Loss Calculation in Federal Fraud Sentencing

By Elizabeth Franklin-Best, Esq., and Christopher Zoukis, JD, MBA

In a federal fraud case, one number can matter more than almost any other: the loss amount. Under the Sentencing Guidelines, the dollar figure attached to a fraud offense can add anywhere from a handful of offense levels to as many as thirty. It is, in many financial cases, the single most consequential determination at sentencing — and it is far more contestable than defendants often realize.

At Elizabeth Franklin-Best, P.C., we treat the loss figure as an argument to be tested, not a number to be accepted. Elizabeth Franklin-Best, our principal attorney, is named a Best Lawyers in America 2026 “Best Lawyer” in Appellate Practice and holds a Chambers USA 2026 ranking for Litigation: White-Collar Crime & Government Investigations — recognition built in exactly the kind of record-intensive, contested litigation that loss disputes demand. Together with Managing Director Christopher Zoukis, our firm brings deep experience in federal fraud defense and sentencing. Loss calculation in fraud cases is governed by U.S. Sentencing Guideline § 2B1.1.

This guide explains how loss is calculated in federal fraud sentencing — the 2B1.1 loss table, the difference between actual and intended loss, the reasonable-estimate standard, credits against loss, gain as an alternative measure, and the sophisticated-means and victim enhancements that stack on top of the loss figure. It also covers Amendment 827, which moved the definition of loss into the guideline text itself effective November 1, 2024. If you are facing a federal fraud sentencing, the loss figure is where some of the most important advocacy will happen.

Quick Answer

Question	Answer
Why does the loss amount matter?	Under Guideline § 2B1.1, the loss amount drives the offense level in a fraud case — it can add from zero to roughly 30 levels, often determining the length of the sentence.
What is actual loss?	The reasonably foreseeable pecuniary harm that resulted from the offense.
What is intended loss?	The pecuniary harm the defendant purposely sought to inflict, which the guideline text now says includes harm that was impossible or unlikely to occur.
What are credits against loss?	Amounts that reduce loss — generally money returned and the value of property or services the victims received before the offense was detected.
Does loss have to be exact?	No. The court need only make a reasonable estimate of the loss — but that estimate must rest on a sound methodology and reliable evidence.
What changed in 2024?	Amendment 827 moved the definition of loss — including intended loss and the gain alternative — from the commentary into the text of § 2B1.1 itself, effective November 1, 2024.
How can our firm help?	We litigate every step of the loss calculation — definitions, methodology, credits, and enhancements — beginning with a paid, one-hour initial consultation focused on the numbers in your case.

Key Takeaways

• In federal fraud sentencing under Guideline § 2B1.1, the loss amount is the dominant driver of the offense level.
• The 2B1.1 loss table adds nothing at $6,500 or less and up to 30 levels for losses above $550 million.
• Loss is the greater of actual loss (foreseeable harm that resulted) or intended loss (harm the defendant sought to inflict).
• The court need only make a reasonable estimate of loss — but the estimate must be sound and evidence-based.
• Credits against loss reduce the figure by money returned and the value of property or services victims received.
• Where loss cannot reasonably be determined, the defendant’s gain may be used as an alternative, usually lower, measure.
• Amendment 827 (effective November 1, 2024) moved the loss definition — including intended loss — from the commentary into the text of § 2B1.1, resolving the circuit split created by United States v. Banks.
• Victim enhancements under § 2B1.1(b)(2) and the sophisticated-means enhancement under § 2B1.1(b)(10)(C) stack additional levels on top of the loss table.
• The loss figure first appears in the presentence report and can be challenged on methodology, causation, intent, credits, and attribution.

Why the Loss Amount Matters So Much

Federal fraud and theft offenses are sentenced under Guideline § 2B1.1. That guideline sets a base offense level, and then — through the loss table in § 2B1.1(b)(1) — increases the offense level in steps as the dollar amount of loss rises. The loss enhancement is the dominant driver of a fraud sentence.

The effect is dramatic. The 2B1.1 loss table starts at no increase when the loss is $6,500 or less, adds 2 levels once the loss exceeds $6,500, and climbs in 2-level steps to a 30-level increase for losses above $550 million. Because every offense level translates into additional months — and several levels can mean years — the loss figure often determines, more than any other single fact, how long a sentence will be.

This is why loss is the central battleground in fraud sentencing. The government will advance a loss figure, usually through the presentence report. But that figure rests on definitions, assumptions, and methodology — all of which can be examined and contested. A loss number is the end of an analysis, and every step of that analysis is open to challenge.

Actual Loss vs. Intended Loss

Since November 1, 2024, the text of § 2B1.1 itself — in Notes to the loss table added by Amendment 827 — defines loss as the greater of actual loss or intended loss. Before that date, the same rule appeared only in the guideline’s commentary, a distinction that produced a circuit split discussed below. Understanding the two measures is essential.

Actual loss is the reasonably foreseeable pecuniary harm that resulted from the offense. The key words are “reasonably foreseeable” and “resulted” — the loss must be a foreseeable consequence of the offense and must actually have occurred. Intended loss is the pecuniary harm that the defendant purposely sought to inflict, and the guideline now specifies that it includes harm that would have been impossible or unlikely to occur — for example, in a sting operation or where a fraudulent claim exceeded the value at stake.

The court uses whichever figure is greater. In many cases the two diverge sharply: a scheme may have been interrupted before causing much actual harm, while the amount the defendant aimed at was far larger — or vice versa. Which measure governs, and how each is calculated, is frequently where loss litigation begins.

Applied Insight: Intended loss is where the most aggressive loss figures often come from. Because it can include harm that was impossible or unlikely to occur, the government can build a large number from what a scheme theoretically aimed at rather than what it actually did. Scrutinizing whether the claimed intended loss reflects a genuine, provable intent — rather than a worst-case projection — is a core defense task.

The Reasonable-Estimate Standard

The Guidelines do not require the loss figure to be calculated with precision. The commentary provides that the court need only make a reasonable estimate of the loss, recognizing that the sentencing judge is well positioned to assess the evidence.

This standard cuts both ways. It means the government does not have to prove loss to the dollar — but it also means the government’s estimate must actually be reasonable, supported by a sound methodology and reliable evidence. An estimate built on speculation, on double-counting, or on an unsound method is not a reasonable estimate, and it can be challenged on exactly that ground.

Because loss is generally found by a preponderance of the evidence, the defense’s task is to test the government’s methodology, to offer a sounder one where the facts support it, and to insist that any estimate rest on evidence rather than assumption. A reasonable estimate is a real standard, and holding the government to it is meaningful advocacy.

Credits Against Loss

One of the most important — and most often overlooked — features of loss calculation is the principle of credits against loss. The loss figure is not simply the gross amount that moved through a scheme; certain amounts must be subtracted.

The Guidelines commentary provides that loss is generally reduced by the value of money returned, and the fair market value of property returned or services rendered, by the defendant or others before the offense was detected. In other words, where a victim received something of genuine value, or where money was paid back, those amounts can reduce the loss figure. In a mortgage fraud case, for example, the value of collateral may be credited; in a goods-or-services case, the value actually delivered may be.

Credits against loss can substantially lower the loss figure, and with it the offense level. Identifying every available credit — every dollar repaid, every item of value the victims received — is a key part of loss litigation, and it is an area where careful, fact-intensive work pays off directly in offense levels.

Gain as an Alternative Measure

In some cases, loss cannot reasonably be determined. The guideline text — since Amendment 827 — addresses this directly: where there is a loss but the amount cannot reasonably be determined, the court shall use the gain that resulted from the offense as an alternative measure.

Gain is generally understood to be a more conservative measure — it ordinarily understates the loss, because a defendant typically does not keep the full amount a scheme caused. The gain measure is an alternative, used only when loss itself cannot reasonably be estimated, not a free-standing choice the government can elect.

For the defense, the gain alternative can be significant. Where the government’s loss figure is genuinely speculative and cannot be reasonably estimated, arguing that gain is the appropriate measure can produce a substantially lower number — and a substantially lower offense level.

Sophisticated Means and Victim Enhancements

Loss is the dominant number in a fraud sentencing, but it is not the only one. Section 2B1.1 contains specific offense characteristics that stack additional levels on top of the loss table, and two of them appear in case after case: the victim enhancements and the sophisticated-means enhancement.

Under § 2B1.1(b)(2), the court applies the greatest of three victim-related tiers: 2 levels if the offense involved 10 or more victims, was committed through mass-marketing, or resulted in substantial financial hardship to one or more victims; 4 levels if it caused substantial financial hardship to five or more victims; and 6 levels if it caused substantial financial hardship to 25 or more victims. Who counts as a victim — and whether anyone truly suffered substantial financial hardship — are contestable questions, and the answers can move the offense level meaningfully.

Under § 2B1.1(b)(10)(C), the court adds 2 levels — and the offense level becomes at least 12 — if the offense involved sophisticated means and the defendant intentionally engaged in or caused the conduct constituting sophisticated means. Sophisticated means refers to especially complex or especially intricate conduct in executing or concealing the offense, such as hiding assets or transactions through shell entities, fictitious transactions, or offshore accounts. Two defense points matter here: ordinary concealment is not sophistication, and the enhancement turns on the defendant’s own intentional conduct, not the scheme’s overall complexity. A participant who had no role in the sophisticated features of a scheme should not receive it.

Legal Disputes Over How Loss Is Defined

Loss calculation spent years at the center of a structural fight. The text of Guideline § 2B1.1 used the bare word “loss,” while the definition that included intended loss — and most of the detailed rules for calculating loss — appeared in the commentary. Under Stinson v. United States, 508 U.S. 36 (1993), Guidelines commentary is authoritative unless it violates the Constitution or a federal statute or is inconsistent with, or a plainly erroneous reading of, the guideline — so for decades courts applied the commentary’s loss definition as a matter of course.

That consensus broke after Kisor v. Wilkie, 588 U.S. 558 (2019), in which the Supreme Court held that an agency’s interpretation of its own regulation receives deference only where the regulation is genuinely ambiguous. Applying Kisor, the Third Circuit held in United States v. Banks, 55 F.4th 246 (3d Cir. 2022), that the ordinary meaning of “loss” in § 2B1.1 is actual loss, accorded the commentary’s intended-loss definition no weight, and vacated a sentence built on intended loss where no actual loss had occurred.

Banks produced a genuine split: defendants in the Third Circuit were sentenced on actual loss alone, while intended loss continued to apply in every other circuit. That geographic disparity — the same failed scheme yielding a materially different offense level depending only on where it was prosecuted — is what prompted the Sentencing Commission to act.

Amendment 827: The Loss Definition Moves Into the Guideline

Effective November 1, 2024, Amendment 827 resolved the split. The Commission created Notes to the loss table in § 2B1.1(b)(1) and moved the general rule — loss is the greater of actual loss or intended loss — out of the commentary and into the guideline text itself, along with the gain alternative and the definitions of actual loss, intended loss, pecuniary harm, and reasonably foreseeable pecuniary harm. Because the rule now sits in the guideline rather than the commentary, the Banks deference argument no longer applies to sentencings under the current manual.

Timing still matters. Under § 1B1.11, the court applies the Guidelines Manual in effect on the date of sentencing unless doing so would violate the Ex Post Facto Clause, in which case the manual in effect when the offense was committed governs. For cases sentenced under pre-November 2024 manuals — including cases now on appeal or in post-conviction review — whether a commentary-deference challenge to intended loss remains available is a live, circuit-specific question that deserves close analysis.

Two further developments round out the picture. The Commission described Amendment 827 as a consistency fix and noted that it may still undertake a comprehensive review of § 2B1.1 in a future amendment cycle, so the fraud guideline may change again. And the Commission’s 2025 simplification amendments, effective November 1, 2025, removed the departure step from the former three-step sentencing process — meaning arguments that a loss-driven range overstates real culpability are now presented as variances under 18 U.S.C. § 3553(a) rather than as Guidelines departures.

Applied Insight: Loss calculation rewards counsel who treat it as live, contested law rather than settled arithmetic. The actual-versus-intended-loss question, the reasonable-estimate standard, the full set of available credits, and the question of which manual — and which loss definition — governs a particular case all create room to argue for a lower figure. In fraud sentencing, the loss number is rarely as fixed as the presentence report makes it appear.

Challenging the Loss Figure

The loss figure first appears in the presentence investigation report, and the defense can object to it. A challenge to loss can take several forms.

• Methodology. The defense can argue that the government’s method of estimating loss is unsound, speculative, or not a reasonable estimate.
• Causation and foreseeability. Actual loss must be reasonably foreseeable harm that resulted from the offense — losses from other causes, or harm too remote to be foreseeable, should not be counted.
• Intended loss. The defense can contest whether a claimed intended-loss figure reflects a genuine, provable intent rather than a worst-case projection.
• Credits. The defense can identify every dollar repaid and every item of value the victims received, and insist those credits be applied.
• Attribution. In a multi-participant case, the defense can argue that only the loss properly attributable to this defendant should count.

Because the loss table moves in steps, even a partial reduction in the loss figure can drop the offense level — and the advisory range — significantly. Loss litigation is detailed, document-intensive work, and it is among the highest-value advocacy available in a fraud sentencing.

How Our Firm Litigates Loss

At Elizabeth Franklin-Best, P.C., we approach the loss figure as the central contest of a fraud sentencing. Our principal attorney, Elizabeth Franklin-Best, has appeared in more than 330 federal proceedings, including over 100 appeals across the federal circuits and at the United States Supreme Court, and she is recognized by Best Lawyers in America 2026 in Appellate Practice and ranked by Chambers USA 2026 in Litigation: White-Collar Crime & Government Investigations. That record matters here because a loss dispute is won on the documents and preserved for appeal in the same motion — so we build the loss record with the standard of review already in mind, treating the government’s figure as a methodology to be dismantled rather than a number to be accepted.

Our loss work includes independently analyzing the government’s loss methodology, contesting actual and intended loss where the facts allow, holding the government to the reasonable-estimate standard, identifying and pressing every credit against loss, arguing for the gain measure where loss cannot reasonably be determined, raising the applicable circuit law on the loss definition, and filing detailed objections to the presentence report. We represent federal defendants nationwide through admission pro hac vice.

For related reading, see our federal sentencing overview, our guide to the Sentencing Guidelines, our companion guides to acceptance of responsibility, role adjustments, and criminal history, and our federal fraud defense hub.

Talk With a Federal Sentencing Lawyer

In a federal fraud case, the loss figure can be the difference between a moderate sentence and a severe one. That figure is not a fixed fact — it is the product of definitions, methodology, and assumptions, every one of which can be tested. Effective loss litigation is among the most valuable work a defense team can do.

Our firm offers a paid, one-hour initial consultation. We will go through the loss figure line by line — the definitions applied, the methodology behind the estimate, the credits that may have been missed, and the enhancements stacked on top — and tell you candidly where the number can be fought. If a loss calculation is driving your exposure, have it examined before sentencing, not after.