Gartenberg v. Merrill Lynch Asset Management, Inc. ( 1983 )

573 F. Supp. 1293 (1983)

Irving L. GARTENBERG, Plaintiff,
v.
MERRILL LYNCH ASSET MANAGEMENT, INC., Merrill Lynch, Pierce, Fenner & Smith Incorporated, Merrill Lynch & Co., and Merrill Lynch Ready Assets Trust, Defendants.
Simone C. ANDRE, Plaintiff,
v.
MERRILL LYNCH READY ASSETS TRUST, Merrill Lynch Asset Management, Inc., Merrill Lynch Funds Distributor, Inc., Merrill Lynch, Pierce, Fenner & Smith Incorporated and Merrill Lynch & Co., Inc., Defendants.

Nos. 82 Civ. 8074(MP), 81 Civ. 7021(MP).

United States District Court, S.D. New York.

September 1, 1983.

Pomerantz Levy Haudek & Block by Stanley Grossman, Bruce G. Stumpf, Vincent K. Eng, New York City, for plaintiff Irving L. Gartenberg.

Silverman & Harnes by Sidney B. Silverman, New York City, for plaintiff Simone C. Andre.

Brown, Wood, Ivey, Mitchell & Petty by James K. Manning, A. Robert Pietrzak, New York City, for Merrill Lynch Ready Assets Trust.

Rogers & Wells by James N. Benedict, Stanley Godofsky, New York City, for all other defendants.

DECISION AND OPINION

MILTON POLLACK, District Judge.

The same parties sue, and are sued again, on the same type of claim as was tried at a Bench trial before this Court in 1981, which resulted in a judgment of dismissal on the merits on December 28, 1981 in favor of the defendants, 528 F. Supp. 1038, aff'd 694 F.2d 923 (2d Cir.1982), cert. denied, ___ U.S. ___, 103 S. Ct. 1877, 76 L. Ed. 2d 808 (1983) (petition filed by plaintiff Andre alone).

These are suits under the Investment Company Act of 1940 (the "Act") by Gartenberg and Andre as shareholders in a money market fund, the Merrill Lynch Ready Assets Trust (the "Fund," or "MLRAT") challenging as excessive, the management and advisory fees received from the Fund in 1982 by Merrill Lynch Asset Management, Inc. ("MLAM") in violation of its fiduciary duty prescribed by Section 36(b) of the Act, 15 U.S.C. § 80a-35(b). MLAM's affiliate, Merrill Lynch Pierce Fenner & Smith, Inc. (the "Broker" or "MLPFS"), which processed more than 7,300,000 orders for the Fund's shareholders in 1981, is also again named as a defendant, as is Merrill Lynch & Co., MLAM's parent corporation.

At the trial of Gartenberg I, Andre attempted to add to the suit, claims under Sections 15(a), 15(b), 15(c), and 20(a) of the Act, 15 U.S.C. §§ 80a-15(a) to 15(c), -20(a). Those claims were considered and found to be improperly asserted and at all events wanting in legal substance. 528 F. Supp. at 1067, aff'd 694 F.2d at 934.^[1]

In the present suit, Gartenberg also asserts a Section 20(a) claim that MLAM is liable for false and misleading proxy statements made to Fund shareholders.^[2]

Plaintiffs, having unsuccessfully attacked MLAM's compensation for 1980-81 as excessive and in violation of its fiduciary duty, filed "new" similar claims against the compensation for 1982 within three days after the affirmance of the judgment dismissing their claims for 1980-81.

The "new claims" assert that the Fund, in 1982, was substantially larger, that the 1982 compensation received by MLAM was excessive and receipt thereof violated its fiduciary duty and that two misrepresentations made by MLAM to the trustees were of "possible significance to the trustees in negotiating the fees," pertaining to fall-out benefits to be considered as constructive additions to the advisory fees and to the actual processing costs incurred by Merrill Lynch. These matters will be discussed infra.

Virtually all fundamentals of the present claims are foreclosed by the doctrines res judicata and/or collateral estoppel, and the affirmed findings on the earlier claims make it unnecessary to consider them de novo.

In the Spring of 1983, just before consideration of the annual renewal of the management contract for 1983-84, a study, ordered by the trustees of the Fund, was completed by the firm of Peat, Marwick, Mitchell & Co. (PMM), the outside independent accountants, which established and quantified a constructive addition to MLAM's fees by reason of so-called "float" interest benefits obtained by MLPFS, the broker affiliate, from proceeds of sales and redemptions of Fund shares.

The Second Circuit had dealt specifically with fall-out benefits from float and fall-out commission revenue in affirming the lower Court's finding that plaintiffs had not carried their burden of proof under § 36(b) of showing that MLAM's fee was so disproportionately large that it was beyond the range of what would have been agreed to after arm's length bargaining. The Appellate Court noted, however:

"These benefits to an affiliate in the Merrill Lynch organization, to the extent quantifiable, should be taken into account in determining whether the Manager's fee meets the standard of § 36(b).

"... It would not seem impossible, through use of today's sophisticated computer equipment and statistical techniques, to obtain estimates of such `fall-out' and `float' benefits which, while not precise, could be a factor of sufficient substance to give the Funds' trustees a sound basis for negotiating a lower Manager's fee.... Indeed, the independent trustees of the Fund might be well advised, in the interests of Fund investors to initiate such studies." 694 F.2d at 932-33.

As will be shown hereafter, the experts on both sides agreed that it was either not possible or practically not feasible to quantify fall-out commission benefits to MLPFS even with the use of today's sophisticated computer equipment and statistical techniques.

In his "new" complaint, Gartenberg also repeats his allegations that the trustees should, in addition to realizing the existence of float benefits, have been advised of the dollar amount of such benefits when they approved the management contract in May, 1982 (some seven months prior to the opinion of the Court of Appeals quoted above, and almost a year prior to the quantification of the float). Instead, the plaintiff charges, the trustees were misled by indications from Merrill Lynch, or MLAM, that such additional benefits to MLPFS were not significant overall.

A connected study was conducted by PMM, the results of which became available to the trustees of the Fund in May, 1983, and consisted of an in-depth reexamination and quantification of the costs of the services supplied to the Fund and its shareholders by the Merrill Lynch organization. The impetus for the study derived from the suggestion of the Court of Appeals in the earlier case. MLAM's expenses and costs so quantified were then compared with the sum of all of the quantifiable fall-out benefits added to the fees received by MLAM to determine whether there was any excessiveness in MLAM's compensation, actual and constructive, in 1982, in violation of Section 36(b).

The evidence in this case firmly established that the compensation received by MLAM was not in violation of Section 36(b) of the Act; that the independent trustees of the Fund were not dominated but rather were properly and adequately informed so that they were able to evaluate MLAM's contract and the benefits actually and constructively obtained therefrom by MLAM and its affiliates; and that the trustees disinterestedly and in good faith scrutinized that compensation and the realization of constructive side benefits to MLPFS and reached a reasonable business decision, satisfying an arm's-length standard of bargaining, that there was no excessiveness or unreasonableness or breach of fiduciary duty involved in such compensation received from and through the Fund. Hindsight results have amply vindicated their judgment. Plaintiff failed to carry his burden of credibly demonstrating that the trustees were imposed upon by any misrepresentation to them by the advisor or through any lack of appreciation of concepts or associated benefits of the overall enterprise, essential to an exercise of a practical informed reasonable business judgment. Similarly, plaintiff failed to establish with credible evidence that the proxy materials to the Fund's shareholders were false, misleading or inadequate.

In more detail, the facts follow:

Res Judicata and Collateral Estoppel Considerations

Defendants strongly maintain that doctrines of res judicata and/or collateral estoppel bar the present action. They urge that no new material facts, changed circumstances or conduct have developed since the previous complaints were dismissed on the merits. They say that plaintiffs charge merely a continuation of the allegedly violative conduct asserted in the earlier suits, and that since the judgment therein is a final determination of the same conduct under the same management contract and what might have been in issue had it been raised therein, the present suits legally state only the same claims and merely a continuation of conduct under the same contract previously adjudicated as not violative of Section 36(b). The so-called additions to the compensation through float and fall-out benefits to the Broker were matters raised but not proved by the plaintiffs in the earlier cases. The defendants contend that plaintiffs' claim that circumstances have changed is merely their attempt to submit proof on those matters unsuccessfully litigated in the earlier cases.

The plaintiffs, however, contend that they are not precluded from maintaining the present suits under either doctrine of law based on allegations that the amount in controversy here is larger; that defendants have instituted new procedures for reimbursement of the Broker's processing costs; that estimates of processing costs have been audited, by independent Certified Public Accountants, which furnish the type of probative evidence recommended by the Court of Appeals; that since 1981 the Manager's fees have increased while the number of orders processed by the Broker has declined; that the Manager's expenses have declined, while compensation increased by 25%; and that economies of scale have not been shared.

Further, plaintiffs assert that they here charge the defendants with new misrepresentations and nondisclosures to the trustees which occurred since January 1, 1982, concerning "net float" benefits and inability to quantify fall-out commission benefits. Plaintiffs also urge that it was not previously determined that the fee contract was fair and reasonable, and that, for all these reasons, the current action is not precluded by the prior determination.

The short answer here is that under Section 36(b), 15 U.S.C. § 80a-35(b), MLAM and the independent trustees have an ongoing obligation to see that MLAM's compensation does not become a breach of fiduciary duty by reason of excessiveness. Changes in circumstances that increase the advisory fee or the benefits associated therewith, or decrease the costs to MLAM and its affiliates associated with earning the fee could result in a finding that the advisory fee violates the fiduciary duty of the manager, even when a fee previously generated by the same advisory agreement was found not to breach that duty.

The doctrine of res judicata provides that a valid and final judgment rendered in favor of the defendant bars another action by the plaintiff on the same claim. See Montana v. United States, 440 U.S. 147, 99 S. Ct. 970, 59 L. Ed. 2d 210 (1979). The doctrine of collateral estoppel provides that a final judgment is conclusive in a subsequent action as to issues of fact or law that were actually litigated in the first trial and were essential to the determination in the first trial. See, e.g., Winters v. Lavine, 574 F.2d 46, 57 (2d Cir.1978). The purpose of these doctrines is to protect litigants from the burden of relitigating identical issues and to promote judicial economy by preventing needless litigation. See Parklane Hosiery Co. v. Shore, 439 U.S. 322, 326, 99 S. Ct. 645, 649, 58 L. Ed. 2d 552 (1979).

These doctrines of preclusion should be applied warily when Congress has acted to create an ongoing duty, such as the fiduciary duty to avoid excessive fees that is created by Section 36(b) of the Act. The fact that a Fund's advisor has received a favorable judicial determination that the advisory fee was not excessive does not insulate the advisor from future claims under changed circumstances that future fees are excessive and in violation of Section 36(b). An advisor would be entitled to the preclusive effects of a final judgment dismissing a claim that the advisory fee was excessive, if the costs and benefits to the advisor of the fee arrangement were no more favorable to the advisor in the subsequent suit than they were in the suit that was dismissed. Where a plaintiff makes a good faith claim that the fee has increased, that the operating costs have decreased, or that the benefits associated with the fee have increased, the doctrines of res judicata and collateral estoppel should not be used to bar an evidentiary determination of whether the fee is in violation of Section 36(b). In this case, application of the doctrines of preclusion would be particularly inappropriate, in light of the Second Circuit's suggestion that the trustees might initiate studies of float and fall-out benefits in preparation for a determination of whether the advisory fee was excessive.

Cases cited by defendant do not suggest a different result. The fact that a party who failed to prove an antitrust conspiracy is precluded from basing a subsequent conspiracy claim on a continuation of the conduct that was the basis for the first suit, see Engelhardt v. Bell & Howell Co., 327 F.2d 30 (8th Cir.1964), is inapposite. While private entities are continuously subject to the antitrust laws, these laws do not impose an ongoing fiduciary duty. Moreover, plaintiffs herein do not merely allege a continuation of the conduct that was the subject of the first suit; rather they claim that changes in circumstances in 1982 made the advisory fee excessive even if it had not been found excessive under the same contract in 1981. Thus cases where courts have barred subsequent claims on the ground that they were collaterally estopped because the identical issue had been litigated are also inapplicable here. See, e.g., Exhibitors Poster Exchange, Inc. v. National Screen Service Corp., 517 F.2d 110, 114 (5th Cir.1975), cert. denied, 423 U.S. 1054, 96 S. Ct. 784, 46 L. Ed. 2d 643 (1976). Similarly, Walsh v. International Longshoremen's Ass'n, AFL-CIO, 630 F.2d 864 (1st Cir.1980), which accorded preclusive effect to the denial of an injunction against a secondary boycott of cargoes bound for or arriving from the Soviet Union, is not dispositive here. In Walsh, the First Circuit was addressing conduct that was simultaneously engaged in by different ILA local unions and the Court held that the conduct sought to be enjoined was identical in the two cases.

The Section 36(b) cases where Courts in this District have barred claims attacking advisory fees under the doctrine of res judicata are also inapplicable because they involve lawsuits by plaintiffs who were attempting to undermine valid settlements of class action shareholder derivative suits. See Ashare v. Brill, 560 F. Supp. 18 (S.D.N. Y.1983), appeal pending No. 83-7276 (2d Cir. filed April 18, 1983); Lerner v. Reserve Management Co., [1981 Transfer Binder] Fed.Sec.L.Rep. (CCH) ¶ 98,036 (S.D.N.Y. 1981). In each of these cases, District Courts refused to overturn advisory fee schedules that had been established in the settlement of earlier derivative actions. Obviously much stronger considerations of fairness to defendants are present in a case where a shareholder derivative suit challenges fees that were created as a result of the settlement of a previous shareholder derivative suit. Plaintiffs herein do not seek to challenge as excessive a fee resulting from a contract that they have agreed to.

While the outcome in Gartenberg I does not bar the present case, the prior determination has a strong effect on the present action due to the doctrine of stare decisis. The findings of fact made by this Court in Gartenberg I, which were accepted by the Court of Appeals, are controlling in this case on all issues where a change in circumstances is not demonstrated by the parties. Similarly, the standard of law to be applied under Section 36(b) is well defined by the opinions in Gartenberg I and the present case does not require a duplicative analysis of the statute or its legislative history.

In analyzing Section 36(b), the Court of Appeals in Gartenberg I stated that "the test is essentially whether the fee schedule represents a charge within the range of what would have been negotiated at arm's-length in the light of all the surrounding circumstances" 694 F.2d at 928. The Court of Appeals further stated that "[t]o be guilty of a violation of § 36(b), therefore, the advisor-manager must charge a fee that is so disproportionately large that it bears no reasonable relationship to the services rendered and could ... not have been the product of arm's-length bargaining. To make this determination all pertinent facts must be weighed." 694 F.2d at 928.

Recognizing the precedential impact of the prior litigation, the parties have stipulated that the record in Gartenberg I, including the Court's opinions therein, is made part of the record herein. Issues proven in the earlier action require no additional proof. Accordingly, familiarity with the Circuit and District Courts' opinions in Gartenberg I is assumed.

The Money Market Industry and MLRAT

At the time of Gartenberg I, in December, 1981, money market funds and the Merrill Lynch Ready Assets Trust in particular were experiencing continued and substantial growth in both shareholders and assets and commensurate increases in the costs and processing of shareholder orders. Total money market assets had reached $185 billion, which represented a twentyfive fold increase in less than four years. See Gartenberg I, 528 F. Supp. 1038, 1042. The MLRAT had grown from its inception in 1975 to become the largest money market fund with almost $20 billion in assets at the time of trial in September, 1981, and over $22.6 billion in assets in December, 1981, when the decision in Gartenberg I was rendered.

The growth of the MLRAT and the industry as a whole continued in the early months of 1982. The number of MLRAT shareholders peaked in March, 1982, at 1,458,128 and MLRAT net assets peaked at $23,235,350,821, in August, 1982. Through the end of November, 1982, the size of the Fund remained relatively constant, with assets in excess of 22 billion dollars. The growth of the Fund paralleled that of the industry as a whole, which peaked in assets at $232.6 billion on December 1, 1982.^[3] The leadership of Merrill Lynch in the money market field was due in large part to the availability of the Merrill Lynch system and its vast organization which could supply the processing services required to satisfy the daily orders and other demands of the shareholders. In November, 1982, Merrill Lynch was processing over 25,000 orders per day on behalf of the MLRAT.

Another reason for the relative success of the MLRAT in attracting shareholders and assets was the strong performance record of the fund. Of the fourteen domestic prime money market funds^[4] in existence for the five-year period beginning in 1978 and ending in 1982, the MLRAT had the fourth best overall performance, with an average rate of return of 11.98% over the five-year period, as compared to the industry average of 11.77%. In 1982, the MLRAT had the seventh best performance out of 51 domestic prime funds, with an annual rate of return of 12.67% as compared to the industry average of 12.30%.^[5]

The spectacular rise in the success of money market funds came to a dramatic end in December, 1982. As the result of Federal action deregulating rates which could be offered by banks and other financial institutions, banking institutions, on December 10, 1982, began to offer "money market deposit accounts" which in some cases afforded rates of return comparable to those offered by the traditional money market funds. These new bank funds were aggressively marketed and this, combined with the rise in stock prices and the dramatic reduction in interest rates, led to a precipitous liquidation and fall in the level of money market assets. In six months the MLRAT lost almost 36% of its assets and 15% of its shareholders. The decline experienced by MLRAT is as follows:

Net Assets        No.             No. Accounts
(millions)        Shareholders    Closed
November 1982         $22,499          1,410,390           37,359
December 1982          19,332          1,338,440          103,413
January  1983          17,766          1,297,060           62,350
February 1983          16,629          1,263,458           50,960

March         1983         15,772          1,235,396       50,927
April         1983         14,766          1,208,736       47,237
May           1983         14,527          1,192,740         - -
June[6]    1983         14,141             - -            - -

Even though there have been very substantial redemptions in the last six months, Merrill Lynch has successfully met the redemption demands of investors; there has been no freeze on any payouts and no processing delays or difficulties have been reported. Donald Cecil, an independent trustee of MLRAT and chairman of the audit committee of the trustees, attributed the lack of difficulty in meeting shareholder demand for redemption to the efficiency of the organization and the expertise of the people handling it. Thus, the Merrill Lynch organization has successfully managed the decline in the fortunes of the MLRAT with the same professional competency it employed during the meteoric rise of the Fund.

It is against this historical background that plaintiff Gartenberg's contentions^[7] that misrepresentations were made to the trustees when they were considering the annual renewal of the management contract relating to processing costs, float and other fall-out benefits and the conclusory. contention that the compensation received in 1982 by MLAM is excessive and received in breach of fiduciary duty, must be considered.

The Approval of the Advisory Agreement by the Trustees (1982)

In Gartenberg I, the Court considered all of the advisory agreements up to and including the 1981 advisory agreement, effective September 1, 1981 through August 31, 1982, as approved by the trustees on May 7, 1981 and by the shareholders on August 6, 1981. The Court found that the trustees engaged in frank and open discussions, that they were not dominated by MLAM and that they exercised an informed judgment in good faith and upon a reasonable basis. 528 F. Supp. at 1064.

In the spring of 1982, the trustees were called on to consider the same proposed advisory agreement that would become effective in September, 1982, if approved by the independent trustees and the shareholders. Thus, they had the benefit of this Court's decision in Gartenberg I, but they conducted their deliberations without the guidance offered by the Second Circuit's opinion in Gartenberg I, since it was not filed until December 3, 1982.

In November, 1981, the board of trustees of the MLRAT was increased in number to ten trustees, eight of whom are non-interested within the meaning of the Act. In Gartenberg I, the Court noted that the six non-interested trustees were "men of maturity and substantial stature with singular qualifications and experience in divers fields, having held positions of importance in the world of business, finance, investment and education." 528 F. Supp. at 1058. These six men have been joined by two additional non-interested trustees who are equally qualified and experienced. The independence and competence of the trustees was not questioned or questionable herein and there is no question that they are completely free of domination or undue influence by MLAM, MLPFS and their affiliates.

In advance of the May 12, 1982 trustees' meeting at which the advisory agreement was approved, the trustees were presented with a 347-page document, known as the 1982 Trustee Booklet, or the 1982 "Green Book." As in previous years, the attention of the trustees was appropriately fixed on their responsibilities at the beginning of the text of the "Green Book":

As the independent Trustees, you bear the principal responsibility for evaluating the advisory and distribution agreements, determining their reasonableness and considering the available alternatives from the standpoint of the best interests of the Trust and its shareholders....

The trustees were accurately apprised in 1982 of the requirements of Sections 15 and 36(b) of the Act in substantially the same manner as had been used in Gartenberg I. In particular, the trustees were advised that the claims asserted by Gartenberg and Andre in Gartenberg I had been dismissed on the merits and that both plaintiffs had filed appeals. A complete copy of this Court's decision in Gartenberg I was presented for review by the trustees. The trustees were advised of the grounds of the plaintiffs' appeals including the claims that (i) it was improper to consider the processing costs of MLPFS in weighing the fairness of the compensation to the Manager; and (ii) that this Court's decision should be reversed because the trustees were not informed of the dollar value of and failed to consider fall-out benefits to MLPFS, including float and fall-out commission benefits. In addition, the Green Book contained the advisory and distribution agreement proposed for approval, comparisons of the proposed MLAM fee with those of other money market funds, information on the processing costs incurred by MLPFS on behalf of the MLRAT based on the cost studies presented in evidence at Gartenberg I, financial statements and other information relating to MLAM as applied to MLRAT and non-MLRAT activities, information regarding MLRAT portfolio transactions and a legal opinion concerning the propriety of considering processing and administrative expenses of MLPFS in determining the reasonableness of the advisory fee.

Donald Cecil, a disinterested Trustee, testified that he believes that he has received from MLAM and those acting on its behalf sufficient information from the time he first became a Board member until the present to make a reasonable business judgment as to the fairness of the investment advisory fee. He has been continuously counselled, as have the other independent Trustees, regarding the fee during the period January 1, 1982 up to May 18, 1983. He believes that he and his fellow trustees have acted very conscientiously, spent a great deal of time in both study and deliberation, and have made good business judgments to the best of their ability, and thinks that the shareholders are getting a very good product at a very economical price.

In summation, at the close of the trial, plaintiff limited his challenge to the information considered by the trustees to a claim that only two misrepresentations were contained in the Green Book or in statements made to the trustees.^[8]

Plaintiff challenged the statement made in the memorandum to the trustees contained in the Green Book as it relates to float, which states as follows:

The plaintiffs have particularly emphasized benefits to Merrill Lynch derived from the "float" with respect to redemption checks. You are aware of the means by which Merrill Lynch customers transact share orders through Merrill Lynch. In the case of purchase orders Merrill Lynch advances immediately available funds to the Trust on behalf of its customers the day after funds are received from the customer even though Merrill Lynch may have received "Clearing House" funds from such customer which may take a considerable time to clear. Because of this practice, and the fact that generally purchases have exceeded redemptions, Merrill Lynch believes that there is no significant net float involved.

Plaintiff claims that the subsequent studies have shown that there is significant net float and thus that the statement contained in the April 21, 1982 Green Book for the May 11-12 meetings of the trustees was false. The truth of the statement in the Green Book can only be judged on the basis of the information available at the time it was made. At that time, no study of float accruing to the benefit of MLPFS had been concluded^[9], nor was there any indication that a legal obligation existed to conduct such a study. The trustees were made aware that the plaintiffs thought that float was deserving of particular emphasis and that they were basing their appeal, in part, on the argument that float should be included in the calculation of benefits accruing to the advisor.

Plaintiff's attack on the statement concerning net float in the 1982 Green Book has the benefit of hindsight with respect to the Second Circuit's later opinion in Gartenberg I which suggested that the trustees might well be advised to initiate studies of float and fall-out commission benefits.

In fact, the trustees did initiate an elaborate and expensive study designed to attempt to quantify float and fall-out benefits almost immediately after the Second Circuit's decision. As will be discussed below, the trustees and their expert concluded that only the former could be quantified and float was quantified for the period of calendar year 1982. There is no evidence of the degree of net float prior to April 21, 1982, the date the Green Book was presented to the trustees. What occurred thereafter was necessarily speculative at that time. It is clear, beyond peradventure of doubt, that no misrepresentation on the subject of float was made to the trustees in advance of their consideration of the renewal of the advisory agreement in 1982. Moreover, as indicated hereafter, the float later determined would not change the validity of the trustees' conclusion that the fee was not excessive since the amount of float is more than offset by the re-examined processing costs.

The second claimed misrepresentation urged on plaintiff's summation, concerns the certified financial statements of MLAM prepared by Deloitte Haskins & Sells (DHS). The MLAM Earnings Statement for the year ended December 25, 1981, contains an expense item entitled "Transaction processing costs reimbursed to affiliated company" in the amount of $29,000,000. At the May 11-12 meeting of the audit committee of the trustees, MLRAT and MLAM President Arthur Zeikel explained that

"the amount of this item had been determined unilaterally and arbitrarily by Merrill Lynch solely for internal accounting convenience and that it had been based upon information derived from the Merrill Lynch CASS system.... [MLAM personnel] did not have an opportunity to discuss the subject with Merrill Lynch and that [I have] undertaken to assure that future MLAM financial statements would not reflect determinations concerning items of this magnitude and sensitivity that had not been discussed with appropriate MLAM personnel prior to the time the determinations were made."

The minutes show that the matter was discussed extensively, and the consensus of the meeting of the audit committee was that greater care should be taken with accounting allocations of this type in the future.

Plaintiff points to the DHS work papers that purport to indicate that MLAM's top officials agreed to the $29,000,000 figure which they considered reasonable in the circumstances. Plaintiff claims that Mr. Zeikel's statements misrepresented the fact that MLAM officials had not been consulted, when in fact they had, and caused the trustees not to rely on the DHS figure when they computed the processing costs that could legitimately be said to offset the advisory fee.

An examination of the DHS figure and the purpose for which it was constructed demonstrates that the trustees were correct in viewing the DHS allocation as not representative of the full costs of processing incurred by MLPFS. The $29,000,000 figure represents 80% of an amount generated by the CASS system as representing costs related to the Fund. The 80% figure was an artificial estimate chosen to reflect an intra-company purpose of the affiliated companies. The CASS system yielded an artificial figure which did not include an allocation for account executive compensation that is chargeable to the Fund nor does it reflect the true magnitude of other processing costs borne by MLPFS on behalf of the Fund. The CASS system was developed for a different purpose entirely, and it is based on production credits, a concept that is similar to commissions, and since the Fund does not generate production credits, certain costs that are fairly chargeable to the Fund are not captured by the CASS system. The $29,000,000 figure is nothing more than the internal estimate of the costs to MLPFS arising out of Fund operations that was used for intra-company transfer of funds and for certain reporting purposes.^[10]

Thomas Lockburner, DHS's national accounting and auditing coordinator for the securities industry, with particular responsibility for the audits of MLAM and MLPFS, made it perfectly clear that he did not view the internal figure agreed upon to represent the full value of processing costs supplied to MLAM, but only the intra-company amount agreed to be deducted on MLAM's reports; he testified that even if he had received the PMM cost study which included the total of the processing costs to MLPFS associated with the Fund, far in excess of those included in the DHS certified financial statements for MLAM, he would nonetheless not have altered his certification. He testified that when certifying a transaction between related parties, an auditor looks merely to see that the transaction took place and that the funds had been transferred (and not at the completeness of the item) DHS certified the amount that was transferred from MLAM to MLPFS for processing services in each of the two years; DHS did not certify that this amount represented the totality of the sum that is fairly chargeable by MLPFS to MLAM for processing costs associated with MLRAT.^[11] For purposes of § 36(b) it is the extent of processing costs that is important, not the degree to which part thereof is transferred to an affiliate's accounts from the Manager.

The 1981 and 1982 DHS estimates of processing costs for internal purposes did not represent the first time such artificial figures were constructed with respect to the Fund, nor did the 1982 trustees' meeting mark the first time that such figures or their purport were considered. As early as April, 1979, Robert Diemer, an official with Merrill Lynch's Diversified Financial Services Group, tried to calculate the costs to Merrill Lynch's branch office arising out of the Fund. At that time, which was before the Gartenberg I complaint was filed, Diemer attempted to calculate an allocation of branch office expenses based on MLAM revenue as a percentage of total Merrill Lynch production credits, a technique very similar to that employed by DHS. While Diemer thought this was the most appropriate technique of those available to him, he admitted that it would be "subjective at best." 528 F. Supp. at 1050 & n. 15.

The trustees were kept fully informed of the distinction between an internal allocation of costs and the total processing costs for use as an offset against the advisory fee. At the October 24, 1979 trustees' meeting, Mr. Ross, one of the interested trustees, stated that he had been given the authority to allocate costs between MLAM and MLPFS. Ross told the other trustees that in his opinion "the fee schedule should be based on the total cost to the Merrill Lynch complex for providing the services and not the internal allocation of costs between MLAM and MLPFS." 528 F. Supp. at 1061.

Against this background, it is unrealistic to assume that the trustees did not appreciate that the DHS $29,000,000 figure was only an intra-company estimate of the total processing costs incurred by MLPFS on behalf of the Fund. Nothing in the record supports plaintiff's assertion that the trustees believed that MLAM had not been consulted and therefore ignored the $29,000,000 figure entirely. The trustees had long been aware that artificial figures, created internally and relied upon for intra-company transfers, did not reflect the true magnitude of processing costs incurred by MLPFS on behalf of the Fund. As the deposition testimony of George James, one of the non-interested trustees, indicates: the inapplicability of the CASS based figures due to their reliance on production credits was again discussed at the May, 1982 Meeting of the audit committee. The only proper inference to be drawn on the basis of all the facts and circumstances is that the DHS figures were understood in the context indicated above and that they did not fully capture total costs. This was not error.

In determining whether there has been a breach of fiduciary duty in violation of Section 36(b), the Court must consider as important factors: the expertise of the independent trustees of the Fund, whether they were fully informed about all the facts bearing on the advisor-manager's services and fee, and the extent and care and conscientiousness with which the trustees performed their duties. Gartenberg I, 694 F.2d at 930.^[12]

Against the overall scope of the materials presented and representations made to the trustees, the two claimed misrepresentations do not create any doubt that the trustees were kept fully informed and were reasonably and practically able to assess the facts bearing on the advisor-manager's fees and costs.

The minutes of the deliberations of the audit committee and the full board of trustees further support a finding that the trustees were fully informed and that they understood and conscientiously considered the issues surrounding the advisory agreement and the compensation received thereunder. The trustees considered alternatives to the then existing relationship between MLPFS, MLAM and the MLRAT, including internalization of the functions performed by MLAM and the selection of another advisor. Both the audit committee and the full board of trustees voted unanimously to continue the advisory and distribution arrangements between MLRAT and MLAM for the year beginning September 1, 1982 on the same terms as existed in the preceding year. This decision was based on, among other things, the nature and extent of services provided to shareholders, operating expense information and fees charged by other investment advisors, a determination that collateral benefits to MLPFS from the MLRAT might exist but that they could not be quantified mathematically, an understanding that "float" on MLRAT redemptions might benefit MLPFS, but that it had not yet been quantified, and an appreciation of the vastness in general of the processing costs necessarily incurred by MLPFS on behalf of MLAM. The decisions were reasonably considered at the time and under the circumstances in which they were reached.

The Shareholders Consideration of the Fee

Plaintiff raises a vague claim in his complaint and proposed findings of fact, a claim which he chose not to advance through testimonial evidence, that the proxy statement to the Fund's shareholders, dated June 18, 1982, failed to reveal material information to the shareholders. The claimed omissions consist of information concerning the value and nature of benefits received by MLPFS from the Fund (presumably plaintiff is referring to fall-out commissions and float), the fact that expenses were declining as a percentage of revenues, and information concerning MLAM's expenses.

This argument is virtually identical to that raised in Gartenberg I, which was rejected in its entirety, 528 F. Supp. 1064-65. In his trial brief in Gartenberg I, plaintiff argued that the proxy materials were incomplete, in part due to their failure to reveal MLAM's costs in managing the Fund and their failure to reveal the "revenues and profits to Merrill Lynch & Co. resulting from the existence of the [Fund]." Here plaintiff offers no new evidence that the proxy statement failed to reveal material information, instead plaintiff points to the statement itself and suggests that the Court find the omissions listed above. For the reasons set forth in Gartenberg I, these allegations do not state or advance a claim under Sections 36(b) or 20(a).

The only aspect of plaintiff's claim concerning omissions in the proxy statement that is arguably new herein is the claimed omission of information concerning the dollar value of benefits received by MLPFS from the Fund. As of June 18, 1982, no reliable information concerning the dollar value of float or fall-out benefits was available for inclusion in the proxy statement. Subsequent events have demonstrated that fall-out commission benefits are not quantifiable and that the careful quantification of net float by PMM did not render the advisory fee excessive. For the purposes of Sections 20(a) and 36(b) of the Act, a proxy statement is not defective for failure to include speculative information. Obviously it cannot be criticized for a failure to discuss numerical values that had not yet been quantified. Plaintiff's claims of omissions in the proxy statement are without merit in all respects.

Costs and Benefits to MLAM and its Affiliates from the Operation of the Fund

The total costs attributable to the operation of the Fund incurred by the defendants in this case can be conveniently separated into the direct costs to MLAM and the costs of processing services which were provided by MLPFS.

a. MLAM Costs

In 1982, MLAM incurred expenses amounting to $2,050,000 directly on behalf of the Fund. Plaintiff argues that $222,000 allocated to "Advertising and Market Development" and $645,000 allocated to "Other" in a document entitled "MLAM Profitability Analysis," which was presented to the trustees, is not properly chargeable as an offset to the advisory fee. Plaintiff claims that the former are promotional expenses which violate SEC Rule 12b-1, 17 C.F.R. § 270.12b-1 and that the latter includes the expenses of the prior Gartenberg litigation, since the category "other" increased during the period of litigation.

Outside of attempting to draw inferences from the titles and size of the categories of expenses, plaintiff offers no proof of the amount, if any, actually spent on litigation. As far as the litigation expenses are concerned, plaintiff suggests no reason why these costs should not be included in a determination of whether the advisory fee is excessive in violation of § 36(b). Litigation expenses are no less real than any other costs paid by MLAM. Similarly, there is no evidence in the record to support a claim that MLAM acted as a distributor of securities in violation of Rule 12b-1. Merely including expenses in the category "Advertising and Market Development" neither proves that they were promotional, nor that the "promotions" were of a type that would violate Rule 12b-1. The plaintiff has not met his burden of proof with respect to a challenge of any portion of the direct expenses incurred by MLAM.

b. Costs of Providing Processing Services

As was the case in Gartenberg I, the vast majority of the costs associated with the operation of MLRAT are borne by MLPFS in the form of processing services that MLPFS supplies to the MLRAT. Both this Court and the Court of Appeals held that a determination of whether an advisor's fee is so disproportionately large that it bears no reasonable relationship to the services rendered must take into account processing costs borne by affiliates of the advisor.

1. Peat Marwick Mitchell & Co. Cost Study

For the benefit of the trustees and in anticipation of Gartenberg I, Peat Marwick Mitchell & Co. ("PMM") conducted an independent evaluation of how costs should be allocated within MLPFS. PMM surveyed MLPFS branches across the country and determined that, exclusive of account executive costs, it cost MLPFS between $6.50 and $9.75 per order to process MLRAT transactions between the third quarter of 1979 and the first quarter of 1981. In light of the December 3, 1982 Court of Appeals decision in Gartenberg I which was immediately followed by the filing and prosecution of the present action, PMM was again retained by defendants to determine and quantify the total costs of processing services as well as to measure the float and fall-out benefits that are discussed below.^[13]

Russell Peppet, the vice president of PMM in charge of management consulting, testified that since the earlier estimates were made a number of operational changes had been effected by MLPFS which required that a full cost study be conducted.^[14] Also, at the time of the 1979 through 1981 cost studies, Mr. Peppet had indicated that the value of a portion of the time Merrill Lynch account executives spent working on the Fund was assignable as a cost of the Fund. At that point in time, however, PMM did not assign a dollar amount, but believed that some amount was assignable. In light of the fact that Peppet sought to obtain an estimate of the economic benefits, including float, inuring to MLPFS from the Fund, there was no question in his mind that from an overall perspective, account executive compensation absolutely had to be included in the costs that were reasonably assignable to the Fund. Peppet concluded that the float and fall-out benefits received by MLPFS would have been impossible if the account executives had not used the Fund as a source for generating additional transactions. The account executives are involved in the very transactions that generate float, free credit balances, and fall-out for MLPFS.

In order to determine the percentage of time that MLPFS account executives and sales assistants spend on activities associated with the MLRAT, PMM surveyed twenty of MLPFS's 430 offices, chosen to create a cross-section on the basis of geography, office size, size of MLRAT accounts and office type (main or branch). PMM determined that Merrill Lynch account executives spent 7.4% of their total working time directly involved with MLRAT in 1982.^[15] These figures do not include any administrative activities that might be related to the MLRAT, and they represent percentages of total time at work. Arguably, the proper percentage would be based on the fraction of productive time, a calculation that would yield higher percentages deemed devoted to MLRAT.

As a result of its study, PMM produced an estimate of the total of all elements of processing costs including the cost of supplied account executive services not theretofore quantified. Omitting account executive service costs, the PMM estimate of the remainder of the 1982 cost of providing processing services for a MLRAT order volume of 7,053,363 orders was $55,509,968, the equivalent of $7.87 per order. This estimate is somewhat below the highest transaction cost estimates produced by PMM in advance of Gartenberg I. (The reduction may reflect some of the cost savings measures enacted by MLPFS).

The inclusion of a charge for that fraction of the account executives' time devoted to MLRAT activities, valued at that share of the account executives earnings, yields a total 1982 cost estimate of $104,530,841, the equivalent of $14.82 per order. Of this, approximately $45 million is account executive compensation and $4 million consists of overhead to that compensation.

Plaintiff does not challenge or offer an alternative to the methodology employed by PMM with respect to the cost calculations, including the computation of the percentage of time devoted to MLRAT activities by account executives. Indeed, plaintiff's counsel expressed a very high regard for Mr. Peppet's credibility and professional competency. The Court is convinced that a substantial portion of account executives' time is devoted to MLRAT processing and other MLRAT related activities. The PMM finding that MLRAT related activities account for 7.4% of an account executive's time is not unreasonable, especially since in 1982 orders for the purchase and redemption of MLRAT shares aggregate in excess of 30% of all orders processed by MLPFS's 430 offices and 8,700 account executives.

Plaintiff challenges the allocation of any of the compensation paid to account executives as attributable to MLRAT processing costs, arguing that because account executives are compensated primarily on the basis of a percentage of commissions that they generate, the time they devote to MLRAT which generates no commissions is of no cost to MLPFS. The fact that MLPFS account executives are not remunerated on a straight time basis does not mean that the time they devote to MLRAT activities has no value to MLPFS or is uncompensated in the payments received by the account executives. MLPFS has chosen to compensate its account executives on an incentive basis that rewards them for generating commissions for the company. When they engage as part of their duties in non-commission generating activities such as processing MLRAT orders, account executives are also benefiting the company. Moreover, every hour spent on MLRAT is an hour not spent prospecting for additional commissions. Thus the opportunity cost of an account executive's time that is spent on MLRAT activities is fairly chargeable as a processing cost and an offset to MLAM's advisory fees. While 7.4% of the total earnings for the account executives is not an exact measure of the value of the opportunity cost to MLPFS, it is not an unreasonable estimate, and plaintiff, who has the burden of proof, has offered no alternative other than ignoring account executive compensation completely. For the reasons already discussed, some allocation of account executive compensation must be made as a proper cost to offset the fees. Under the facts and circumstances of this case, and based on the Court's assessment of the credibility of the expert testimony, the Court finds that the estimate of account executive compensation allocable as an offset to the advisor's compensation from the Fund, as determined by PMM, is reasonable and allowable.

In related argument, plaintiff attacks the PMM study for employing an absorption as opposed to a direct technique of cost accounting which did not distinguish between variable costs and fixed costs. PMM employed a modified full-cost approach, specifically designed for this case, which admittedly did not distinguish variable from fixed costs. Plaintiff argues that only variable costs, here, those costs that increased directly because of the MLRAT should be allocated to the MLRAT. Mr. Peppet testified that there was insufficient information available for a division of costs between variable and fixed, and that the approach he employed was absolutely appropriate. Moreover, he stated that for some purposes it is proper to allocate fixed costs even where a direct technique of cost accounting is employed.

For the purpose of determining whether the fee charged by MLAM to the MLRAT is excessive, there is no question that a portion of the fixed costs incurred by MLPFS are allocable to the MLRAT. It is true that, when considering whether to add a new product line, a company is interested in the marginal profit of the new line, that is the extra income generated minus the incremental or variable costs allocable to the product. However, a company must generate sufficient revenues to cover both its fixed and variable costs — the fixed costs must be paid by contributions from all the product lines.^[16] An advisory fee is not excessive merely because it generates some income in excess of the variable costs that can help offset the fixed costs. Under plaintiff's reasoning, if all of MLPFS's operations were subject to the fiduciary requirements of Section 36(b), each product would only be allowed to include variable costs as an offset to its revenues and fees with the result that the entire company would be obligated to perpetually lose vast sums of money to avoid breaching its fiduciary obligations. There is no question that defendants are entitled to calculate processing costs associated with the MLRAT using techniques that apportion a fair share of both variable and fixed costs to the MLRAT.

Again, plaintiff, who bears the burden of proof, offers no allocation of costs as an alternative to the PMM study. Under the circumstances, including the fact that the MLRAT is not merely an insignificant appendage of MLPFS, but rather one that accounted for over 30% of all transactions in 1982, the PMM allocation of processing costs is not unreasonable.

c. The Fee

Pursuant to the investment advisory agreement with the MLRAT, MLAM receives an advisory fee from the MLRAT which is calculated as a percentage of the net asset value of the Trust. The same fee schedule is set forth in both the 1981 and 1982 advisory agreements which span the periods from September 1, 1981 to August 31, 1982 and September 1, 1982 to August 31, 1983, respectively.

The total investment advisory fee received by MLAM from the Fund from September 1, 1981 through August 31, 1982 was $63,320,825, the equivalent of .2847% of average total Fund assets for that period. The total fee received from September 1, 1982 through May 31, 1983, was $41,083,777, corresponding to approximately the same average annual percentage of Fund assets as in the earlier period. Plaintiff's Section 36(b) claim challenges the advisory fee for the calendar period 1982 and the expert quantification of the processing costs and float covers the calendar year 1982. The advisory fee earned by MLAM in calendar year 1982 is $64,367,393, the equivalent of .2851% of average total Fund assets.

d. Float and Free Credit Balances

In its generic sense, "float" denotes a timing difference that accrues to the benefit or disadvantage of an organization in that it may receive funds and disburse funds at different points in time, and consequently, either it receives an advantage from having the availability of those funds or suffer a disadvantage from the unavailability of the funds. The value of the "float" is determined by the interest rate that could be earned on the available funds (or the interest rate that is payable on funds that are borrowed to offset a "negative" float).

In this case the term float has been used to describe the advantage or disadvantage arising out of the purchase or redemption of MLRAT shares. For example, if a shareholder decides to ask his or her broker to redeem a portion of the MLRAT account and mail a check therefor, the broker will process the order and cause a sum to be deducted from the MLRAT account on the following day, known as the settlement day. On the settlement day, a check drawn on a Merrill Lynch account will be issued to the shareholder. MLPFS benefits from the float until the shareholder presents the check for payment. The converse, a negative float, occurs when an investor purchases MLRAT shares and the investor's account is credited with the shares and begins to accrue interest before the investor's check clears.

In light of the suggestion by the Second Circuit to the trustees that they obtain a numerical estimate of "float benefits" as they could be a "factor of sufficient substance to give the [trustee] a sound basis for negotiating a lower [advisory fee]," 694 F.2d at 932, the trustees commissioned PMM to quantify all aspects of float as it relates to the MLRAT. PMM thereupon calculated the float on redemptions, for each type of redemption^[17], and the float on purchases, for each type of purchase. PMM separately determined the amount of redemptions and purchases in each category and the average period of float, if any, for each category. PMM determined that the average daily funds flow that generated float to MLPFS's advantage was $234.7 million and the average daily funds flow to MLPFS's disadvantage was $12.4 million. The net float to the advantage of MLPFS is therefore $222.3 million per day. PMM applied the average 30-day commercial paper rate for 1982 (11.9%) to the net float and determined that MLPFS earned a net float benefit of $26.5 million in 1982.

The plaintiff's expert calculated the gross level of float benefit accruing to MLPFS on check redemptions alone to be 14 million dollars. This figure excludes other forms of redemptions and ignores negative float. PMM estimates that plaintiff's calculations on this segment of the float are consistent with its overall calculations. In light of the fact that defendant's evidence is more favorable to plaintiff than is plaintiff's own evidence, plaintiff accepts defendant's proof on this issue.

It is one thing to calculate net float benefits and quite another to decide whether they constitute a valid offset to the costs incurred by MLPFS in operating the MLRAT. All of the float benefits received by MLPFS are voluntarily contributed by the shareholders since each shareholder can completely eliminate all float generated by his/her transactions simply by exercising the option to withdraw funds by using the personal checks that are available to all shareholders, or by the other means available to them on request for "expedited redemption."^[18] When a shareholder draws a personal check on his/her MLRAT account, the funds remain in the shareholder's account earning interest for the shareholder until the check clears. Thus, as both parties agree, all float that benefits MLPFS can be eliminated through the use of personal checks. Float only arises because the redeeming shareholder wants the Merrill Lynch name on a check, instead of his/her personal name as the drawer of the check. Consequently, it is at the shareholder's absolute option whether float is or is not created.

Thus, unlike the advisory fee which is automatically deducted from the return paid on MLRAT accounts, the "float benefits" are not fairly chargeable as an offset to the costs incurred by MLPFS. If a shareholder, for whatever reason, thinks that it is better to have a check with the standing of the broker as the drawer than it is to use the check which he/she draws, then MLPFS is entitled to the float benefits thus created as compensation for the value of the optional check service it is called upon to provide the customer. In principle, voluntary float should not be counted as an offset to the processing costs incurred by MLPFS on behalf of the MLRAT.

Free credit balances are similar to float. They arise when a shareholder tells the broker to redeem MLRAT shares but does not tell the broker what to do with the funds after redemption. Eventually, the shareholder may tell the broker to transmit the funds, to buy other securities, or to reinvest the funds in the MLRAT. In the interim, the funds are, by the customer's choice, uninvested; they earn no interest for the shareholder, but their availability may generate interest income for the benefit of MLPFS. Obviously, free credit balances are voluntarily created by the shareholders and, should they so desire, can be avoided completely and continue to earn money market interest credits if the shareholder simply waits until he/she knows that he/she wants to do with redemptions from the MLRAT before requesting redemption. There is no reason for a shareholder to sell MLRAT shares until the precise minute that the shareholder knows that he/she wants to apply the redeemed funds elsewhere.

PMM nonetheless quantified the free credit balances in a manner similar to that used to quantify the float. PMM calculated that on average, $84.0 million per day of free credit balances became available in 1982 from which a benefit could flow to MLPFS. They further calculated that application of the 11.9% commercial paper interest rate for 1982 could yield an annual $10.0 million free credit balance benefit to MLPFS, assuming that they used and put such funds to their best use. For the reasons expressed above concerning the inappropriateness of counting float benefits as an offset to processing costs associated with the MLRAT, free credit balance benefit would be an equally inappropriate item of offset to processing costs. The advisory fee paid to MLAM was not excessive or received in breach of fiduciary duty as a result of the shareholders' choice to utilize optional services available to them which resulted in voluntary creation of unasked for float and free credit balances in the MLPFS account with the investor.

e. Fall-out Benefits

Plaintiff claims that MLPFS has gained large "fall-out" financial benefits in the form of commissions on non-MLRAT securities business generated by MLRAT customers. Plaintiff argued that the fall-out commissions received by MLPFS that would not have been earned but for the fact that MLPFS was able to solicit MLRAT shareholders, is fairly chargeable as an offset to the costs associated with operating the MLRAT. Two issues are presented: the quantification of the fall-out commissions and the appropriate share, if any, that is fairly chargeable as an offset to the costs incurred by MLPFS in operating the MLRAT. The burden of proof is on plaintiff to quantify the fall-out benefits and demonstrate the appropriate share for allocation of said benefits as an offset to costs.

Plaintiff has not attempted to quantify the fall-out benefits in this case and thus has not met his burden of proof. Even if fall-out benefits were fairly allocable as an offset to MLRAT operating costs, no such offset can be considered on the record of this case.

The reasons for plaintiff's failure to quantify the fall-out benefits merit brief examination. Experts for both parties agree that it would not be possible, even through use of today's sophisticated computer equipment and statistical techniques, to obtain estimates of fall-out benefits using data that was available to Merrill Lynch or that was otherwise generally available. This conclusion was reached by Robert R. Nathan Associates and Peat Marwick Mitchell & Co. after hundreds of hours of study at a cost of tens of thousands of dollars to the parties and their counsel. Both experts agreed that the use of computer and statistical techniques would involve regression analysis and that regression analysis, using currently available data, would not be able to prove a causal relationship between MLRAT accounts and fall-out commissions earned by MLPFS.^[19] At most, regression analysis can show how much fall-out business was associated with the MLRAT accounts; it cannot delve into the conscious or subconscious rationale behind individual investors' decisions. Data on the level of non-MLRAT securities purchases by MLRAT shareholders does not shed any light on the reasons behind said purchases.

Plaintiff suggests that these defects could be remedied if survey research were conducted to develop new data that is presently unavailable. Plaintiff's expert, Mr. Gary French, testified that he had made a preliminary analysis concerning the design of a survey of MLRAT shareholders. Mr. French testified that his proposed survey population would consist of MLRAT shareholders in 1982 who had transacted no other business with MLPFS for at least two years prior to opening their MLRAT accounts and who had subsequently engaged in other investment business that generated commissions for MLPFS in 1982. Mr. French testified that he would have to rely on defendants to cull the appropriate survey population from the universe of MLRAT shareholders. It was estimated that such a task might take five months to complete^[20]. He speculated that it would take his firm, Robert R. Nathan Associates, approximately four months thereafter, at a cost of as much as two hundred thousand dollars, to prepare and conduct a survey of a subsample of the survey population in order to estimate the gross fall-out benefits accruing to MLPFS in the form of commissions on non-MLRAT securities purchases. Mr. French admits that he has never attempted to ascertain fall-out benefits from money market funds or bank accounts, nor is he aware of any attempts to estimate fall-out benefits using survey research.

In light of the costs and time requirements involved in conducting such a survey, plaintiff chose not to attempt to quantify fall-out benefits using this technique. While such a survey might be "very much fun" (sic) for Mr. French, the conclusion that it would amount to nothing more than yet another patently frivolous expenditure of time and funds by both parties is supported by PMM's estimate of the size of the survey population and by the inherent unreliability of a gossamer speculation of the type proposed.

Peat Marwick Mitchell surveyed a sample of new MLRAT accounts opened between April 1, 1982 and September 30, 1982. Of the 10,259 new accounts sampled during this period, only 1661 or (16.2%) consisted of investors who had no previous accounts with MLPFS. Of the 1661, only 443, or (26.7%) engaged in non-MLRAT investment with MLPFS in the succeeding six-month period after opening the MLRAT account. Therefore, approximately 4.3% of the new MLRAT accounts sampled would have been included in the survey population that plaintiff contended was the starting point for a calculation of gross fall-out benefits^[21]. The universe of shareholders who potentially generated gross fall-out benefits is relatively insignificant.

In any event, plaintiff's proposed survey is fatally flawed because it relies on the answer to the question: "Would you have made these other [non-MLRAT] transactions in 1982 had you not been contacted or solicited about the possibility of making them by MLPFS?" This question asks an investor to speculate about what he or she would have done if certain events had not occurred. Answers to such a counter-factual question would be nothing more than idle speculation and conjecture. An expert's opinion that is grounded in mere speculation might be of limited interest to academics, but it would provide no credible evidence upon which an evaluation of the reasonableness of MLAM's fees could be made. See, e.g., Zenith Radio Corp. v. Matsushita Electric Indus. Co., 505 F. Supp. 1313, 1324-26 (E.D.Pa.1981); see also Gilbert v. Gulf Oil Corp., 175 F.2d 705, 709 (4th Cir.1949).

Even if the results of plaintiff's survey were more than speculation, they would shed no useful light on the issue faced by the Court in this case because they measure the wrong quantity. Both sides agree that only net fall-out benefits, which consist of gross fall-out benefits minus the costs involved in soliciting and actually conducting said business, could arguably be allowed to offset the operating costs to MLPFS of the MLRAT. Plaintiff's proposed study did not explicitly envision the techniques that would be necessary to calculate the costs of conducting the fall-out business. Plaintiff's expert had never conducted such a cost study, nor could he state at the time of trial how long such a study would take or what it would cost. It would be a total waste of resources to engage in an extensive survey that would, in a light most favorable to plaintiff, only yield part of the necessary information, long after the time for annual approval of the advisory agreement.

The entirety of the preceding analysis of fall-out benefits addresses the quantification issue. Even if net fall-out benefits could be quantified, the Court would still have to determine what share, if any, is fairly allocable as an offset to the MLRAT operating costs. Plaintiff maintains that all the net fall-out commissions earned by MLPFS from MLRAT customers are allocable as an offset to MLRAT costs. This ignores the fact that the fall-out benefits would not accrue absent the use of the sophisticated solicitation techniques that are employed by MLPFS that are amply demonstrated in the record of this case. Plaintiff's position is no different than that of a landowner who demands all of his sharecroppers' net profits on the grounds that without the land there would be no crops or profits. Obviously, MLPFS is entitled to allocate a substantial portion of the net fall-out benefits to itself as compensation for its own entrepreneurial skill. All that MLRAT provides to MLPFS is a mailing list of potential investors in non-MLRAT securities. The value of such a list is far less than the profits that its users can ultimately earn by skillful solicitation and follow-up efforts.

In short, the Court has grave doubts that substantial fall-out benefits would be allocable as an offset to the operating costs of a money market fund, even if they could be accurately quantified. In any event, no attempt at quantification was made in this case and thus plaintiff has not met his burden of proof to show that the presence of fall-out benefits to MLPFS make the advisory fee so unreasonable as to prove that the fee could not have been the product of arm's-length bargaining.

f. The Fairness of the Advisory Fee under Section 36(b)

The foregoing computations of the advisory fee, the float, free credit balances and costs associated with the operation of the Fund supply the raw materials with which the Court can determine if the advisory fee received was so disproportionately large that it bore no reasonable relationship to the services rendered and could not have been the product of arm's-length bargaining. A summary of the relevant data is as follows for the calendar year 1982:

Costs:
MLAM costs                    $   2,050,000
MLPFS Processing Costs          104,530,841
_____________
($14.82/order)
Total                         $ 106,580,841
Revenues:
Advisory Fee                     64,367,392
Float                            26,500,000
Free Credit Balances             10,000,000
_____________
Total                         $ 100,867,392
Net Loss from Fund Operations ($ 5,713,449)

The above summary, in the form most favorable to plaintiff since it includes the quantified value of float and free credit balances even though these items may not be properly includable, indelibly confirms the determination that was reached by the independent trustees in May, 1982, namely, that the advisory fee was not calculated to, and in fact does not, violate Section 36(b) of the Act.

Plaintiffs have failed in this case as in the earlier one to meet their burden of proving that any flaw existed in the oversight and action of the independent trustees with respect to the management contract or that the trustees were in anywise imposed upon by the advisor or failed to appreciate or consider or evaluate the elements and circumstances at the time. There was no excessiveness or unfairness with respect to the advisory fee that would amount to a breach of fiduciary duty. The compensation to MLAM and its affiliates for the calendar year 1982 was fair to the Fund, reasonable in amount and fairly related to the subject matter from which it was derived. The advisory fee is consistent with a fee that would be the outgrowth of arm's-length bargaining. In short, there is no doubt that the Fund and its shareholders received and continue to receive their money's worth from MLAM and MLPFS.

Examination of the information presented to and the deliberations conducted by the trustees (complete in all respects), the relative performance of the Fund in terms of return on investment (above average) and the advisory rates charged by MLAM compared to those charged by other advisors (among the lowest)^[22], all factors which can be considered in determining if a Section 36(b) violation has occurred, further support the conclusion that the advisory fee was fair in all respects.

While the summary figures are very useful in comparing the costs and benefits associated with the Fund, it would be error to become overly concerned with their precision. The task presented to the Court by Section 36(b) is to examine all the circumstances surrounding the Fund to determine whether the fees are so excessive that they could not reflect the results of arm's-length bargaining. That question is not resolved by indulging carping criticisms of the degree of precision involved in measuring processing costs. Even if the PMM estimate of the account executive compensation arising out of the Fund's operations were substantially in excess of the true magnitude of such costs, the Court's conclusion concerning the fairness of the advisory fee would be the same. Merrill Lynch and its affiliates are entitled to recoup their costs and to make a fair profit without having to fear that they have violated Section 36(b) of the Act. ¶ Despite ample opportunities in two trials, plaintiff has not offered a yardstick for measuring when the profits generated by the Fund would render its fee excessive. In the absence of a suggested metric, the Court has considered all of the circumstances surrounding the operation of the Fund and has reached the inescapable conclusion that even when considered in combination with the quantifiable and unquantifiable float and commission benefits associated with the Fund, the advisory fee is not excessive in violation of Section 36(b) of the Act. Indeed, a careful analysis of the costs and benefits associated with the Fund makes it clear that the Fund operates at or near the break-even point and may even be a loss leader^[23]. Evidently, the trustees have reached the same conclusion, since the 1983 advisory contract, which is not sub judice, provides a circumstantial indication that the trustees think that the Fund's shareholders have been receiving unpaid for benefits at the expense of MLAM and its affiliates.

Accordingly, the Court concludes that these actions should be, and they are, dismissed on the merits, with costs to be taxed by the Clerk. A judgment shall be entered accordingly.

The foregoing shall, together with the separate findings filed herewith,^[*] agreed upon by the parties or resolved by the Court where disputed, constitute the findings of the fact and conclusions of law required by Rule 52(a), Federal Rules of Civil Procedure.

NOTES

[1] In this case, plaintiff Andre again filed in her amended complaint claims under §§ 15(a), (b), (c), and 20(a) of the Investment Company Act. Andre asserted that these claims were not decided in Gartenberg I. On April 18, 1983, the Court struck the charges under Sections 15 and 20 as having been found in the earlier litigation to be devoid of legal merit and as having been asserted in bad faith. Plaintiff Andre, and her counsel, were assessed attorneys' fees in the amount of $5,000.00 for reasserting those claims frivolously. Plaintiff Andre has appealed that dismissal, and the imposition of the sanctions. The claims in Andre II, brought under § 36(b) were scheduled to be tried along with the same claims in Gartenberg II. However, at the opening of the trial, plaintiff Andre moved for a discontinuance with prejudice of its claims under § 36(b). Since the Andre complaint was cast in terms of a class suit, and to avoid any possible impairment of a class action status of Gartenberg II, the Court deferred a ruling on Andre's motion until Gartenberg II had been completed. Plaintiff Andre withdrew from further participation in the case during the trial. Andre's motion to dismiss her claims, with prejudice, is now granted.

[2] MLPFS and Merrill Lynch & Co. are sued under Section 36(b) and the latter is also charged as a controlling person under Section 48, 15 U.S.C. § 80a-47.

[3] N.Y. Times, June 24, 1983, p. D3, at col. 4.

[4] Donoghue's Money Fund Report presently groups money market funds according to the types of money market instruments in which they invest. The MLRAT follows a conservative investment strategy and only invests in domestic prime instruments. Even when compared with funds that invest in riskier instruments such as Eurodollar securities, the MLRAT ranks 11 out of the 38 funds in existence during the five-year period.

[5] Donoghue's Money Fund Report performance results are reported with effective annual yields which assume reinvestment of dividends. The yield on the MLRAT for 1982 excluding reinvestment of dividends was 11.95%.

[6] June figure as of June 22, 1983. See N.Y. Times, June 24, 1983, p. D3, at col. 3.

[7] Plaintiff Gartenberg has nonetheless significantly increased his MLRAT holdings and those of his family since Gartenberg I. As of April 22, 1983 Gartenberg personally held 1,338 shares of the MLRAT and held 950 shares in an UGMA account for his son. Gartenberg's wife owned 809 shares as of that date. In addition, Gartenberg, who is an attorney, held 111,000 shares on behalf of his clients as of April 25, 1983 and had held 235,455 shares on July 30, 1982.

[8] In his brief submitted before the trial, plaintiff argued that defendants' failure to quantify the float in advance of the trustees' 1982 deliberations violated MLAM's duty under Section 15 of the Act. Section 15 requires that the advisor "furnish, such information as may reasonably be necessary to evaluate the terms [of the advisory agreement]." Nothing in Section 15 suggests that the advisor must conduct elaborate and expensive studies to create information that could not reasonably have been expected to be necessary in the trustees' deliberations. Cases cited by plaintiff did not support the suggestion that the advisor is to conduct expensive studies to provide information that is not usually the type relied on by independent trustees. See, e.g., Tannenbaum v. Zeller, 552 F.2d 402, 418 (2d Cir.), cert. denied, 434 U.S. 934, 98 S. Ct. 421, 54 L. Ed. 2d 293 (1977); Fogel v. Chestnutt, 533 F.2d 731, 745, n. 13 (2d Cir.1975), cert. denied, 429 U.S. 824, 97 S. Ct. 77, 50 L. Ed. 2d 86 (1976). The 1982 trustees' meeting occurred seven months prior to the Court of Appeals suggestion to the trustees that they consider quantifying the float. Immediately after that suggestion, the trustees commissioned the suggested study and ultimately their judgment that the fee was not excessive was amply vindicated by the credible evidence on the total costs and benefits associated with the Fund.

Plaintiff also complains that the trustees were incorrectly informed that fall-out commission benefits could not be quantified and that they were not informed that MLPFS had concluded that Fund customers were particularly receptive to "cross-selling" attempts. The evidence makes clear that the trustees were aware of the concept that such benefits were not quantifiable. The efforts of both PMM and Robert R. Nathan Associates, examined in detail below, confirm that such fall-out benefits are not readily and accurately quantifiable. These additional claims of misrepresentation have no merit.

[9] The minutes of the May 12, 1983 trustees' meeting indicate that Mr. Cecil stated that "the trustees had been advised that [the question of float] was presently being studied by Merrill Lynch & Co., Inc. and that it was not presently known what percentage of Merrill Lynch's free credit balances arose from float on redemption checks." At trial, Mr. Cecil testified that it was his understanding at the time of the May, 1982, meetings that Merrill Lynch & Co., Inc. was not conducting and had not conducted a study with respect to benefits to MLPFS from free credit balances or float. In all events, the trustees were correct in assuming that no study of float had been completed with results that were available for consideration.

[10] In 1982, an even more arbitrary technique of allocation was employed, resulting in a calculation of $32,183,696 for "Transaction processing costs." This figure represents 50% of the advisory fee.

[11] The failure of DHS to include an amount for processing costs that fully reflects the total magnitude of such costs has no effect on the fee charged by MLAM to MLRAT, since the fee is a function of MLRAT assets, not MLAM profits.

[12] Sections 15(a) and 15(c) of the Act also require that the terms of the advisory agreement be approved annually by a majority of the disinterested trustees. Under Section 15(a)(3) either the trustees or the shareholders can terminate the advisory agreement on sixty days written notice to the advisor.

[13] As in the first trial, the PMM studies are comprehensive, involving more than 40 professionals at a cost of over $400,000.

[14] MLPFS had enacted changes that simplified the process of handling MLRAT transactions. For example, MLPFS now employs an encoded deposit feature that allows a MLRAT shareholder to purchase additional shares directly through a cashier. MLPFS also simplified the process for selling MLRAT shares in order to purchase other securities and conversely, MLPFS also established Regional Operations Centers which control many of the transactions emanating from the MLPFS branches which formerly were processed directly in New York.

[15] The PMM survey was conducted in the first quarter of 1983 at a time when MLRAT transactions represented 23% of MLPFS transactions. In 1982, MLRAT transactions represented 31% of MLPFS transactions. PMM actually found that 5.6% of account executive time was devoted to the MLRAT in the first quarter of 1983. This figure was then adjusted to reflect the fact that MLRAT transactions made up a greater share of total transactions in 1982. The result is the PMM calculation that 7.4% of account executive total time was allocated to the MLRAT.

In his direct testimony, Mr. Peppet referred to the 5.6% figure as the basis of the PMM cost allocation calculation. On cross examination, Mr. Peppet referred to a 7.6% figure which represents the percentage of product related time (total time minus executive time not spent working on MLPFS products) as a basis for the cost allocation calculation. The exhibits clearly indicate, and on redirect Mr. Peppet confirmed, that PMM employed the more conservative 5.6% figure and then adjusted it to reflect the fact that MLRAT transactions were proportionally greater in 1982 with the result that 7.4% of the costs associated with account executive compensation were allocated to the MLRAT.

[16] The documentary evidence relied upon by plaintiff reaches the same conclusion. See Marple, The Relative Contribution Approach to Management Reporting in National Association of Accountants on Direct Costing — Selected Papers 421 (R. Marple ed.).

[17] Redemptions can be made directly by check, wire or transfer to another account; funds can also be redeemed for the purchase of securities; shareholders can also write personal checks on their MLRAT accounts.

[18] There is also no float on wire transfers which result in wire of funds to the shareholder's bank account within a single day.

[19] Mr. Peppet also testified that the results of the regression analysis might be unreliable due to the presence of multicolinearity. This problem arises when the explanatory variables in the regression equation are correlated with each other in violation of the underlying statistical assumption that all explanatory variables are independent of each other. Peppet also testified that the data necessary for a regression analysis would be difficult and expensive to gather. Mr. French of Robert R. Nathan testified that he had not discovered any way to quantify fall-out benefits by using computer equipment and the data stored on that equipment.

[20] At a pre-trial hearing in this case, Mr. William Stewart testified that isolation of the survey population from the MLRAT universe would involve making extractions from and merging numerous computer data files, an exercise that would take approximately five months to do at a cost of $60,000.

[21] It should be noted that the 4.3% figure covers the time period which includes August, 1982, which was the month where common stock prices made a dramatic surge upward. The number of MLRAT shareholders making investments in other securities would be expected to be unusually high in this period, without regard to "fall-out" investments.

[22] Effective net compensation at a rate of up to 0.50% has been sanctioned by Courts as appropriate under § 36(b). See, e.g., Levy v. Bernstein-McCauley, Inc., No. 80 Civ 2657 (S.D.N.Y. settlement signed Apr. 14, 1982).

[23] The same possibility was recognized in Gartenberg I, 528 F.Supp. at 1053 & n. 20.

[*] The findings of fact, which are on file in the office of the clerk for the Southern District of New York, have been deleted for purposes of publication.