General Publications March 16, 2022

“How to Mitigate Sentencing Liability in Complex Fraud Cases,” Law360, March 16, 2022.

Extracted from Law360

Elizabeth Holmes' recent convictions, as well as the U.S. Department of Justice's current focus on individual culpability for corporate fraud offenses,[1] have generated renewed interest in how these cases are sentenced, particularly when they involve substantial financial loss to numerous victims.

The first line of defense is to limit the accounting and assignment of loss, and the application of what are often overly punitive enhancements under the federal sentencing guidelines.

The amount of loss, including so-called relevant conduct, is the primary driver of prison exposure as a guidelines matter. In market-related schemes, that loss can be tens or even hundreds of millions of dollars.

The guidelines also enhance punishment based on executive status and whether the offense involved securities fraud at a public company.

We'll soon see how these issues play out in the Holmes prosecution. But they also routinely play out in less-publicized cases, as defense counsel struggle with cabining loss and limiting the application of guideline enhancements that don't discriminate by market cap.

While the guidelines are advisory, they remain the most important factor informing judicial sentencing discretion.[2]

Favorable litigation of loss and the applicability of what can often be arbitrary guideline enhancements is therefore critical. After all, the guidelines range determines the height of the diving board from which a defendant must leap in seeking a guideline variance to arrive at a reasonable sentence.

Loss Considerations and the SOX Swing

Sentencing in federal fraud cases is enhanced, if not dictated, by actual or intended loss — whichever is greater — which need only be estimated. According to the U.S. Sentencing Commission's Guidelines Manual, actual loss is "reasonably foreseeable pecuniary harm," while intended loss is "pecuniary harm that the defendant purposely sought to inflict."[3]

Reasonably foreseeable pecuniary harm is "harm that the defendant knew or, under the circumstances, reasonably should have known, was a potential result of the offense."[4]

Loss under the guidelines must also account for relevant conduct. In conspiracy and scheme cases, this means all acts that were within the scope of jointly undertaken criminal activity.[5]

Federal prosecutors further seek to apply specific offense characteristics, or SOCs, under the fraud offense guideline, many of which were adopted as part of the Sarbanes-Oxley Act and have become ubiquitous in complex fraud cases.

They apply, for example, if the fraud involved substantial victim financial hardship; sophisticated means; more than $1 million in gross receipts from a financial institution; insolvency of a publicly traded company; or a securities law violation by a public company officer or director.[6]

Application of only some of these SOCs, in combination with tens or even hundreds of millions of dollars in reasonably foreseeable losses, can lead to decades of prison exposure under the guidelines.

Limiting Loss

The reasonably foreseeable loss standard implicates a but-for causation requirement, meaning a defendant is responsible solely for losses that would not have occurred but for his offense conduct.

If, for example, the CEO of a publicly traded company misrepresents his company's earnings, he is responsible solely for trading directly attributable to this misrepresentation — not all losses claimed by investors.[7]

And in so-called hub-and-spoke conspiracies, where one or more hub conspirators have conspired with spoke conspirators, courts must treat each spoke separately. If spokes operated independently — that is, if the government cannot rim the wheel — then losses associated with one spoke are not reasonably foreseeable to others.[8]

Indeed, courts must determine the scope of a defendant's criminal agreement before assessing responsibility for third-party acts.[9]

The relevant conduct guideline makes clear that

[a]cts of others that were not within the scope of the defendant's agreement, even if those acts were known or reasonably foreseeable to the defendant, are not relevant conduct.[10]

Courts thus routinely distinguish between mere awareness of risk versus actual intent to generate harm.[11] This is true even if the offense and relevant conduct involve effectively identical illegal activity.[12]

Evidentiary Tools for Cabining Fraud Loss

The DOJ has stated that it will rely on parallel agency enforcement and whistleblowers in seeking to prosecute corporate fraud offenses.

The experts who opine on causation and scope of loss in civil cases are similarly relevant and persuasive at sentencing.

In addition, event studies — statistical tools that measure market reliance — are not only highly relevant, but also foreign to many prosecutors.[13] They distinguish what is frequently the limited impact of misstatements from other market forces.[14]

Especially in securities fraud prosecutions involving market reactions to discrete announcements, event studies focusing on causation facilitate exclusion of unforeseeable losses.

Because actual loss is not a required criminal liability element, it is rare for the government to ensure, or even to evidence, that a loss figure at sentencing corresponds to actual victims.

Conducting — or inquiring if the government has conducted — a statistically relevant sampling of alleged victim reliance is a powerful tool in limiting loss.

In the absence of such sampling, the government, which bears the burden, will simply ask the court to extrapolate from limited proof of victimization offered during a plea or at trial.

Avoiding the SOX Swing

The fraud guideline SOCs that have a stratospheric effect on prison exposure were adopted in response to the notorious corporate fraud scandals of the early 2000s.

But they apply equally to the mine-run of federal fraud prosecutions implicating the solvency of smaller public companies and executives who happen to be convicted of securities fraud.

Courts recognize that these SOX swing enhancements often dramatically overstate culpability.

When the SOX swing is attributable merely to status — simply being an executive at a small public company — rather than culpability, a downward departure or variance is appropriate.[15]

Fraud committed at a Fortune 50 company is likely more aggravated in terms of intent and public harm than fraud involving a small public company.

Yet both can, and often do, implicate sophisticated means; substantial financial harm to multiple victims; and executive-level securities or commodities fraud convictions.

Sentencing statistics show that courts frequently sentence below the applicable guideline range if rote application of loss and SOCs would result in excessive prison exposure.

From 2016 to 2020, departures or variances occurred in 65% to 80% of cases involving at least 324 months of guidelines exposure.[16]

And fraud defendants confronted by a guidelines range of 360 months to life received an average sentence of 186 months, a nearly 50% reduction.[17]

The sentencing statistics also reflect significant departures and variances in more typical fraud prosecutions. In 2020, courts downwardly varied or departed from the bottom of the fraud guideline range in 40.8% of cases, corresponding to an average sentence of 21 months.[18]

Invocation of Civil Fraud Precedents

In addition to importing damages experts and event studies from the civil securities fraud context to limit loss, invocation of civil precedents also cabins liability.

In Janus Capital Group Inc. v. First Derivative Traders, for example, the U.S. Supreme Court held in 2011 that the maker of a false statement for purposes of U.S. Securities and Exchange Commission Rule 10b-5 is one who has "ultimate authority over the statement, including its content and whether and how to communicate it."[19]

Primary liability attaches solely to the maker of a false statement, not someone who has participated either in drafting or the making of an oral false statement by another.[20]

This is a particularly important distinction when, to establish specific intent, most federal securities fraud indictments sound in maker liability for written and oral misrepresentations.

And when criminal defendants face greater exposure to loss of liberty than their civil or administrative counterparts who face only financial penalties, this disparity should factor at sentencing.

Materiality is an essential element of both civil and criminal securities fraud. But in the criminal context, materiality does not mean that anyone was actually deceived.[21]

Misrepresentations that are incapable of belief remain criminally actionable,[22] but the reason courts look to but-for causation, reliance and reasonable foreseeability is that prison time has to account for these concepts post-conviction.


Loss and guideline enhancements that inflate fraud liability require that defense counsel actively litigate whether loss was foreseeable.

This implicates issues of causation and reliance — typically the domain of civil litigation.

They also require that counsel rationalize and mitigate the effect of other guideline enhancements by ensuring they correspond to factual and moral culpability, rather than simply status.

1] See "Deputy Attorney General Lisa O. Monaco Gives Keynote Address at ABA's 36th National Institute on White Collar Crime," Oct. 28, 2021,

[2] See United States Sentencing Commission, The Influence of the Guidelines on Federal Sentencing: Federal Sentencing Outcomes, 2005-2017, at 27-31 (Dec. 2020) (noting that sentences imposed under U.S.S.G. Section 2B1.1, the principal fraud offense guideline, averaged between 75% to 80% of the minimum custodial guideline range between 2005 and 2017).

[3] U.S.S.G. Section 2B1.1, cmt. 3(A)(i)-(ii).

[4] Id., cmt. 3(A)(iii)-(iv).

[5] U.S.S.G. Section 1B1.3.

[6] U.S.S.G. Sections 2B1.1(b)(2), (10), (17)(A)-(B), and (20).

[7] See, e.g., United States v. Stein, 846 F.3d 1135, 1153-54 (11th Cir. 2017) (remanding for resentencing because the government failed to prove but-for causation); United States v. Lonich, 2022 U.S. App. LEXIS 623 (9th Cir. Jan. 10, 2022) (remanding for resentencing because the government failed to demonstrate that the defendant's concealment of millions of dollars in illicit loans actually caused his bank to fail).

[8] See generally United States v. Chandler, 388 F.3d 796, 808 (11th Cir. 2004) ("A rimless wheel conspiracy is one in which various defendants enter into separate agreements with a common defendant, but where the defendants have no connection with one another, other than the common defendant's involvement in each transaction.").

[9] U.S.S.G. Section 1B1.3(a)(1)(B).

[10] Id., cmt. 3(B).

[11] See, e.g., United States v. Studley, 47 F.3d 569, 575 (2d Cir. 1995) ("The fact that the defendant is aware of the scope of the overall operation is not enough to hold him accountable for the activities of the whole operation.").

[12] See, e.g., United States v. McClatchey, 316 F.3d 1122, 1128 (10th Cir. 2003) (holding that the government had "misperceive[d] the nature" of the relevant conduct inquiry by advocating that all improper kickback payments were reasonably foreseeable to a defendant who had paid bribes in connection with only one of several fraudulent contracts at issue).

[13] See generally Jill E. Fisch and Jonah B. Gelbach, "Power and Statistical Significance in Securities Fraud Litigation," 11 Harvard Business Law Review 55 (2021) (discussing use of event studies).

[14] See, e.g., United States v. Heine, No. 3:15-cr-238-SI, 2018 U.S. Dist. LEXIS 99922, at *17 (D. Ore. June 14, 2018) (observing that the defendants' misrepresentations were not the sole cause of bank loss and that "an event study or regression analysis might have quantified the loss to the Bank caused by Defendants' relevant conduct, as opposed [to] other factors").

[15] See, e.g., United States v. Parris, 573 F. Supp. 2d 744, 746 (E.D.N.Y. 2008) (sentencing defendants to 60 months of incarceration, despite an applicable guidelines range of 360 months to life, because they "were engaged in a rather typical 'pump and dump' scheme in the world of the high-risk penny-stock investor" and were not engaged in conduct "of the same character and magnitude as the securities-fraud prosecutions of those who have been responsible for wreaking unimaginable losses on major corporations and, in particular, on their companies' employees and stockholders").

[16] United States Sentencing Commission, Judiciary Sentencing Information (JSIN), available at

[17] Id.

[18] United States Sentencing Commission, 2020 Sourcebook of Federal Sentencing Statistics, "2020 Sentences Under the Guidelines Manual and Variances," Table 31, available at United States Sentencing Commission, Interactive Data Analyzer, available at

[19] Janus Capital Group, Inc. v. First Derivative Traders, 564 U.S. 135, 142 (2011).

[20] Id. at 145; but see Lorenzo v. SEC, 139 S. Ct. 1094, 1099 (2019) (holding that dissemination of false or misleading statements with intent to defraud can fall within the scope of secondary Rule 10b-5 liability, even if the defendant did not in fact "make" the false statement). Courts have also disallowed the "scheme" provisions of subsections (a) and (c) of Rule 10b-5 to serve as a back door to liability for those who help others make a false statement or omission in violation of Rule 10b-5(b). See, e.g., In re Parmalat Sec. Litig., 376 F. Supp. 2d 472, 503 (S.D.N.Y. 2005).

[21] Compare Basic, Inc. v. Levinson, 485 U.S. 224, 238 (1988) (holding that for civil liability, "[i]t is not enough that a statement is false or incomplete, if the misrepresented fact is otherwise insignificant" to a reasonable investor), with Neder v. United States, 527 U.S. 1, 25 (1999) (materiality in the federal mail and wire fraud context encompasses mere capability of influencing action, even if a reasonable person would not be so influenced).

[22] United States v. Thomas, 377 F.3d 232, 243 (2d Cir. 2004) (victim gullibility is irrelevant to federal mail and wire fraud liability); see also United States v. Maxwell, 920 F2d 1028, 1036 (D.C. Cir. 1990) ("Appellant is simply wrong, however, if she means to assert that the wire fraud statute does not apply where the persons defrauded unreasonably believed the misrepresentations made to them.").

Meet the Authors
Media Contact
Alex Wolfe
Communications Director

This website uses cookies to improve functionality and performance. For more information, see our Privacy Statement. Additional details for California consumers can be found here.