Auditing Forensic Accounting Reports: Tracing Holes in Asset Valuation Models

An asset-tracing chart is not a picture of the money. It is a summary of voluminous records, and under Federal Rule of Evidence 1006 it stands in for the underlying bank statements, wire confirmations, and ledgers only if those records were produced and are themselves admissible. The clean box-and-arrow diagram the jury sees is the last link in a chain that starts with raw transaction data and passes through reconciliation, tracing-method selection, fee treatment, and currency conversion. Every one of those steps is a place the number can drift.

Under the 2023 amendment to Federal Rule of Evidence 702, the proponent must show by a preponderance of the evidence that the expert's method was reliably applied to the facts of this case, not merely that the method is valid in the abstract. Daubert v. Merrell Dow Pharmaceuticals frames the reliability inquiry; Frye governs in general-acceptance jurisdictions. The audit target for an attorney or adjuster is the seam between the underlying records and the demonstrative: does each node on the chart tie, penny for penny, to a specific statement line and transaction reference, or does the visualization smooth over gaps the workpapers cannot close.

• Ask first: were the underlying records produced, and are they admissible on their own footing under Rule 1006.
• Then ask: does the summary add anything the records do not contain, such as inferred causation, color-coded intent, or a rounded figure that the ledger does not support.

A tracing model is only as sound as the reconciliation feeding it. Reconciliation matches the account's internal ledger to the bank's record and resolves the differences. Three categories of difference are where integrity leaks:

1. Timing differences. Deposits in transit and outstanding transfers post on different dates in the ledger and the statement. A tracing chart that treats the ledger date as the settlement date can show funds arriving before they cleared, breaking the chronological logic the whole exhibit depends on.
2. Unmatched items. A debit with no corresponding credit, or a credit with no source, is an open item. If the expert closes it with an assumption rather than a document, that assumption is now load-bearing and undisclosed.
3. Adjusting and plug entries. When the reconciliation will not balance, a plug entry forces it. A plug is an admission that the records do not fully explain the flow. Any node downstream of a plug inherits that uncertainty.

The mechanism to probe is simple to state and hard to fake: for each arrow on the chart, there should be a reconciliation workpaper that connects a named debit on one statement to a named credit on another, by amount, date, and reference. Where that linkage runs through an unmatched item or a plug, the chain is inferential, not documentary.

Once funds pass through a commingled account, dollars lose their identity. To keep tracing, the expert must apply a tracing convention, and the convention chosen materially changes which later withdrawals are treated as the traced funds. The main conventions are established doctrine, not opinion:

• The rule in Clayton's Case (first in, first out). The first money into a running account is treated as the first money out. Applied mechanically across a shell layer, it can attribute early legitimate deposits to later suspect withdrawals, or the reverse, purely by ordering.
• The lowest intermediate balance rule. Traced funds cannot exceed the lowest balance the account fell to between deposit and withdrawal. If the account was drained to near zero at any point, the claim on later funds is capped there, regardless of how large the later balance grew.
• Pro rata or pari passu allocation. Commingled contributors share withdrawals proportionally rather than by sequence.

The audit point: the report must state which convention was used at each commingled node and why. A forensic accountant who applies first in, first out at one layer and lowest intermediate balance at another, without disclosing the switch, has embedded a choice that flatters the conclusion. In a multi-layered shell structure, ask the expert to re-run the trace under an alternative convention. If the traced amount moves, the convention, not the evidence, is doing the work.

Money does not move intact across institutions. Cross-border and correspondent-bank transfers deduct value in flight, and a tracing chart that shows the same figure entering and leaving a hop is describing a transaction that did not happen. The mechanism of leakage:

• Correspondent and intermediary bank fees. A wire routed through one or more correspondent banks can have a lifting fee deducted at each hop, so the credit received is smaller than the debit sent.
• Wire and SWIFT charges. Sending and receiving institutions each levy fees, sometimes debited separately from the principal, sometimes netted out of it.
• Nostro and vostro settlement spreads. Value moving between an institution's correspondent accounts can carry deductions that never appear on the originator's statement.

The tell is a chart where large sums move between accounts as identical, unbroken figures. Real transfers rarely tie to the penny across institutions. The audit action is to place the debit-side and credit-side records side by side for each hop and require the expert to account for the difference. A difference explained by a documented fee strengthens the trace. A difference the expert cannot source, or one they silently absorbed to make the arrow balance, is a hole. So is the opposite error: an expert who ignores fees and treats gross amounts as net overstates the traced sum at every hop it accumulates.

Any cross-currency step forces a conversion, and conversion is never a single objective number. Margin enters through three levers, each of which the expert has chosen:

• Rate source. A mid-market reference rate, a central-bank fixing, and the rate the bank actually applied to the transaction can all differ. Using a reference rate when the account was charged a worse commercial rate misstates the amount that actually moved.
• Value date. The exchange rate on the instruction date, the value date, and the posting date differ. Picking the most favorable date shifts the converted figure.
• Spot versus contracted rate. If the parties used a forward or a contracted rate, the market spot rate is the wrong input entirely.

Across a multi-hop trace that converts and reconverts, small per-step margin compounds. An expert who converts everything to a single presentation currency at one blended rate can hide directional bias in that choice. The audit action is to require the rate source, rate type, and value date for every conversion, confirm they are applied consistently across all steps rather than cherry-picked per step, and check whether the transaction records show the actual rate the institution used. Where the actual rate is available, a modeled rate that departs from it needs a stated reason.

The visualization is where analysis becomes persuasion, and where an accurate trace can still become an inadmissible or impeachable exhibit. A Rule 1006 summary must reflect the underlying records without adding argument. Common failures that turn a defensible trace into a cross-examination opening:

• Editorializing design. Red arrows, labels like "diverted" or "concealed," and directional emphasis assert conclusions the records do not state. That is advocacy embedded in a summary.
• Aggregation that hides variance. Collapsing many transactions into one bold arrow can bury the timing differences, fees, and rate margins discussed above. The jury sees certainty the workpapers do not support.
• Untraceable nodes. Every box and arrow should map to a specific, produced record. A node that exists only in the model, with no statement behind it, is an assumption drawn as a fact.

The clean audit standard: a reviewer should be able to take any element of the chart and walk it back to a bank statement line, a transaction reference, a stated tracing convention, a documented fee, and a disclosed conversion rate. If any of those five is missing for any element, that element is where cross-examination starts.

The audit produces the questions. Each mechanism above maps to a line of examination that does not require the attorney to out-accountant the accountant, only to expose the choice behind the number.

• On reconciliation: establish that unmatched items and plug entries exist, then establish that specific nodes on the exhibit sit downstream of them.
• On tracing method: get the convention on the record, then ask whether the traced amount changes under an accepted alternative. Undisclosed method-switching between layers is a credibility point as much as a technical one.
• On fees: put the debit-side and credit-side records together and ask the expert to reconcile the difference from the documents.
• On currency: pin the rate source, type, and date, and ask whether the actual transaction rate was available and used.
• On the exhibit itself: separate what the records show from what the design asserts.

None of this guarantees exclusion or a favorable ruling, and it is not a substitute for legal judgment on your matter. It is a procurement and preparation framework: a way to pressure-test a forensic accounting report before it reaches a jury, and to brief the retained expert on the standard their asset-tracing exhibit will have to survive.