Maryland National Bank v. Cummins

588 A.2d 1205, 322 Md. 570, 1991 Md. LEXIS 75

• Court: Court of Appeals of Maryland
• Decided: April 18, 1991
• Docket: 36, September Term, 1990
• Status: Published

Opinion

RODOWSKY, Judge.

In this class action the Circuit Court for Baltimore City held that investment practices for personal trusts followed by a corporate trustee for nearly ten years violated the prudent investor rule. We consider multiple issues relating to liability and to the relief granted.

The corporate trustee is the appellant, Maryland National Bank (MNB). Appellees, the plaintiffs below (Plaintiffs), are income beneficiaries of a testamentary trust administered by MNB which was funded in September 1972. Plaintiffs sued on April 6, 1983. The circuit court certified Plaintiffs as representatives of a class consisting of “the life tenants of all personal trusts (inter vivos and testamentary) where [MNB] was a trustee at any time between September 1, 1972 until July 26, 1982 [the Class Period].” The Class Period ends as of the date when, by retroactive adjustments, MNB modified the practices complained of in the complaint.

During the Class Period MNB administered an average of 2,000 personal trusts under the policies hereinafter described. Cash receipts for all personal trusts were initially deposited to a demand deposit account (DDA) which paid no interest. It was MNB’s policy to leave income cash in the DDA, after receipt and prior to distribution, for those trusts in which income was distributed on a regular schedule, usually quarter annually, but sometimes monthly. Also MNB’s policy was to invest only in increments of $1,000 the principal cash of those trusts which had assets exceeding approximately $150,000. In these trusts principal cash left uninvested in the DDA at any given time could range from $0 to $999. 1 For a majority of smaller trusts, those having assets of roughly $150,000 or less, MNB’s policy encouraged investment in collective investment funds (CIFs) in increments of $500. CIFs are in-house, mutual funds which provide diversification of investments, particularly by smaller trusts. 2 See 12 C.F.R. § 9.18 (1990).

These policies were in effect at MNB in September 1972, were reduced to writing in early 1976, and were not substantially altered during the remainder of the Class Period. Plaintiffs contended that these practices were an imprudent failure to invest cash held in trusts.

At trial MNB sought to justify its policies principally on the ground that investing cash more fully would not have been cost effective during the Class Period. MNB never satisfied the trial judge that, as a practical matter, it could not have invested substantially all of the available cash. The heart of the trial court’s finding is

“that although computers might have simplified the job, MNB had the ability, manually throughout the class period, to invest these monies for the benefit of the beneficiaries but chose not to. The result was there were high cash balances in the DDA account that the bank was able to use for its own gain.”

(Emphasis added).

The trial court entered judgment against MNB for $3,857,129.69, consisting of lost return to the Plaintiffs on uninvested trust cash, of compounded prejudgment interest, and of a surcharge of ten percent of trustee’s commissions. MNB appealed, and the Plaintiffs cross-appealed, to the Court of Special Appeals. We issued certiorari on our own motion prior to consideration of the case by the Court of Special Appeals.

MNB contends that the circuit court erred, both in finding a breach of the trustee’s duty and in fashioning the relief. In computing its judgment the circuit court used as factors the stipulated average daily balances in the DDA for each calendar year of the Class Period. The circuit court directed that, for the years 1972 through 1976, MNB pay an amount equal to five percent on those average DDA balances, an investment return analogous to passbook savings account interest earnings. For the years 1977 through the end of the Class Period in September 1982 the court ordered MNB to pay amounts to be determined, in the manner testified to by Plaintiffs’ expert, by applying to the respective average DDA balances certain percentages that represented the average dividend rate paid by a money market mutual fund in the particular year. The court directed that the end products so calculated for each year bear annual interest thereafter at the rate of ten percent and that that interest be compounded to the date of judgment.

In this appeal MNB challenges liability, the base figures used in calculating damages, the rate of prejudgment interest, the compounding of prejudgment interest at any rate, and the surcharge against commissions. The trustee also raises a laches defense. Plaintiffs, on the other hand, assert that the circuit court erred in using a hypothetical return to the Plaintiffs as the model for relief. They say that the court should have awarded Plaintiffs an amount representing the “profit” realized by MNB on the trusts’ uninvested cash.

I

This case, in large measure, concerns the banking business, trust accounting, and data processing systems. Per sonal trust accounting segregates principal and income. In an entirely manual system there would be separate ledger cards for each of those two components of a trust account. At MNB, trust accounting was automated throughout the Class Period. Further advances in automation for trust accounting later came about. MNB now uses advanced systems to invest fully principal and income cash. This case, however, must be decided within the framework of circumstances during the Class Period.

Under MNB’s policy of carrying trust cash in the DDA, MNB received the considerable advantage of having trust cash available for lending by MNB without incurring any interest cost. An internal study by MNB in 1981 computed the value of the use of uninvested trust cash to be $149,300 per $1 million of deposit balances. This represented the then current Federal Reserve Bank interest charge to borrowing banks of 13.18% plus a processing fee of 1.75%.

MNB’s system in effect treated separate trusts as one common trust in certain aspects of the receipt and disbursement of income and of the purchase, retention and sale of principal assets. In this system the DDA was a common, non-interest bearing, checking account for all personal trusts administered by MNB. 3 Securities in which individual trusts were invested, or in which a CIF invested for participating trusts, were held in the name of MNB as trustee, or of its nominee. If MNB’s investment decision were to move out of a given security, MNB would sell at one time a lot of that security held in MNB’s or its nominee’s name and representing holdings of a number of per sonal trusts. On the settlement date MNB would credit the selling trust accounts with principal cash in the amount of their respective net sales proceeds. On receipt of the check for the total sales proceeds, MNB would deposit it to the DDA for clearing. When purchasing securities to be held directly by trusts, principal cash of a number of trusts, to the extent available under MNB’s policy, could be pooled for the acquisition of a security in one lot. Principal cash of the buying trusts would be respectively debited on the settlement date by the amount of each trust’s proportion of the purchase price, and principal assets would be credited for each trust’s proportion of the lot. On receipt of the certificate MNB would pay by a check drawn on the DDA which would in due course clear. The certificate was placed in MNB’s vault.

From the standpoint of income, interest or dividends on given securities in which multiple trusts might be invested were received by MNB in the form of a check payable to MNB which was deposited to the DDA for clearance. In any particular trust, the portion of the income receipt representing that trust’s investment in the security was credited to the trust, automatically under a computer program, when the dividend or interest was scheduled to be paid by the security issuer. Disbursements of income on behalf of a particular trust account were debited to that trust account when the disbursement was ordered, or programmed, to be paid. The check in payment was then drawn on the DDA and ultimately was presented for payment from the DDA.

At all times during the Class Period, MNB furnished to members of the Plaintiffs’ class periodic statements reflecting, for each trust in which the beneficiaries were interested, the trust’s income receipts, its principal or assets, including principal cash, its disbursements, and its income cash on hand, if any. MNB was able to make all of the allocations to individual trust accounts necessary for the system to function. MNB was also able to comply with 12 C.F.R. § 9.13(b) (1990) which, throughout the Class Period, provided:

“(b) The 11 investments of each account shall be either:

(1) Kept separate from those of all other accounts, except [for CIFs], or

(2) Adequately identified as the property of the relevant account.”

This trust accounting was to an extent theoretical because of the differences in timing of debits and credits to the trust account in relation to cash in the DDA. Principal transactions were recorded in trust accounting when the transaction was scheduled to close. In sales by a trust, principal cash was credited in trust accounting on the settlement date but the check in payment would not have cleared. In purchases by a trust, principal cash was debited in trust accounting on the settlement date, but the check from the DDA was not issued until the security was received. With respect to income, cash was credited in trust accounting on the date when the interest, dividend, or other income item was payable, although the check in payment would not have cleared. Income cash in trust accounting was debited when a distribution to an income beneficiary was scheduled to be made, or when payment of an expense was ordered, although cash in the DDA would not yet have been debited. As a consequence, the DDA ordinarily trails trust accounting. In transactions involving the receipt of principal or income cash the trust account reflects cash which may not have been received in the DDA. In transactions involving the disbursement of principal or income cash from the DDA, the fact that the DDA trails trust accounting can result in the trust statement sent to the income beneficiaries reflecting less cash than is actually in the DDA. Receipts produce a float favorable to the trust beneficiaries, and disbursements produce a float favorable to the bank. 4

Another benefit derived by MNB from invested trust cash carried in the DDA was a reduction of MNB’s reserve requirements. Under Federal Reserve and Office of the Comptroller of the Currency (OCC) procedures, this was accomplished by MNB classifying, for reserve purposes, part of the DDA as a time deposit, even though the deposit was non-interest-bearing. See E. Herman, Conflicts of Interest: Commercial Bank Trust Departments, at 111 (The Twentieth Century Fund, 1975) (Herman). Herman estimates that “for the large banks, between about 50 percent and over 90 percent of all in-bank trust deposits seem to be classified as non-interest-bearing time.” Id.

It has long been settled under Maryland law, however, that a bank-trustee does not commit a per se breach of trust by depositing with itself, on the banking side of its operations, cash which it holds as trustee in its trust department operations. See Newark Distributing Terminals Co. v. Hospelhorn, 172 Md. 291, 191 A. 707 (1937); Corbett v. Hospelhorn, 172 Md. 257, 191 A. 691 (1937); Ghingher v. O’Connell, 165 Md. 267, 167 A. 184 (1933); Real Estate Trust Co. v. Union Trust Co., 102 Md. 41, 61 A. 228 (1905); Safe Deposit & Trust Co. v. Magruder, 34 F.Supp. 199 (D.C.Md.1940). 5 No party to this action challenges, at least directly, the continued vitality of this rule of Maryland law. The Plaintiffs do not urge, and the trial court did not find, that MNB engaged in prohibited self-dealing by lending, as trustee, trust cash to itself, as banker. Rather, the liability issue is whether MNB, by failing to obtain or pay a reasonable return on the trust cash, violated its duty prudently to invest on behalf of the class.

II

“Maryland follows a ‘prudent person’ standard for investment by fiduciaries.” Attorney Grievance Comm’n v. Owrutsky, 322 Md. 334, 350 n. 7, 587 A.2d 511, 519 n. 7 (1991). In Board of Trustees of the Employees’ Retirement Sys. v. Mayor & City Council of Baltimore City, 317 Md. 72, 103, 562 A.2d 720, 735 (1989), cert. denied, — U.S. -, 110 S.Ct. 1167, 107 L.Ed.2d 1069 (1990), and in Shipley v. Crouse, 279 Md. 613, 621, 370 A.2d 97, 102 (1977), we quoted G. Bogert, The Law of Trusts and Trustees § 541 (2d ed. 1960) for the principle that in all management of the trust a trustee is required to manifest “the care, skill, prudence, and diligence of an ordinarily prudent [person] engaged in similar business affairs and with objectives similar to those of the trust in question.” This duty “is not necessarily to maximize the return on investments but rather to secure a ‘just’ or ‘reasonable’ return while avoiding undue risk.” Board of Trustees, 317 Md. at 107, 562 A.2d at 737.

Judicial decisions rendered over the years preceding the Class Period, and during the Class Period, presented ample illustrations of fiduciaries who were liable to surcharge for failure to invest funds entrusted with them to the extent required by the duty undertaken. See Cheyenne-Arapaho Tribes of Indians of Oklahoma v. United States, 512 F.2d 1390, 206 Ct.Cl. 340 (1975) (summary judgment for Government denied where Indian funds left in Treasury at no interest, or at four percent, when outside investments available at higher returns); Manchester Band of Pomo Indians, Inc. v. United States, 363 F.Supp. 1238 (N.D.Cal.1973) (summary judgment on liability against United States where Indian funds remained in Treasury at four percent between 1938 and 1959 when short term government bonds paid higher rate); Estate of Orrantia, 36 Ariz. 311, 285 P. 266 (1930) (self-depositing without interest for six months by executor bank); New England Trust Co. v. Triggs, 334 Mass. 324, 135 N.E.2d 541 (1956) (self-deposit without interest by trustee bank for over two years); In re Doyle’s Will, 191 Misc. 860, 79 N.Y.S.2d 695 (1948) (testamentary trustee deposited in commercial bank account paying approximately .75% when savings banks paid 1.5%, compounded semi-annually); In re Haigh’s Estate, 133 Misc. 240, 232 N.Y.S. 322 (1928) (executor bank self-deposited at 2% and 2.25% funds which decedent had had invested at four percent); Reid v. Reid, 237 Pa. 176, 85 A. 85 (1912) (bank trustee of security in commercial transaction self-deposited funds without interest and was chargeable with interest rate paid by it to third parties depositing in similar accounts).

In the instant matter the Plaintiffs’ proof of the policy practiced by MNB throughout the Class Period is legally sufficient to support the allegation that MNB did not act as a prudent investor. Reasonable persons do not, as a matter of policy, continuously leave uninvested sums up to $999 (if principal cash in a trust account is viewed in isolation) and perhaps ranging in six figures (if principal cash in the trust accounts is considered in the aggregate). 6 Reasonable persons do not, as a matter of policy, leave uninvested for up to nearly three months income cash aggregating hundreds of thousands of dollars. 7

The burden was therefore on the trustee to persuade the trial court that a prudent investor would not invest the trust cash which MNB left in the DDA. See Goldman v. Rubin, 292 Md. 693, 713, 441 A.2d 713, 724 (1982); Lopez v. Lopez, 250 Md. 491, 501, 243 A.2d 588, 594 (1968).

The circuit judge found that MNB had available a number of feasible modes of investment. A master passbook savings account or individual passbook savings accounts could have been used for trust cash during the Class Period at a return of approximately five percent. CIFs could have been used more fully and, in lieu of a minimum participation of $500, the price of a unit could have been greatly reduced, for example, to $3.00. By using a lower minimum participation, more trust cash could have been invested in master notes. MNB invested in this form of commercial paper by pooling principal cash from various trusts, but MNB did so only in minimum participations of $1,000 per trust. A further fact-finding was that, as early as 1976, trust cash lying in the DDA could have been invested in money market mutual funds. This could have been done after the checks for receipts had been cleared through the DDA. Thereafter, the mutual fund could have transferred cash periodically to an MNB disbursement account to cover income distributions and expense disbursements of trusts.

MNB advances to us four reasons why it says the circuit court clearly erred in finding a breach of duty. They are (A) benefits to the beneficiaries of the DDA system justified the non-payment of interest; (B) MNB’s policy complied with OCC requirements; (C) MNB’s policy conformed to the virtually universal practice of the trust industry; and (D) investing trust cash would have required a prohibitively expensive allocation of the investment income to individual trust accounts.

MNB produced sufficient evidence to support the findings which it now urges us to make. Witnesses who opined that MNB’s policy was reasonable include the former Chief National Trust Examiner of the OCC and the chief executive officer of a leading developer of software for trust department accounting, in addition to various trust department officers from MNB. Fatal to MNB’s appeal are that the totality of the evidence does not present a question of law and that MNB’s witnesses never persuaded the fact finder that MNB’s policy was that of a prudent investor.

A

MNB submits that its cash management policy provided significant benefits to the income beneficiaries of personal trusts. One might expect MNB’s analysis to focus on those benefits which would have been lost to the class members had MNB invested trust cash which lay in the DDA. MNB, however, presents a check list of some nine benefits, most of which describe economies of scale manifested in the bulk trading of securities, in the use of nominees, and in arrangements for the physical custody of securities. These benefits result from management of many trusts, and not from foregoing paying interest on trust cash.

MNB asserts that the beneficiaries received free checking and interest-free coverage of overdrafts which, it submits, alone are sufficient to make MNB’s policy reasonable. As explained above, under the dual accounting system, depending on the transaction, there could be in the DDA cash originating with a given trust which was not reflected on that trust’s account, or there could be cash reflected on a trust’s account which had not yet reached the DDA. The result is not, however, simply a wash, because MNB sought to control overdrafts. MNB’s audit report of January 15, 1976, on uninvested trust cash sets forth MNB’s cash management policy. It states flatly that “[tjhere shall be no principal overdrafts in any accounts,” and that “[tjhere shall be no income overdrafts, except those caused by the payment of accruals on the purchase of securities.”

B

Throughout the Class Period 12 C.F.R. § 9.10(a) (1982) provided:

“Funds held in a fiduciary capacity by a national bank awaiting investment or distribution shalí not be held uninvested or undistributed any longer than is reasonable for the proper management of the account.”

The evidence is uncontradicted that MNB was regularly examined by the OCC and that the examiners took no exception to MNB’s policy. Indeed, OCC’s former Chief National Trust Examiner testified that retention of principal cash below $1,000 was in compliance with the regulation. This is evidence that MNB acted prudently, but it neither compels that finding nor renders the trial judge’s contrary conclusion clearly erroneous. In this respect the instant matter is analogous to Ellsworth v. Sherne Lingerie, Inc., 303 Md. 581, 495 A.2d 348 (1985), a products liability case involving flammable fabrics, where we reversed a verdict for the defendant based on misdirection of the jury and on the erroneous exclusion of evidence. Significant here is that, although the fabric involved in Ellsworth complied with the federal standard for the flammability of clothing textile, that factor was not determinative on liability — not even as to the plaintiff’s cause of action sounding in negligence.

C

MNB relies on what it characterizes as the “virtually universal practice of the trust industry.” The trustee also acknowledges, however, that “industry practice cannot control legal obligation.” Appellant’s Brief at 16. Consequently, this argument likewise is to be evaluated under the clearly erroneous standard.

In support of its position MNB emphasizes the 1975 study for The Twentieth Century Fund, see Herman, supra, at 112-13, and the testimony of the former OCC official, both to the effect that, among banks, the minimum for investment of principal cash was most frequently $1,000, and that income cash need not, and perhaps could not, be invested.

At MNB the origins and continuation of the cash management policy do not reflect conscious attention to the duty prudently to invest. The trust department’s legal counsel explained that in trusts “where the income was paid out on a regular basis, it was felt that income was a sacrosanct item during this class period, [and] that it had to be there on call for beneficiaries who may call up in the middle of the month and say, T need my income.’ ” We have neither been cited to, nor found, any judicial authority supporting that notion. A corollary of this notion was that MNB did invest the income of trusts, referred to as “complex” trusts, in which income was accumulated for discretionary payment. As to principal cash, MNB’s policy of investing in increments of $1,000 antedated the Class Period. It arose when investment officers were expected to invest short term principal cash in the commercial paper of Sears, Roebuck and of General Motors Acceptance Corporation which was only available in $1,000 multiples.

The leading Maryland case on self-depositing by a trustee of trust funds, decided in 1905, casts doubt on MNB’s policy. Real Estate Trust Co., 102 Md. 41, 61 A. 228, involved a bank which, as trustee, was mortgagee of real estate securing corporate bonds. After default on the bonds, a portion of the realty was condemned, and the trustee deposited the condemnation proceeds with itself in a non-interest bearing account. In rendering a court accounting of these funds, counsel for the trustee directed the court auditor to charge the bank, as trustee, with two percent per annum interest. The circuit court increased the rate to six percent. Before this Court the unarticulated concession was that the bank-trustee should pay interest on the self-deposited funds, and the dispute was limited to the proper rate between the alternatives presented by the record.

For many years preceding, and during most of, the Class Period national banks were prohibited by federal law from paying interest on demand deposits. See 12 U.S.C. § 371a. It was not until Public Law 96-221, effective December 31, 1980, 94 Stat. 146-47, that authorization to pay interest on “NOW” (negotiable order of withdrawal) accounts was extended to depository institutions nationwide. “Nevertheless, it remains hornbook [trust] law that cash should not be left unproductive except for overriding liquidity needs.” Lybecker, Regulation of Bank Trust Department Investment Activities: Seven Gaps, Eight Remedies, Part I, 90 Banking L.J. 912, 929 (1973). Thus, the prohibition of interest on the DDA is not defensive. The issue here is whether MNB should have invested funds which it allowed to remain in the DDA or whether, as MNB contends, its cash management policy satisfied its duty prudently to invest.

In answering that issue the trial court was not required to accept as controlling the practices of even a majority of commercial banks which operated trust departments, because of the inherent conflicts of interest. Writing in 1973, Lybecker said:

“For trust accounts which do not meet the short-term securities funds participation standards’, and for banks without the computer time or mechanical equipment necessary to administer short-term securities funds, the alternative employed most frequently has been to deposit the ‘uninvestable’ cash in the trust department’s demand deposit account on the commercial side of the bank complex. Because banks are prohibited from paying interest on demand deposits, the bank complex ‘saves’ the interest on the cash which might otherwise be deposited in a savings account, and has significantly more money available for commercial department operations. The bank complex sometimes gives the trust department, not the individual trusts, credit for the amount saved, thereby increasing the profitability of trust department operations. Thus, those banks lacking short-term securities funds arrangements sacrifice trust income for liquidity while enjoying the use of the uninvested cash. The Hunt Commission Report has expressed concern that more frequent analysis and review of bank policy may be desirable whenever a bank lacks short-term securities funds or sets minimum participation requirements for such funds at levels too high for some accounts to qualify.” 8

Lybecker, Regulation of Bank Trust Department Investment Activities, 82 Yale L.J. 977, 985 (1973) (footnotes omitted).

In the 1975 study for The Twentieth Century Fund, Herman presented the following analysis concerning bank practices generally:

“Most telling, perhaps, is the slowness with which banks have developed and improved machinery for keeping trust cash to a minimum, and even within banks the application of these improvements has lagged. Master Note plans spread rather slowly in the 1950s and 1960s, and more sophisticated vehicles for investing trust cash in bank-managed pools of money market instruments have developed even more slowly. Computerization has made it possible for the larger trust banks to keep their trust accounts pretty close to fully invested. Although progress along these lines was significant after the mid-1960s, some billion dollar trust banks did not even have Master Note plans in 1973. The tendency to apply the most sophisticated cash management techniques (such as liquidity pools) to pension funds, while preserving the distinction between income and principal cash for personal trusts, can be explained in part by the smaller size and more complex demands of trust accounts, by statutory limits on personal trust fees, and by legal obstacles to a more efficient handling of their cash. But it is also evident that trust founders and beneficiaries are weaker bargainers, less well informed, and less aggressively pursued by the banks, and that the banks have a stake in preserving such distinctions as that between income and principal. A truly ‘undivided loyalty’ to beneficiaries would have resulted in a more rapid advance in the cash management of personal trust accounts.”

Herman at 116 (emphasis added; footnote omitted).

Similarly, Lybecker recommended that trust examiners “should serve as a focal point for the accumulation of technical information, thereby making it possible to lower the floor for short-term securities fund participation to zero and to extend the possibility of collective cash investment to all sizes of cash balances.” Lybecker, Regulation of Bank Trust Department Investment Activities: Seven Gaps, Eight Remedies, Part II, 91 Banking L.J. 6, 13 (1974) (footnote omitted).

In the only recent decision dealing with issues of the type presented here, the bank trustee self-deposited “cash awaiting permanent investment or distribution and not needed for immediate disbursements ... in passbook savings accounts____” Van de Kamp v. Bank of America Nat’l Trust & Savings Ass’n, 204 Cal.App.3d 819, 841, 251 Cal.Rptr. 530, 538 (1988). This practice had existed for a period of time, unspecified in the opinion, prior to January 1976 when the defendant implemented an advance in electronic data processing which permitted automatic daily investment of principal cash. 9

An internal MNB report of September 3, 1982, surveyed nine competitors’ trust department investment increment practices. The report does not disclose how long the reported practices had been in effect. Philadelphia National Bank was investing principal and income cash down to the last dollar. Four banks invested principal and income cash in increments of $100, one bank in increments of $500 as to both sources, and one bank in increments of $1,000 each. An eighth bank did not invest income cash but invested principal cash in increments of $100. Only one other bank in the study failed to invest income cash at all and invested principal cash only in $1,000 increments.

Thus, although the practices followed by MNB may well have been widespread in the banking industry, particularly in the early years of the Class Period, the trial court did not err by declining to go along with the crowd.

D

The major theme of MNB’s explanation of its practices is that doing otherwise would have been prohibitively expensive. During the Class Period to November 1979 data processing for the MNB trust department was a batch system which produced written reports. That system was capable of producing a report, within twenty-four hours, of the uninvested cash in any given account, or in a number of accounts. It could produce a report of uninvested cash in all personal trust accounts in a turnaround time of somewhat more than one day. In contrast, “cash sweeping” of trust accounts is the instantaneous, computerized investment and disinvestment of cash. A program for principal cash sweeping was designed in 1975 and became available for banks of MNB’s size around 1977 when MNB contracted for that system. The system was fully operational by November 1979. Income sweeping was not available until after the Class Period. 10

Essentially MNB submits that it could not allocate income to the trusts if the DDA cash were invested, and that, prior to cash sweeping, it could not precisely account for income earned, if DDA cash were invested. In evaluating the testimony concerning prohibitive expense, the trial judge could consider that, when principal cash sweeping became available, MNB’s system was programmed to continue the practice of investing principal cash only in $1,000 increments, except for estates invested in CIFs, and that when income cash sweeping became available, MNB would not initially accept its use as part of the duty which the trustee assumed in consideration of commissions. See n. 10, supra.

One prong of the trustee’s argument assumes that MNB would have to account for the income earned, trust by trust, dollar for dollar, and day by day. MNB says that this degree of precision is required by the trial court’s opinion which used 100% of the stipulated annual average daily balances of the DDA as the bases to which rates of return were applied. We interpret the trial court’s opinion as holding that MNB did not reasonably invest the trust cash. The trial court neither directed the form of investment, nor ignored the need to convert to cash before reinvestment or distribution, nor disregarded the concept of de minimus.

The remedy ordered here is necessarily an approximation of what might have been earned had MNB chosen to invest cash available for investment, instead of leaving it in the DDA. MNB destroyed records reflecting the actual balances in the DDA so that the figures presented by MNB and accepted by the Plaintiffs in a stipulation are necessarily reconstructed in some fashion. For example, for 1982 the trial court used in its computation an average daily balance, taken from the stipulation, of $1,096,479, which was then adjusted for the partial year. An MNB study of September 3, 1982, determined that approximately $2.4 million of income cash alone would become available for invest ment if personal trust income cash were invested in increments of $100. An equity court may afford a remedy by approximation, particularly where the defendant’s conduct makes precision impossible. See Hanley v. Stulman, 212 Md. 273, 277, 129 A.2d 132, 136 (1957) (description of right-of-way by prescription in injunction against encroachment may approximate course of right-of-way in area where defendant’s construction activity obliterated actual path).

Further, the trial court’s use of five percent interest, by analogy to passbook savings interest, was conservative. MNB invested some principal cash directly in master notes and, indirectly through CIFs, in various forms of short-term investments. The trial court could have woven the higher rates of return realized on those investments into the relief granted for the earlier years of the Class Period on the theory that lower investment increments could have been used.

Some six years before the end of the Class Period Professor Scott published his conclusion that a trustee had a duty to invest income cash. In the 1976 supplement to 2 A. Scott, The Law of Trusts § 181, at 70-71 (3d ed. 1967), the author included a discussion concerning the temporary production of income.

“The question arises not only as to funds awaiting investment or distribution, but also as to income payable to the beneficiaries at stated periods, quarterly or yearly. What should be done with income which has been received by the trustee but which is not yet payable to the beneficiaries? Should it be deposited by a trustee in a savings account? Should it be deposited by a corporate trustee in a savings account of its own, or that of another bank? Or may it credit the trust account and use the funds as its own? It would seem that the general duty to make trust property productive of income should be applicable under present conditions.”

(Footnotes omitted).

In the 1979 supplement to his work, Professor Scott further explained the reasons for his opinion.

“It is to be borne in mind that in earlier times there was a difficulty in the case of savings accounts in that generally interest began only at certain periods and that no interest was allowed if the money was withdrawn during the periods and ordinarily the bank could insist on a time notice before withdrawal. But now interest may run from the time of deposit to the time of withdrawal without preliminary notice of withdrawal. Hence it would seem not unreasonable to require the trustee to make the income productive of income.

“In any case there do not seem to be any practical impediments today to the making of trust funds temporarily productive of income, and retaining their liquidity, whether by deposits in savings accounts or by creating some form of short term security funds.”

2 A. Scott, The Law of Trusts § 181, at 89-90 (3d ed. 1967 & Supp.1979) (footnote omitted). See currently 2A A. Scott & W. Fratcher, The Law of Trusts § 181, at 549 (4th ed. 1987).

MNB, in part of its argument, visualizes that the trial court required a system of individual passbook accounts, in which each passbook would constitute a sub-account for income on income within the trust accounting system and in which each entry of a debit or credit in the passbook would require the transfer of the passbook out of, and back into, the security vault. The trial court, of course, also found that a master passbook could have been used. 11 We agree with the trial court. MNB has not convincingly demonstrated that interest earned in a master passbook account could not periodically be allocated to the trusts in the proportions which each trust’s cash bore to total trust cash in the trust accounting system. 12 Admittedly, allocation would be imprecise, because the passbook account, just as the DDA, would trail trust accountings. While there may be differences, at any given time and as to specific trusts, between an allocation of gross interest and the specific tracking of each dollar, an allocation system should basically level the differences between accounts over time. From the standpoint of a trustee’s duty to act as a prudent investor, such a system is certainly preferable to paying no interest whatsoever on the DDA cash.

At oral argument MNB contended that such a system would violate the requirement of 12 C.F.R. § 9.13(b) that the investments of each trust account be “Adequately identified as the property of the relevant account.” In the DDA system, a given trust in theory loans, without interest, its cash to MNB which then uses that cash in its commercial banking operations. The amount of MNB’s repayment obligation to a given trust is reflected only in the trust accounting. We see no reason why MNB’s obligation, under a master passbook system, periodically to apportion the gross interest earned could not similarly be reflected by the trust accounting records and by proportions derived therefrom. 13

By utilizing in the remedy for the later years of the Class Period rates of return paid by a money market mutual fund, the circuit court was simply applying the principle that the conditions and circumstances of the particular case govern the duty prudently to invest. See Carey v. Safe Deposit & Trust Co., 168 Md. 501, 514-21, 178 A. 242, 247-50 (1935). One of the factors which was made explicit in 12 C.F.R. § 9.10(a) by the 1982 amendment, but which is implicit in the common law prudent investor rule, is the “anticipated return that could be obtained while the cash remains uninvested or undistributed____” The returns available in the short-term money market, as reflected in the mutual fund dividend rates that were utilized in the judgment, were 7.5% for 1978, 11.2% for 1979, 13.4% for 1980, 17.4% for 1981, and 12.9% for 1982, with the last converted to 7.52% to represent the seven-twelfths of that year included in the Class Period.

If we interpret the circuit court’s holding to be that cash in the DDA should have been invested in a money market fund during the later years of the Class Period, then the analysis is similar to that involving use of a master passbook. The difference is that a DDA would have to be used to clear deposits of checks received before the cash was invested in the money market fund. Although some increased cost to MNB would be involved in the added step, the dramatic rise in the return in the short-term money market during those years compels having that cash invested.

In trying the liability aspects of this case MNB drew a line at its cash management policies, maintained that any greater realization of income for the beneficiaries would have been economically unfeasible, and consequently presented no evidence supporting a fallback position. Determining the cost of business operations is never a science and is often elusive. MNB’s evidence in general was that keeping trust cash relatively fully invested would cost, conservatively, approximately $80 per account per year. This evidence was intuitive, with the exception of an internal time and motion study done in 1982 which arrived at a unit cost of $80.15 per year for passbook servicing with a minimum number of transactions. Another 1982 MNB study of uninvested cash concluded that $82.50 per year would be the cost of operating computerized principal and income cash sweeping for trust accounts which required principal and income cash to be separately stated. Under the prodding of the OCC, MNB adopted those computerized programs without a special charge to the trusts.

Any change from MNB’s cash management policies to a policy of fully investing trust cash did not involve simply some additional accounting expense. A change to full investing in some fashion would have eliminated MNB’s use of the DDA cash as a commodity at no cost to MNB. The circuit court did not err when it concluded that MNB had the ability, without computerized cash sweeping, to invest the trust cash, but that MNB “chose not to.”

Ill

In their cross-appeal the Plaintiffs assert that the trial court should have awarded the “profit” realized during the Class Period by MNB on the uninvested trust cash. Plaintiffs’ analysis is that MNB used trust assets in its business, for the personal gain of the trustee, so that the trustee must disgorge those profits. The principle relied upon by the Plaintiffs does not apply, because, under Maryland legal theory, the funds employed by MNB in its commercial banking operations are not trust assets, as such. The trust asset is the receivable from MNB, as banker, to the trust. The trust, in theory, makes a loan to MNB as banker. That those loans were without interest implicates the prudent investor rule discussed in Part II, supra. But, under Maryland law dealing with the unique situation of the self-depositing bank trustee, the loan from trust to bank is not prohibited self-dealing and is not a per se breach of trust.

The foregoing is the rule established in Real Estate Trust Co. v. Union Trust Co., 102 Md. 41, 61 A. 228. We there said that if the trustee bank “could take money from others on deposit, it surely could open an account with itself when acting as trustee and could lawfully take and hold in its capacity as banker, money in its hands in its capacity as trustee. There was nothing irregular in its doing so.” Id. at 54, 61 A. at 233. We said that the funds in the bank “were to the credit of the trustee and were not used by the trustee. The trustee as such made no profit out of them.” Id. at 55, 61 A. at 233. We held that the bank trustee could “keep its trust accounts in its own bank without running the risk of paying as a penalty therefor the highest legal rate of interest.” Id. Consequently, the trustee bank was obliged to pay only the two percent per annum interest with which it had charged itself in the court accounting, and not the highest legal rate of six percent.

In states which exclude from the prohibition against self-dealing the self-depositing of trust funds by a bank fiduciary, courts reject the argument that the fiduciary is liable for profits realized on the use of the funds in the banking operations. See Van de Kamp, 204 Cal.App.3d at 840, 251 Cal.Rptr. at 537 (statute authorizing self-depositing); Estate of Smith, 112 Cal.App. 680, 297 P. 927 (1931); Hayward v. Plant, 98 Conn. 374, 119 A. 341 (1923); In re People’s Trust, 169 A.D. 699, 155 N.Y.S. 639 (1919); Stahl v. First Pennsylvania Banking & Trust Co., 411 Pa. 121, 191 A.2d 386 (1963); Moore’s Estate (No. 3), 211 Pa. 348, 60 A. 991 (1905); Mills v. Swearingen, 67 Tex. 269, 3 S.W. 268 (1887). 14 The weight of authority at common law sanctioned the self-depositing practice. See G.G. Borgert & G.T. Bogert, The Law of Trusts and Trustees § 598, at 488 (rev. 2d ed 1980) (Bogert).

Although Restatement of Trusts § 170, comment m (1935) and Restatement (Second) of Trusts § 170, comment m (1959) have espoused a rule contrary to that adopted in Real Estate Trust, neither the General Assembly of Maryland nor this Court has ever changed the Maryland rule.

Accordingly, we shall affirm the circuit court on the cross-appeal.

IV

MNB also attacks each of the components used by the trial court in constructing the monetary relief. We have already explained in Part II, supra, the appropriateness, within the trial court’s discretion, of using as a factor 100% of the stipulated annual DDA average daily balances in approximating the loss to the income beneficiaries for each of the years in the Class Period. MNB does not question that some prejudgment interest may be allowed on the principal amount awarded for each year of the Class Period, but MNB does question the compounding of prejudgment interest, as well as the use of ten percent per annum as the interest rate. The trustee further claims that the trial court erred in reducing by ten percent its trustee’s commissions.

The trial court based the compounding of prejudgment interest on a well established principle, but it does not apply to this case. The principle is stated in dicta in Riggs v. Loweree, 189 Md. 437, 56 A.2d 152 (1947), cited by the trial court. Riggs involved an unsuccessful effort to charge a testamentary trustee with the loss of trust funds invested in common stock. We said:

“If the trustee is chargeable with interest, he is charged with simple interest, not compound interest, unless (1) he has received compound interest, or (2) it was his duty to accumulate the income, or (3) he has received a profit which cannot be ascertained, but is presumably at least equal to compound interest. 1 Restatement, Trusts, sec. 207. It is reasoned that when a trustee uses trust funds in connection with his own business, and he is unable or unwilling to show what profit, if any, he has made by the use of the funds, the court should impose compound interest as a substitute for the profit he is presumed to have made by their use.”

Id. at 445, 56 A.2d at 157.

The circuit court also cited Brown v. Tydings, 149 Md. 22, 130 A. 337 (1925), where the rule was applied to an administrator pendente lite who from time to time borrowed cash from the estate, paying the estate interest ranging from three percent to four percent. These rates were comparable to those paid by savings banks on deposits. We held that the fiduciary was chargeable with compound interest at six percent. In Brown v. Tydings it could not be determined how the administrator had used the estate funds. Citing to Ringgold v. Ringgold, 1 H. & G. 11 (1826), we said that “[t]he allowance of compound interest is ... a substitute for the appropriation of actual profits made, or presumed to have been made, by the use of the trust funds, when lack of information from the fiduciary prevents determination of the amount to be appropriated as profits.” Brown, 149 Md. at 28, 130 A. at 339.

Bogert concludes that “[p]erhaps the most common instance of the collection of compound interest from the defaulting trustee is found where he has used the trust fund in his own business and the actual profits earned by the trust fund are not claimed or are impossible of computa tion____” Bogert, § 863, at 56 (footnotes omitted). Scott reaches the same conclusion. “Ordinarily, [the trustee] is liable only for simple interest.” 2A A. Scott & W. Fratcher, The Law of Trusts § 207.2, at 263 (4th ed. 1988) (footnote omitted). But if the trustee “uses trust funds in his own business and it does not appear what return he has received, it has been held in a number of cases that he is chargeable with compound interest____” Id. at 264. And see Restatement (Second) of Trusts § 207(2) (1959); Perley on Interest, at 77 (1893).

The rule that compound interest may be awarded against a trustee who has used trust funds in the trustee’s personal business, as illustrated by the holding in Brown v. Tydings, supra, is inconsistent with the rule applied to the self-depositing bank trustee in Real Estate Trust Co. v. Union Trust Co., 102 Md. at 55, 61 A. at 233. There we said that the trust funds “were not used by the trustee” and that “[t]he trustee as such made no profit out of” the trust funds. Compounding of interest, as a substitute for actual profits realized, is not permitted against the self-depositing bank trustee because, under the special rule of Real Estate Trust, the trustee is not considered to have used trust funds for its own purposes. As we have seen, “profits” realized by MNB on the DDA were not recoverable. See Part III, supra.

Where the trustee, in breach of trust, has allowed the principal to remain idle, the trustee has been charged with simple interest. See Glenn v. Cockey, 16 Md. 446, 455 (1860); Ringgold v. Ringgold, 1 H. & G. 11, 80 (1826); Dame v. Catlett, 6 H. & J. 475, 482 (1823). Thus, the prejudgment interest in this case should be simple interest.

We agree with MNB that the trial court erred in using a rate of prejudgment interest of ten percent per annum, which the trial court described as “the legal rate.” There is no statute which fixes the rate of prejudgment interest for this type of claim. Absent a contractual stipulation or a statute, the rate of prejudgment interest may not exceed the general legal rate of six percent. See, after July 1, 1975, Md.Code (1975, 1990 Repl.Vol.), § 12-102 of the Commercial Law Article and, during the Class Period prior to that date, Md.Code (1957, 1972 Repl.Vol.), Art. 49, § 3. We have so held in First Virginia Bank v. Settles, 322 Md. 555, 588 A.2d 803 (1991). This holding in First Virginia Bank was accurately predicted in Federal Savings & Loan Ins. Corp. v. Quality Inns, Inc., 876 F.2d 353 (4th Cir.1989), rev’g 674 F.Supp. 522 (D.Md.1987) and in Fox v. Kane-Miller Corp., 398 F.Supp. 609 (D.Md.1975), aff'd on other grounds, 542 F.2d 915 (4th Cir.1976).

The circuit court ordered that MNB refund ten percent of the commissions earned on personal trusts during the Class Period. Whether, and to what extent, a trustee should be ordered to refund commissions is within the discretion of the trial court “if a trustee has been guilty of a serious breach of trust.” Stone v. Stone, 230 Md. 248, 255, 186 A.2d 590, 595 (1962). And see Cosden v. Mercantile-Safe Deposit & Trust Co., 41 Md.App. 519, 547, 398 A.2d 460, 476, cert. denied, 285 Md. 728 (1979), cert. denied, 444 U.S. 941, 100 S.Ct. 295, 62 L.Ed.2d 308 (1979); Madden v. Mercantile-Safe Deposit & Trust Co., 27 Md.App. 17, 51, 339 A.2d 340, 359-60 (1975). MNB argues that the violation is not sufficiently serious to warrant any reduction of commissions. In the aggregate, and over the nearly ten years of the Class Period, this case involves a considerable amount of lost interest to the Plaintiffs’ class. In view of the public policy permitting separate claims of proposed members of a class to be aggregated to meet the minimum amount in controversy required for circuit court jurisdiction, see Md.Code (1974, 1989 Repl.Vol.), § 4-402(d)(l)(ii) of the Courts and Judicial Proceedings Article, the degree of seriousness here may be measured by the aggregate effect of the breach.

MNB also makes an argument based upon the evolution of the provision in the trial court’s judgment dealing with commissions. In the principal opinion of November 21, 1989, the court ordered a twelve percent return of commissions, followed by a parenthetical reference to MNB’s response to Plaintiffs’ proposed conclusions of law. Plaintiffs had sought a twenty percent reduction in commissions. In response MNB had argued that the ratio of the DDA average daily balances to total personal trust assets managed by MNB was approximately twelve hundredths of one percent (.12%). When the November 21, 1989, opinion was circulated MNB took the position that the twelve percent figure in the opinion should be .12%. The circuit court thereupon revised the opinion to reduce commissions by ten percent. It is clear that in the judgment appealed from the intended reduction is ten percent.

In Diffenderfer v. Winder, 3 G. & J. 311 (1831), the trustee had not made any distribution for thirteen years, despite having received large sums of income. He “continually employed [the income] in trade and speculation” for his own account. The Court of Chancery charged the trustee with the income received from year to year, together with compounded interest, but had allowed the trustee full commissions. This Court affirmed the relief granted and, in addition, reduced the commissions by fifty percent. Here, although there was no prohibited self-dealing, there was a breach of trust which caused serious loss. A ten percent reduction of commissions was not an abuse of discretion.

V

In its brief in this Court MNB argues that the claim of the Plaintiffs is barred by laches for most of the Class Period inasmuch as periodic statements furnished by MNB to the class members presented the effects of the cash management policy. The trial court caused the parties to file proposed findings of fact and conclusions of law after all of the evidence had been received. In its proposals MNB did not incorporate any request that the court find any portion of the claim barred by laches, although the defense had been raised in MNB’s answer. The circuit court therefore did not address at all the issue of possible laches, and the issue is not before us on this appeal. See Maryland Rule 8-131(a). 15

VI

For the foregoing reasons the judgment is affirmed in part and reversed in part, and the case is remanded to recompute the judgment by allowing only simple prejudgment interest at the rate of six percent per annum on the respective annual amounts representing lost income to the Plaintiffs’ class for the particular years. The prejudgment interest “shall be separately stated in the ... decision and included in the judgment.” Maryland Rule 2-604(a).

JUDGMENT OF THE CIRCUIT COURT FOR BALTIMORE CITY AFFIRMED IN PART AND REVERSED IN PART. CASE REMANDED TO THAT COURT FOR FURTHER PROCEEDINGS CONSISTENT WITH THIS OPINION. SEVENTY-FIVE PERCENT OF THE COSTS TO BE PAID BY THE APPELLANT AND CROSS-APPEL-. LEE, MARYLAND NATIONAL BANK, AND TWENTY-FIVE PERCENT OF THE COSTS TO BE PAID BY THE APPELLEES AND CROSS-APPELLANTS, CAROL H. CUMMINS et al.

1 . Special circumstances applicable to a particular trust at a given time might require larger accumulations of uninvested cash.

2 . During the Class Period, MNB had six CIFs, i.e., beginning in 1972 a long-term corporate bond fund, a municipal bond fund, and a balanced equity fund, beginning in 1974 an aggressive growth equity fund, in 1976 a short-term investment fund, and in 1978 an intermediate-term corporate bond fund.

3 . Operationally there were three separate accounts comprising the DDA. One of these accounts was used to clear checks representing cash receipts and seems to have been the general repository of uninvested cash. The other two accounts were used for disbursements. One of these two reflected automated disbursements and the other manual disbursements. The purpose of three accounts was apparently to facilitate reconciliation of disbursements to a zero balance and the tracking of outstanding checks. See Comptroller's Handbook for National Trust Examiners § 301.1, at 2 "Trust Cash" (Jan.1986).

4 . With respect to principal cash the float favorable to the bank arises when securities are not delivered by the settlement date, inasmuch as the bank will not pay for securities until delivery. This has been called the "fail float,” see Van de Kamp v. Bank of America Nat’l Trust & Savings Ass'n, 204 Cal.App.3d 819, 832, 251 Cal.Rptr. 530, 532 (1988), or the “lag period float,” see S. Levmore, Bank Trust Departments and “Float” Revenue: Finding the Proper Procedures, 98 Banking L.J. 817, 820 (1981).

5 . The rationale of these cases is discussed in Part III.

6 . At some point in a year principal cash would be needed to pay trustee’s commissions. Trustee’s commissions on the corpus, as well as the income, of personal trusts, have been regulated throughout the Class Period. Accounting from July 1, 1981, whether or not the trust was in existence at that time, commissions on corpus, absent agreement or court order, are .4% on the first $250,000, .25% on the next $250,000, .15% on the next $500,000, and .1% upon any excess. See Md.Code (1974, 1990 Cum.Supp.), § 14-103(c) of the Estates and Trusts Article.

7 . The stipulated average daily balances of the DDA for the years of the Class Period are set forth below. The amounts represent principal and income cash.

For the Year The Average Class Checking Daily Balance

1982 $1,096,479

1981 1,018,656

1980 872,704

1979 510,529

For the Year The Average Class Checking Daily Balance

1978 427,285

1977 593,458

1976 551,767

1975 510,075

1974 468,383

1973 426,692

1972 385,000

8 . The Hunt Commission Report is the Report of the President’s Commission on Financial Structure and Regulation (1971).

9 . The program implemented in 1976 by the bank in Van de Kamp invested principal cash in excess of $100. In 1982 all cash was daily invested to the penny. The trial court found these practices to be reasonable, and the intermediate appellate court affirmed.

10 . The end of the Class Period, June 25, 1982, is the effective date of an amendment (47 FR 27831) to 12 C.F.R. § 9.10 dealing with “[f]unds held by a national bank in a fiduciary capacity which are awaiting investment or distribution____’’ The amendment effectively required income cash to be invested. MNB took the position that it would invest income cash in increments of $1,000 for a $100 per year fee. The OCC did not consider that arrangement to be in compliance with the amended regulation.

11 . The master passbook account could be used to clear checks received. Amicus, American Bankers’ Association, points out that a DDA would still be required, in addition to the passbook account, for disbursements, inasmuch as NOW accounts were not available during much of the Class Period. It appears, however, that MNB used a disbursing DDA under the cash management system challenged here. MNB operated separate accounts (part of what we are calling the DDA) for checks written by computer order and for those written by manual order.

12 . Below is the core of MNB’s attack on the trial court’s master passbook alternative to the trustee’s keeping all of the income earned on the DDA.

"The basic misconception of the court below was to think it reasonably possible to continuously ‘allocate’ the cash in the DDA to the individual trusts for the purpose of calculating interest on the cash of each trust actually in the DDA for as long as the cash stayed there. Both the plaintiffs expert and the trial court erroneously assumed that the DDA funds belonged to the trust department and the beneficiaries, when in fact the DDA balances were not trust funds but constituted a secured debt to the trusts assuring the payment of the disbursements required by MNB’s trust accounting system. Again and again, the court, assuming that the DDA cash balances belonged to the individual trusts, reiterated that the trust accounts at some point had to be debited or credited with what is put into or taken from the DDA and insisted that each trust account could be given a proportional credit for each receipt deposited in the DDA. The court below consistently failed to grasp the essential fact that payments from the DDA to the trusts were not allocations at all. The trust accounts and the DDA were independent accounting systems and never reflected the same totals at the same time.”

Appellant’s Brief at 23 (footnotes and record references omitted).

13 . The ultimate reason given by the trustee's expert on trust department accounting for rejecting allocation of interest among the separate trusts was that it would be too disruptive of trust department procedures. The disruption which that expert foresaw was that any errors in allocation would affect every trust account’s periodic statement. The trial court was not required to accept this reason as a justification for the decision not to invest the DDA funds. Auditors typically make adjustments to correct errors made at an earlier time.

14 . Relevant to the issue in Part II of this opinion is that the banks or bankers in the cited cases paid interest on the self-deposited funds. The claim against them was for the profit made, as bank, over and above the interest paid for use of the funds employed in the commercial banking operations.

15 . Trial counsel for MNB, who was not lead counsel on this appeal, undoubtedly was conscious of the Maryland law on the laches defense. ”[E]xpress trustees usually are denied the right to plead limitations,” so that “ ‘[t]o cause the statute to begin running during the life of the trust there must be some act of repudiation of the trust by the trustee,’ ” for example, a refusal to account or appropriating principal or income. Ridgely v. Pfingstag, 188 Md. 209, 234-35, 50 A.2d 578, 590 (1946) (quoting Bogert, § 951).