Why does tracing a commingled account need a rule at all?
Because once dollars are mixed, no fact in the record identifies which ones left the account. A statement showing a deposit of one party’s funds followed by a series of withdrawals is equally consistent with every allocation an advocate might want, so the convention is supplied by law rather than found in the ledger. In In re Columbia Gas Systems Inc., 997 F.2d 1039, 1063 (3d Cir. 1993), the Third Circuit called the lowest intermediate balance rule “a legal construct” that “allows trust beneficiaries to assume that trust funds are withdrawn last from a commingled account.”
The reason the presumption runs in the claimant’s favor is older than the bank account. It descends from the English decision the United States Supreme Court discussed in Cunningham v. Brown, 265 U.S. 1 (1924) as Knatchbull v. Hallett, L.R. 13 Ch. D. 696, in which, the Court wrote, “it was decided … that, where a fund was composed partly of a defrauded claimant’s money and partly of that of the wrongdoer, it would be presumed that in the fluctuations of the fund it was the wrongdoer’s purpose to draw out the money he could legally and honestly use rather than that of the claimant.” The Second Circuit described the same idea functionally in United States v. Banco Cafetero Panama, 797 F.2d 1154, 1159 (2d Cir. 1986).
The tracing convention is therefore a legal question a court decides, not a preference an accountant exercises. The forensic accountant builds the transaction record the convention operates on, computes what each convention produces, and documents why one was advanced and the others set aside. What a witness may then say about that work is a separate question, covered in Is tracing testimony expert or lay testimony? and in Investigative Testimony.
Where does the lowest intermediate balance rule come from?
From Schuyler v. Littlefield, 232 U.S. 707 (1914). Justice Lamar, writing for the Court on 23 March 1914, framed the case as “an application of the rule that where one has deposited trust funds in his individual bank account, and the mingled fund is at any time wholly depleted, the trust fund is thereby dissipated, and cannot be treated as reappearing in sums subsequently deposited to the credit of the same account.” That is the hard half of the doctrine, stated as settled law more than a century ago.
Schuyler also supplies the evidentiary point that decides most modern disputes. The claimants there argued in the alternative that their funds could be followed out of the account and into collateral that reached the trustee in bankruptcy, and the Court refused: they were “under the burden of proving their title,” and “[i]f their evidence left the matter of identification in doubt, the doubt must be resolved in favor of the trustee, who represents all of the creditors.” Tracing failures are resolved against the claimant, not against the estate.
One attribution should be handled with care, because law firm publications and trade press repeat it. Cunningham v. Brown, 265 U.S. 1 (1924) — the Ponzi case — does not announce the lowest intermediate balance rule, and the phrase does not appear in it. Chief Justice Taft’s opinion treats the payments as unlawful preferences recoverable by the trustee, declines to extend the Knatchbull v. Hallett presumption where “the fund with which the wrongdoer is dealing is wholly made up of the fruits of the frauds perpetrated against a myriad of victims,” and rests on the principle “that equality is equity, and this is the spirit of the bankrupt law.” The Fourth Circuit does cite Cunningham alongside Schuyler when stating the rule in In re Dameron, 155 F.3d 718, 723–24 (4th Cir. 1998), which is how the misattribution travels, but the doctrinal root to cite is Schuyler.
If more money is deposited later, does the traceable claim come back?
As a general rule, no. The Fourth Circuit put it without qualification in In re Dameron, 155 F.3d 718, 724 (4th Cir. 1998): “In no case is the trust permitted to be replenished by deposits made subsequent to the lowest intermediate balance.” The same opinion sets out the three outcomes in order — a fund that “has at all times equaled or exceeded the amount of the trust” returns the trust funds in full; a fund “depleted entirely” means “the trust is considered lost”; and a fund “reduced below the level of the trust fund but not depleted” entitles the claimant to the lowest intermediate balance.
Dameron is also the clearest published illustration of the arithmetic. A settlement agent commingled lender funds in a general corporate account. One lender’s deposit of $71,475.40 was followed within four days by unrelated withdrawals that created a deficit in the account of $5,313.05, and the Fourth Circuit affirmed a lowest intermediate balance of $66,162.35 for that lender, while later lenders whose funds saw no subsequent withdrawals kept their full amounts. A four-day dip of roughly five thousand dollars permanently reduced one claim and left the others untouched.
There is a narrow and contested exception worth knowing about rather than relying on. The dissent in Columbia Gas pointed to Restatement (Second) of Trusts § 202 comment m (1959), under which a trustee who makes later deposits “manifesting an intention to make restitution of the trust funds withdrawn” may leave the beneficiary’s lien not limited to the lowest intermediate balance. It was not the majority’s position, and no matter should be planned on the assumption that a later deposit will be credited.
What are the other tracing conventions, and which courts use which?
The Second Circuit set out three basic approaches in United States v. Banco Cafetero Panama, 797 F.2d 1154, 1159 (2d Cir. 1986). The first it called a “drugs-in, last-out” rule, which, the court said, is “more properly called the ‘lowest intermediate balance’ rule.” The second is a pro rata or “averaging” rule, under which each withdrawal carries a share of the tainted deposit fixed by the ratio of that deposit to the funds in the account immediately after it. The third is a “drugs-in, first-out” rule, under which any one withdrawal is treated as carrying the tainted funds up to their amount. In that forfeiture case the government claimed the benefit of the first or the third at its option, and the court concluded that its position was “sufficiently correct to enable it to prevail on this appeal,” while adding that “[w]hich approach reflects reality in any particular case will depend on the precise circumstances.”
Last-in-first-out is the fourth, and it has been approved on its own facts. In United States v. Henshaw, 388 F.3d 738, 741 (10th Cir. 2004), the Tenth Circuit affirmed a district court’s use of last-in-first-out to tie an attorney’s fee to an immediately preceding deposit, and rejected the argument that doing so was inconsistent with the availability of the lowest intermediate balance method. Its sentence on method selection is the one to carry: “There are several alternative methods, none of which is optimal for all commingling cases; courts exercise case-specific judgment to select the method best suited to achieve a fair and equitable result on the facts before them.”
First-in-first-out is the rule of Clayton’s Case, which the United States Supreme Court cited as Clayton’s Case, 1 Merivale 572 (Ch. 1816) and described as holding “that, in a fund in which were mingled the moneys of several defrauded claimants insufficient to satisfy them all, the first withdrawals were to be charged against the first deposits, and the claimants were entitled to be paid in the inverse order in which their moneys went into the account.” That same Court refused to apply it to a fund built out of many victims’ payments: “The rule in Clayton’s Case has no application.” Cunningham v. Brown, 265 U.S. 1 (1924). Priority-of-time ordering does have a settled American home in the appropriation of payments on a running account, where United States v. Kirkpatrick, 22 U.S. (9 Wheat.) 720 (1824) held that in “cases like the present, of long and running accounts,” payments “ought to be applied to extinguish the debts according to the priority of time.” No United States jurisdiction was confirmed for this piece as applying Clayton’s Case to trust tracing. The courts confirmed here as applying the lowest intermediate balance rule are the Third Circuit in Columbia Gas and the Fourth Circuit in Dameron; Columbia Gas collects the First Circuit’s Connecticut General Life Insurance Co. v. Universal Insurance Co., 838 F.2d 612, 619 (1st Cir. 1988) and the Sixth Circuit’s First Federal of Michigan v. Barrow, 878 F.2d 912, 916 (6th Cir. 1989) as having embraced it, and the Second Circuit described it in Banco Cafetero.
How do bankruptcy courts apply the rule to constructive trusts and property of the estate?
Through 11 U.S.C. § 541, which sets the boundary of the estate. Section 541(a)(1) sweeps in “all legal or equitable interests of the debtor in property as of the commencement of the case,” and § 541(d) excludes property in which the debtor holds “only legal title and not an equitable interest.” The Fourth Circuit explained that exclusion in Dameron, quoting Begier v. IRS, 496 U.S. 53, 59 (1990): when a “debtor does not own an equitable interest in property he holds in trust for another, that interest is not ‘property of the estate.’” Tracing converts a claim of that kind from an assertion into an identifiable interest, and the lowest intermediate balance rule fixes its size.
Whether the trust is express or constructive changes the analysis substantially, and the circuits do not agree. In Dameron the Fourth Circuit found an express trust from the closing instructions and so had no occasion to reach constructive trust. The Sixth Circuit took a restrictive view in In re Omegas Group, Inc., 16 F.3d 1443 (6th Cir. 1994), holding that “a constructive trust is not really a trust” but “a legal fiction, a common-law remedy in equity that may only exist by the grace of judicial action,” and that a creditor’s claim of entitlement to one “is not an ‘equitable interest’ in the debtor’s estate existing prepetition, excluded from the estate under Sec. 541(d).”
Omegas reserved property already impressed with a constructive trust prepetition by a court in a separate proceeding, and it reached its position in open disagreement with the Fifth Circuit’s In re Quality Holstein Leasing, 752 F.2d 1009 (5th Cir. 1985). Where the debtor’s receipt of the funds is documented as an escrow or agency arrangement, the argument runs on express trust and the tracing convention decides the amount; where it depends on a court imposing a constructive trust after the petition, the forum may decide the claim before any arithmetic is reached. That is the same equitable-claim-to-specific-property question discussed in Can we freeze the accounts before we have a judgment? and in Tracing & Asset Recovery.
What does a lowest intermediate balance analysis establish, and what does it not?
It establishes a ceiling on an identifiable claim to money in a particular account, and nothing about anyone’s conduct or state of mind. The output is a number tied to a named account over a stated period, built from statements, items and transfers, together with the reason the convention applied was the appropriate one on those records. It is not a finding that funds were taken, that any person acted wrongfully, or that fraud occurred. American Institute of Certified Public Accountants Statement on Standards for Forensic Services No. 1 provides at paragraph 10 that “[t]he ultimate decision regarding the occurrence of fraud is determined by a trier of fact; therefore, a member performing forensic services is prohibited from opining regarding the ultimate conclusion of fraud.”
It is also not a measure of loss. What was lost, what it is worth, and what the claimant would have had in a world where the transactions had not occurred are damages questions, and they belong to our Economic Damages Institute rather than here. The two halves fit together the way the rules of evidence expect: the reconstructed factual record is the predicate the damages analysis relies on. A matter needing both is usually two engagements, and often two experts.
The planning point that follows is about sequence. The cap falls with the account’s low point and, under Dameron, is not replenished by later deposits, so the analysis is worth more the earlier the records are pulled — and 31 C.F.R. § 1010.430(d), which requires records kept under the Bank Secrecy Act chapter to be retained “for a period of five years,” limits how far back the primary records can be assumed to exist. Related reading: How do you prove theft of something that was never recorded? and How much stolen money is actually recoverable?