Four allocation conventions applied to identical records produce materially different traceable amounts. Which one governs is a question of law, not of accounting.
Start a conversation with the Forensic Concierge, already scoped to tracing & asset recovery. Pick a starting point, or describe the matter directly.
Tracing looks like arithmetic and is not. Once misappropriated money is commingled with clean money, no accounting fact determines which dollars left the account — so a court adopts an allocation presumption instead, and four are in common use: first-in-first-out, last-in-first-out, pro rata, and the lowest intermediate balance rule. Applied to the same transaction record they produce materially different traceable amounts. The Tenth Circuit put the choice where it belongs, stating that courts exercise case-specific judgment to select the method best suited to achieve a fair and equitable result on the facts before them. United States v. Henshaw, 388 F.3d 738, 741 (10th Cir. 2004). The practitioner’s job is not to run a preferred convention. It is to justify one against the transaction record, and to know where that record stops.
Six things decide whether a dollar in an account is that dollar, and whether a court can reach it.
The precondition for every allocation rule. Without it there is direct tracing and no presumption to argue about.
FIFO, LIFO, pro rata and the lowest intermediate balance rule. Pro rata carries no timing element; the other three turn on it.
Where that rule governs, the traceable interest is capped at the lowest balance the account reached afterward, and later deposits do not restore what was spent.
A court-adopted presumption rather than a firm preference, which is why the selection has to be argued from the withdrawal record.
Conscious wrongdoer, defaulting fiduciary or innocent recipient — the axis Restatement (Third) of Restitution § 59 sorts commingling cases along. It sets the ceiling on recovery, not the tracing presumption.
Constructive trust, equitable lien and restitution over specific property — the route around the general bar on freezing assets to secure a money claim.
The work, roughly in the order it has to happen.
The trace is usually the difference between a judgment and a collection.
Once trust money is removed "it is not replenished by subsequent deposits" — In re Columbia Gas Systems Inc., 997 F.2d 1039, 1063 (3d Cir. 1993). The Fourth Circuit is blunter: where the account is depleted entirely, "the trust is considered lost." In re Dameron, 155 F.3d 718, 724 (4th Cir. 1998). Millions moving through afterward do not bring it back.
Not by the accountant’s preference. The conventions are allocation presumptions a court adopts, and Henshaw states that courts exercise case-specific judgment to select the one best suited to a fair and equitable result on the facts before them — the same opinion approved both last-in-first-out and the lowest intermediate balance rule as alternatives suited to different sources of commingled funds. That leaves the practitioner a burden of justification rather than a default: evaluate the conventions against the actual transaction record, explain why the others were set aside, and document it. Pro rata tends to prevail where many similarly situated claimants trace to one account, which is the pattern in a mature multi-victim scheme. Presenting more than one convention, and the range it produces, is a defensible posture where the record does not clearly favor one.
Not the statements. A statement shows an amount and a date; it does not show the source. The Department of Justice publishes a Subpoena for Bank Records Checklist that separates deposit tickets from the checks in deposit for exactly that reason — the deposit item is the instrument itself, and it is what establishes where the money came from. The same checklist runs to account opening documents, signature cards, wire transfers and account closure records, which is a useful standard to measure your own subpoena against. One hard limit: Suspicious Activity Reports carry an unwaivable privilege under 31 U.S.C. § 5318(g)(2), so a subpoenaed institution must decline to produce one and must not confirm one exists. The underlying transaction records remain producible.
Considerably less far than a prosecutor, and the gap is structural. Mutual legal assistance treaties run government to government and are unavailable to private civil litigants. So is 31 U.S.C. § 5318(k), which lets US authorities subpoena a foreign bank holding a US correspondent account for records kept abroad, with penalties up to $50,000 per day of non-compliance. A private party is left with letters rogatory, which rest on comity and are slow, and with 28 U.S.C. § 1782 — narrowed by ZF Automotive US, Inc. v. Luxshare, Ltd. (2022) to governmental or intergovernmental adjudicative bodies. In England, Norwich Pharmacal orders compel a third party to identify a wrongdoer and Bankers Trust orders target bank disclosure to trace misappropriated funds.
Neither as easily as vendors imply nor as little as defense arguments suggest, and the honest answer is about method. Blockchain tracing is litigated under Daubert and Rule 702. In United States v. Sterlingov, the Bitcoin Fog prosecution in the District of Columbia, the court held after a hearing that expert testimony resting on Chainalysis Reactor was reliable enough to admit, rejecting the argument that the clustering was an untestable black box — one district court, on one record, and not a holding that blockchain analytics is categorically admissible. The mechanics matter on cross. CoinJoin is the known counter to co-spend clustering, though the government’s witness in that record testified the tool has controls to detect and skip CoinJoin co-spends — and part of the attribution there came from off-chain intelligence the court itself noted "is not actually a heuristic at all."
Describe the accounts, the rough dates and what has already been subpoenaed. The Institute will help you see what a trace of this kind requires, and where it usually stops.