mario Kravetz, named in C-94-3660, purchased 500 shares in the proposed class period. Jay L. Aid-rich, named in C-94—3662, purchased 500 shares in the proposed class period. Marjorie E. Sharmat, named in C-94-3696, purchased 600 shares in the proposed class period. Alan H. Ray, named in C-94—4065, purchased 125 shares in the proposed class period. Rita's Enterprises, Inc, and Rita's Place, Inc, named in C-94—4253, purchased 1,000 shares in the proposed class period. Joseph A. Delaney, named in C-94-4484, purchased 200 shares in the proposed class period. Finally, Victor Heresniak, Evelyn Harris and Kenneth E. Baranski, named in C-95-0560, purchased, 1,500, 400 and 500 shares, respectively, during the proposed class period.

9 . Indeed, there is reason to question whether the named plaintiffs have exerted any influence at all on the manner in which LCH & B or other counsel representing class members have prosecuted this action. To take a particularly vivid example, plaintiff Alan Ray has purportedly authorized Gold, Bennett & Cera to file an eleven page motion, accompanied by twenty eight exhibits, which seeks to “correct” that part of the court’s August 4, 1995, Order which was critical of Gold Bennett’s bid in this case. The granting of this motion could in no conceivable way benefit Mr. Ray, who has a small financial interest in this case and whose economic interest in class counsel's fees is directly opposite to that of Gold Bennett; rather, the motion would only remove what the Gold Bennett firm apparently regards as a smirch on its reputation as a securities class action firm. This motion, in other words, was filed by Gold Bennett in the name of Mr. Ray but was designed solely to benefit the Gold Bennett firm itself.

10 . COLPERA purchased 20,000 shares of CAMD stock on September 8, 1993, and sold all of these shares within the class period. As the court noted in section II A 1, supra, it has not yet been presented with evidence regarding the true value of CAMD stock during the class period; it is thus impossible to say with certainty at this time whether COLPERA, or any other in and out plaintiff, is a claims class member. If during further prosecution of this action it is discovered that COLPERA is not a claims class member, the court at that time will appoint additional class representatives who are claims class members.