UNITED STATES v. BOLER (2024)

Maggie Boler (“Appellant”) was convicted of six counts of presenting false claims against the United States by submitting false tax returns to the Internal Revenue Service (“IRS”), and one count of making a false statement on her fraudulent Paycheck Protection Program ¹ (“PPP”) loan application. Appellant submitted six tax returns to the IRS but only received refunds on four of those returns. As a result of her convictions, Appellant was sentenced to 30 months of imprisonment.

The sole issue in this appeal is whether Appellant's United States Sentencing Guidelines (“Guidelines”) sentencing range can rely on the entire financial harm Appellant intended to cause, even though she never received the funds from the two tax returns denied by the IRS.

We conclude that the complete extent of Appellant's intended financial harm can be utilized in determining her Guidelines sentencing range. Thus, we affirm because the district court correctly incorporated Appellant's full intended loss amount into the sentencing calculation.

I.

A.

Appellant was prosecuted for her involvement in a fraudulent tax scheme. Part of her scheme was filing fraudulent tax returns to the IRS, claiming larger refund amounts than she and others were entitled to. Appellant used fictitious interest income and fabricated federal income tax withholdings to claim these large tax refunds. For the 2016 tax year, Appellant submitted six fraudulent tax returns on behalf of herself and others, falsely claiming a total of $159,389 in tax refunds. The IRS denied two of the fraudulent tax returns but paid the other four tax refunds, which totaled $116,106.

In 2021, Appellant applied for a PPP loan. In her PPP loan application, Appellant stated that her business, named “Maggie A Boler,” had an average monthly payroll of $9,500 for one employee -- herself. In her application, Appellant requested $20,833 based on this information. This information was false. There was no evidence supporting the existence of any business in Appellant's name. Namely, there were no “business expenses,” “required withholdings for a business such as payments to Social Security,” or any “tax payments.” J.A. 93.² And the IRS could not locate “any payroll payments to [Appellant] for $9,500 a month.” Id. at 92–93. As a result of her fraudulent PPP loan application, Appellant received $20,833 to which she was not entitled.

B.

In 2022, Appellant was indicted on six counts of presenting false claims against the United States in violation of 18 U.S.C. § 287 and one count of making a false statement via the PPP loan application in violation of 18 U.S.C. § 1014. Appellant proceeded to trial and was found guilty on all counts.

Prior to sentencing, the United States Probation Office (the “Probation Office”) prepared Appellant's Presentence Investigation Report (“PSR”). In the PSR, the Probation Office calculated the loss associated with Appellant's fraudulent scheme in order to determine the recommended sentencing enhancements pursuant to Guidelines § 2B1.1(b)(1). In doing so, the Probation Office relied on the commentary to section 2B1.1, which defines “loss” as the “greater of actual loss or intended loss.” Guidelines § 2B1.1 cmt. 3(A). The commentary further defines “[a]ctual loss” as the “reasonably foreseeable pecuniary harm that resulted from the offense,” whereas “[i]ntended loss” is defined as “the pecuniary harm that the defendant purposely sought to inflict,” including any “intended pecuniary harm that would have been impossible or unlikely to occur.” Id. § 2B1.1 cmt. 3(A)(i)–(ii).

Based on the commentary, the Probation Office calculated Appellant's intended loss amount as $180,222. This amount combined all six fraudulent tax return refunds Appellant submitted to the IRS, which amounted to $159,389, plus the $20,833 PPP loan. Although the IRS did not pay two of the claimed refunds, the Probation Office nevertheless counted all six tax returns toward Appellant's sentencing enhancements pursuant to Guidelines § 2B1.1 cmt. 3(A)(ii). That is because all six tax returns, plus the PPP loan amount, represented the full amount of harm Appellant intended to cause. On the other hand, the actual loss Appellant caused equaled the total refunds she received from the IRS ($116,106) plus her PPP loan amount ($20,833) for a total of $136,939. Because the commentary requires the sentencing court to apply the higher of a defendant's actual or intended loss, the Probation Office relied on the intended loss amount of $180,222. As a result, the Probation Office recommended a ten-level sentencing enhancement because the loss amount exceeded $150,000 but was not more than $250,000. See Guidelines § 2B1.1(b)(1)(F).

Appellant objected to the calculated loss amount contained in the PSR, arguing that the loss calculation should not include the two tax returns rejected by the IRS. The Probation Office rejected Appellant's contention and maintained the recommended ten-level enhancement in the PSR.

At the sentencing hearing, Appellant articulated her objection to the calculated loss amount. Appellant relied on Kisor v. Wilkie, 588 U.S. 558, 139 S.Ct. 2400, 204 L.Ed.2d 841 (2019), which requires a court to determine that a regulation is genuinely ambiguous before deferring to an agency's interpretation of its own regulation -- in this case, the Guidelines' commentary. Appellant argued that the ordinary meaning of “loss” as written in Guidelines § 2B1.1 is not ambiguous. She contended that “loss” encompasses the actual amount lost by the victim and not any intended loss. Therefore, in her view, the unambiguous meaning of “loss” only included actual loss and the commentary could not expand the plain text. Thus, Appellant argued that the two tax returns rejected by the IRS could not be included in the loss calculation.

The district court overruled Appellant's objection. The district court held that “loss” as written in section 2B1.1 does not “draw any distinction between” actual and intended loss. J.A. 282. While the district court did not explicitly state that the term “loss” is ambiguous, it did observe that nothing in the text of the Guideline “says [that] loss can't be both” actual and intended. Id. Thus, in the district court's view, the commentary's definitions of “loss” simply provides “an explanation of the types of loss that you can have, and both would be appropriate to determine what the guideline range would be.” Id. at 283. Based on a total offense level of 19, which included the ten-level enhancement for the loss calculation, and a criminal history category of I, the district court calculated Appellant's Guidelines sentencing range to be 30–37 months of imprisonment. The district court imposed a sentence of 30 months of imprisonment.

This timely appeal followed.

II.

We review de novo the district court's legal interpretation of the term “loss.” United States v. Miller, 316 F.3d 495, 498 (4th Cir. 2003). But “to the extent that the determination of the amount of loss is a factual matter, we review only for clear error.” United States v. West, 2 F.3d 66, 71 (4th Cir.1993) (internal quotation marks omitted).

III.

A.

The Guidelines set forth certain increases to a defendant's offense level depending on the amount of “loss” related to the fraud committed. See Guidelines § 2B1.1(b)(1). The relevant Guideline, section 2B1.1, directs a sentencing court to increase the offense level pursuant to the Guidelines where “the loss exceeded $6,500.” Id. Section 2B1.1 does not define what constitutes “loss” for purposes of calculating the applicable sentencing enhancement. But the Guidelines' commentary to section 2B1.1 lists several definitions of loss and instructs the sentencing court as to how to calculate the appropriate loss amount. See id. cmt. 3(A). The Guidelines' commentary defines “[a]ctual loss” as the “reasonably foreseeable pecuniary harm that resulted from the offense.” Id. cmt. 3(A)(i). And it defines “[i]ntended loss” as “the pecuniary harm that the defendant purposely sought to inflict,” including “intended pecuniary harm that would have been impossible or unlikely to occur.” Id. cmt. 3(A)(ii). The commentary instructs the sentencing court to apply the loss that is “the greater of actual loss or intended loss.” Id. cmt. 3(A).

When deciding whether to defer to the Guidelines' commentary, we apply the framework set forth in Kisor v. Wilkie, 588 U.S. 558, 139 S.Ct. 2400, 204 L.Ed.2d 841 (2019). See United States v. You, 74 F.4th 378, 397 (6th Cir. 2023). Prior to Kisor, our analysis of the Guidelines' commentary was guided by the standard outlined in Stinson v. United States, 508 U.S. 36, 113 S.Ct. 1913, 123 L.Ed.2d 598 (1993). In Stinson, the Supreme Court “held that the commentary deserves the same deference that courts give agencies' interpretations of their own rules” -- commonly known as Seminole Rock/Auer deference.³ United States v. Campbell, 22 F.4th 438, 444 (4th Cir. 2022) (citing 508 U.S. at 44–45, 113 S.Ct. 1913). Based on Stinson, we were to defer to the commentary unless its interpretation is “plainly erroneous or inconsistent” with the plain language of the Guideline. Id. at 45, 113 S.Ct. 1913; see also id. at 45–47, 113 S.Ct. 1913 (discussing Stinson, Seminole Rock/Auer deference, and the “line of cases governing this type of deference”).⁴

But Stinson's analysis was altered by Kisor. As we noted in United States v. Campbell, “Kisor limited when courts will afford Seminole Rock/Auer deference” to an agency's interpretation of its own regulations, and in turn, to the commentary to the Guidelines. 22 F.4th at 445 (citing 588 U.S. 558, 139 S.Ct. 2400).

Pursuant to Kisor's directive, “Auer deference ‘can arise only if a regulation is genuinely ambiguous.’ ” Romero v. Barr, 937 F.3d 282, 291 (4th Cir. 2019) (quoting 588 U.S. at 573, 139 S.Ct. 2400). And a regulation can only be deemed “genuinely ambiguous” if uncertainty exists “even after a court has resorted to all the standard tools of interpretation,” including consideration of the “text, structure, history, and purpose of a regulation.” Kisor, 588 U.S. at 573, 575, 139 S.Ct. 2400. “If uncertainty does not exist, there is no plausible reason for deference. The regulation then just means what it means—and the court must give it effect, as the court would any law.” Id. at 574–75, 139 S.Ct. 2400.

Thus, concluding whether a rule is “genuinely ambiguous” is step one of the analysis. See Kisor, 588 U.S. at 573, 139 S.Ct. 2400. Once a regulation is determined to be genuinely ambiguous, “the agency's reading must still fall ‘within the bounds of reasonable interpretation’ ” in order to be accorded deference. Id. at 559, 139 S.Ct. 2400 (quoting Arlington v. FCC, 569 U.S. 290, 296, 133 S.Ct. 1863, 185 L.Ed.2d 941 (2013)). In other words, the agency's interpretation of its regulation “must come within the zone of ambiguity” created by the ambiguous text of the regulation. Id. at 576, 139 S.Ct. 2400.

Still, Kisor warns that “not every reasonable agency reading of a genuinely ambiguous rule should receive Auer deference.” Kisor, 588 U.S. at 576, 139 S.Ct. 2400. We must also “make an independent inquiry into whether the character and context of the agency interpretation entitles it to controlling weight.” Id. This inquiry ensures that courts defer only to those agency readings “that Congress would have wanted [them] to.” See id.

While there is no “exhaustive test” to determine whether the character of an agency's interpretation is entitled to Auer deference, the Supreme Court has “laid out some especially important markers.” Kisor, 588 U.S. at 576–77, 139 S.Ct. 2400. First, the interpretation must be the “official position” of the agency. Id. at 577, 139 S.Ct. 2400 (internal quotation marks omitted). “Next, the agency's interpretation must in some way implicate its substantive expertise.” Id. at 579, 139 S.Ct. 2400. “Finally, an agency's reading of a rule must reflect its ‘fair and considered judgment.’ ” Id. (quoting Auer, 519 U.S. at 462, 117 S.Ct. 905).

B.

The issue we confront in this case is whether the term “loss” in Guidelines § 2B1.1 is genuinely ambiguous, and if it is, whether we should defer to the Guidelines' commentary to ascertain its meaning.

1.

At step one of the Kisor analysis, we must determine whether a genuine ambiguity exists in section 2B1.1's use of the term “loss.” If the term is ambiguous, we then consider the commentary. Appellant, relying on what she views as the ordinary meaning of loss as well as the canons of statutory interpretation, argues that the term “loss” is not ambiguous and, therefore, the district court erred by deferring to the commentary's definitions of “loss.” In response, the Government argues that “loss” can mean several different things, particularly in light of the context and history of the Guidelines. Thus, the Government contends that in the face of a genuine ambiguity, the district court properly deferred to the commentary.

a.

We begin with the ordinary meaning of the term “loss.” See United States v. Haas, 986 F.3d 467, 480 (4th Cir. 2021) (“[W]e construe terms in the Sentencing Guidelines according to their ordinary meaning.”).

Appellant argues that after considering the ordinary meaning of the term “loss,” it is not genuinely ambiguous. Appellant relies on dictionary definitions to argue that “loss” can only mean the “tangible, actual loss” suffered, for instance, “where [a] victim actually lost money.” Appellant's Opening Br. at 14. Appellant also points to a dictionary definition of loss as “the disappearance or diminution of value” to support that “loss” can only encompass “measurable actual loss.” Id. at 13–14 (citing Loss, Black's Law Dictionary (11th ed. 2019)). And Appellant argues that where “loss” plainly means actual loss, the commentary's definition of “intended loss” inappropriately expands the plain text of the Guidelines to include harm that did not occur.

In response, the Government asserts that the term “loss” is “not as limited as [Appellant] would claim.” United States Resp. Br. at 14. The Government points out that “loss” has a “myriad” of different definitions. Id. For instance, the Government cites the Merriam-Webster Dictionary, which alone has “seven distinct definitions of ‘loss,’ with those definitions often having multiple sub-definitions.” Id. And the Government disagrees with Appellant's contention that the definitions of “loss” exclusively refer to measurable actual loss. Instead, the Government points out that “loss” also includes intangible harm such as the “absence of a physical capability,” “the harm or privation resulting from losing or being separated from someone or something,” or a “failure to gain, win, obtain, or utilize” a goal. Id. (quoting Loss, Merriam-Webster, https://www.merriam-webster.com/dictionary/loss [https://perma.cc/J6HK-YEYJ]).

As noted by the Government, dictionary definitions do not reveal any single definition of the term “loss.” For example, the 1993 edition of Webster's New International Dictionary defines “loss” as any of the following:

(a) the act or fact of losing;

(b) a person or thing or an amount that is lost;

(c) the act or fact of failing to gain, win, obtain, or utilize;

(d) [a] decrease in amount, magnitude, or degree;

(e) the state or fact of being destroyed or placed beyond recovery; and

(f) the amount of an insured's financial detriment due to the occurrence of a stipulated contingent event.

Webster's New International Dictionary 1338 (1993); see also United States v. Banks, 55 F.4th 246, 257–58 (3d Cir. 2022) (discussing the same). In attempting to collect dictionary definitions of “loss,” the Sixth Circuit opined:

One dictionary defines the word to mean, among other things, the “amount of something lost” or the “harm or suffering caused by losing or being lost.” American Heritage Dictionary of the English Language 1063 (3d ed. 1992). Another says it can mean “the damage, trouble, disadvantage, [or] deprivation ․ caused by losing something” or “the person, thing, or amount lost.” Webster's New World College Dictionary 799 (3d ed. 1996). A third defines it as “the being deprived of, or the failure to keep (a possession, appurtenance, right, quality, faculty, or the like),” the “[d]iminution of one's possessions or advantages,” or the “detriment or disadvantage involved in being deprived of something[.]” 9 Oxford English Dictionary 37 (2d ed. 1989).

United States v. Riccardi, 989 F.3d 476, 486 (6th Cir. 2021) (alterations in original). These multiple and varied definitions alone demonstrate that “loss” could mean a number of different things depending on the dictionary of one's choice. Thus, there is “no single right answer” to the meaning of “loss” based on its plain reading. See Kisor, 588 U.S. at 575, 139 S.Ct. 2400; see also You, 74 F.4th at 397 (“[T]he term ‘loss’ in § 2B1.1 has no one definition and can mean different things in different contexts.” (internal quotation marks and citation omitted)); United States v. Rainford, 110 F.4th 455, ––––, 2024 WL 3628082 at *7 (2d Cir. Aug. 2, 2024) (same).

b.

Nor do the various cannons of statutory interpretation render the term “loss” as written in section 2B1.1 any less ambiguous.⁵ When interpreting a regulation, we do not read its language “in isolation.” Lynch v. Jackson, 853 F.3d 116, 121 (4th Cir. 2017). Rather “the words ․ must be read in their context and with a view to their place in the overall [regulatory] scheme.” Id. (internal quotation marks omitted). In interpreting a regulation's plain language, we look to “the specific context in which the language is used, and the broader context of the” regulation as a whole. Hurlburt v. Black, 925 F.3d 154, 158 (4th Cir. 2019) (en banc) (internal quotation marks omitted).

Appellant argues that the context of section 2B1.1 unambiguously limits “loss” to actual loss. Appellant first emphasizes that because loss must have “exceeded” $6,500 in order for an enhancement to apply, “loss” cannot include harm that was not actually inflicted. See Guidelines § 2B1.1(b)(1). Appellant asserts that “[b]y using the verb ‘exceed’ in its past tense in conjunction with a value, [section] 2B1.1(b)(1) anticipates the enhancement applying where the victim actually lost money.” Appellant's Opening Br. at 14. And Appellant avers that “[a] defendant's intent has nothing to do with measuring whether loss exceeded $6,500.” Id. (internal quotation marks omitted).

In response, the Government asserts that there is no need to delve into the guesswork urged by Appellant in order to discern the context of section 2B1.1 because the United States Sentencing Commission (“Commission”) has already provided ample context for “loss” in Guidelines § 1B1.3 -- the relevant conduct Guideline. In section 1B1.3, the Commission directs a sentencing court to consider not only the actual harm caused by a defendant but also “all harm that was the object of such acts and omissions.” Guidelines § 1B1.3(a)(3) (emphasis supplied). In the Government's view, this broad interpretation of “harm” expands the definition of “loss” within section 2B1.1 without resort to the commentary. In other words, “loss” is ambiguous where it can mean different types of harm, including actual or intended harm.

We agree with the Government's view that the context of section 1B1.3 applies to our interpretation of “loss” as written in section 2B1.1. The relevant conduct Guideline directs courts as to how to consider “harm” in the context of specific offense characteristics “[u]nless otherwise specified.” Guidelines § 1B1.3(a). In this instance, section 2B1.1 does not explicitly limit “loss” to actual or intended loss. Therefore, the language of section 1B1.3 provides a starting point for interpreting the section 2B1.1 enhancements. When read together, Guidelines sections 1B1.3 and 2B1.1 support the understanding that “loss” must encompass both the actual harm and the harm that was the object of Appellant's scheme.

Where, as here, the Commission has provided an expansive context for interpreting “harm” and the text of the Guideline at issue does not specify the type of “harm,” “loss” cannot be as limited as Appellant contends. That is true especially where it would undermine the Commission's directive. See Kisor, 588 U.S. at 590–91, 139 S.Ct. 2400 (directing the court to carefully consider not only the text but also the history and purpose of a regulation when assessing whether ambiguity exists). Indeed, section 1B1.3 lends more ambiguity to the meaning of “loss” because it can encompass more than the actual harm caused by a defendant.

c.

Appellant further argues that “loss” should be limited to “actual loss” where the Commission omitted any reference to intended harm. In Appellant's view, the decision to omit explicit language including “intended loss” from section 2B1.1 was intentional where similar language is present in other sections of the Guidelines.

In support, Appellant cites the tax evasion Guideline, where the Commission defines tax loss to include “the total amount of loss that was the object of the offense (i.e., the loss that would have resulted had the offense been successfully completed).” Guidelines § 2T1.1(c)(1). To support her argument, Appellant attempts to liken this case to the facts and Guideline at issue in United States v. Campbell, 22 F.4th 438. In Campbell, we held that where Guidelines section 4B1.2 defined “controlled substance offense[s]” to the exclusion of any attempt crimes, the commentary could not expand the plain text of the Guidelines to include inchoate crimes. Id. at 445. We reasoned that common sense and the canon of expressio unius dictated that “a definition which declares what a term means excludes any meaning that is not stated.” Id. (internal quotation marks omitted).

But the present case differs from Campbell because section 2B1.1 does not define “loss” and does not suggest that intended loss should be excluded from its definition. And, as previously discussed, the relevant conduct that Guideline § 1B1.3 directs the sentencing court to consider is more than just the tangible harm -- it includes unrealized but intended loss.

Further, contrary to the dissent's contention, our reading of section 2T1.1(c)(1) in comparison to section 1B1.3(a) does not render the language of other Guidelines sections “superfluous.” See Post at ––––. Reference to Guidelines § 2T1.1(c)(1) is not a useful comparator. It simply does not apply here.

Section 1B1.3(a) instructs sentencing courts as to how to interpret “harm” as used throughout the Guidelines when the definition of harm is not “otherwise specified.” Guidelines § 1B1.3(a) (emphasis supplied). The fact that other sections, such as Guidelines § 2T1.1(c)(1), specify that the harm at issue in that particular section also includes intended loss does not make our interpretation of section 1B1.3 inappropriate in this case. That is because here, the Guideline at issue does not define the type of harm the court should consider. Indeed, section 1B1.3 does not apply to the tax section cited by both the Appellant and the dissent because there, unlike here, the type of harm the court must consider is clearly defined in the text of that Guideline. Thus, courts can apply the language of sections 2T1.1 and 1B1.3 simultaneously without negating any instruction by the Commission. However, if we were to adopt the dissent's reasoning, we would, in turn, read out the language of section 1B1.3 that instructs courts to consider both actual and intended harm, unless otherwise specified.

In sum, the structure of the Guidelines supports the conclusion that the meaning of “loss” is broad enough to encompass intended loss.⁶ And, at minimum, the dissent's comparison of sections 1B1.3(a) and 2T1.1(c)(1) creates more confusion as to the meaning of the term “loss” in section 2B1.1, supporting our holding that “loss” is genuinely ambiguous.

d.

The purpose of the Guidelines further cuts against Appellant's narrow reading of “loss.” In promulgating the Guidelines, the Commission set out to “establish sentencing policies and practices for the Federal criminal justice system that ․ avoid[ ] unwarranted sentencing disparities among defendants with similar records who have been found guilty of similar criminal conduct.” 28 U.S.C. § 991(b)(1)(B). The Commission relies on the intended loss amount to further this goal where it encompasses both “the seriousness of the offense and the defendant's relative culpability.” Guidelines § 2B1.1 cmt. (background); see id. App. C Supp. at 104–05 (Amend. 792) (noting “the Commission's continued belief that intended loss is an important factor” because it focuses “specifically on the defendant's culpability”).

Were we to accept Appellant's limited reading of “loss,” it could result in sentencing disparities for defendants with similar culpability. And this is contrary to the Guidelines' purpose. For instance, if sentencing courts only considered “actual loss” when deciding the specific offense characteristics, defendants with equal culpability for the same criminal acts would receive longer sentences merely because, for one reason or another, one succeeded in carrying out their fraudulent scheme while the other did not. See United States v. You, 74 F.4th 378, 398 (6th Cir. 2023) (noting potential sentencing disparities if “loss” were to be so limited).

Because no single meaning of “loss” is evident from the plain text of section 2B1.1, even after employing the traditional tools of interpretation, we conclude that a genuine ambiguity exists.

2.

Given that ambiguity exists as to the Guidelines' definition of “loss,” we next consider whether the commentary's definition of “loss” falls within the “zone of ambiguity” such that it should be given deference. Kisor, 588 U.S. at 576, 139 S.Ct. 2400; You, 74 F.4th at 398. In doing so, we ask whether the commentary falls “within the bounds of reasonable interpretation” by the relevant agency; here, the Commission. Id. (internal quotation marks omitted). If it does, we independently inquire as to whether the commentary's “character and context” entitles it to “controlling weight.” Id. To make this assessment, we consider whether the commentary is the Commission's “official position,” “implicate[s] its substantive expertise” in some way, and reflects the “fair and considered judgment” of the Commission such that it is not simply a “convenient litigating position.” Kisor, 588 U.S. at 577–79, 139 S.Ct. 2400 (citations omitted).

The character and context of the commentary entitle it to controlling weight. The commentary's definitions of loss are the Commission's “official position.” Kisor, 588 U.S. at 577, 139 S.Ct. 2400. In 2001, the Commission amended the Guidelines to consolidate the theft and fraud sections into section 2B1.1 (the “2001 Amendment”). Office of Gen. Counsel, U.S. Sent'g Comm'n, Loss Calculations Under § 2B1.1(b)(1), at 1 (2020). Also known as the “Economic Crime Package,” the 2001 Amendment was “the outcome of Commission review of economic crime issues over a number of years.” Office of Educ. & Sent'g Practice, Amendments to the Federal Sentencing Guidelines: Highlights of Key Points [of the 2001 Amendment] 1 (2001). As a part of the Economic Crime Package, the Commission “modified the definition of loss [in the commentary] such that it would be based on reasonably foreseeable pecuniary harm and would include intended loss.” Loss Calculations Under § 2B1.1(b)(1), at 1. Thus, the inclusion of intended loss in the Economic Crime Package has been the longstanding position of the Commission.⁷ And the use of “intended loss” in the commentary was the result of the well-reasoned decision making by the Commission.

Moreover, the commentary reflects the “substantive expertise” of the Commission and its “fair and considered judgment.” Kisor, 588 U.S. at 577, 579, 139 S.Ct. 2400. Each year, the Commission “collects a vast amount of federal sentencing data” and “also reviews and refines the [G]uidelines in light of decisions from courts of appeals, sentencing-related research, congressional action, and input from the criminal justice community.” U.S. Sent'g Comm'n, About Us (2023), https://www.ussc.gov/sites/default/files/pdf/about/overview/2023_About-Us-Trifold.pdf [https://perma.cc/2ZUW-N62T]. Indeed, the Commission conducts ample research on section 2B1.1 crimes and includes its findings on its website. See U.S. Sent'g Comm'n, Quick Facts: Theft, Property Destruction, and Fraud Offenses (2023).

In sum, because the commentary defines loss within the “zone of ambiguity” created by Guidelines § 2B1.1 and because the “character and context” of the commentary support that it is deserving of “controlling weight,” see Kisor, 588 U.S. at 576, 139 S.Ct. 2400, the district court properly deferred to the commentary to determine the appropriate loss amount in Appellant's case.

C.

In conclusion, because a genuine ambiguity exists as to the meaning of “loss” in Guidelines § 2B1.1 and the character of the commentary supports that it is deserving of deference, we affirm the district court's application of the commentary and its reliance on Appellant's full intended loss amount to calculate her Guidelines sentencing range pursuant to section 2B1.1.

IV.

For the foregoing reasons, the judgment of the district court is

AFFIRMED.

Sometimes lawyers and judges overcomplicate things. This is one of those occasions. Today, the majority holds that the meaning of the word “loss” in U.S.S.G. § 2B1.1(b)(1) is ambiguous, such that we should defer to the Guideline commentary's definition of the word that encompasses intended losses. The majority thus concludes that harm a wrongdoer intended to cause but did not achieve is nevertheless a loss. I see no way to read the word “loss,” as used in § 2B1.1(b)(1), that way. “Loss” is not ambiguous. “Loss” means actual loss, not intended loss. Because the Guideline commentary's interpretation of “loss” in § 2B1.1(b)(1) to include intended loss is both inconsistent with and a plainly erroneous reading of the Guideline, no deference to the commentary is warranted. And because the district court applied a § 2B1.1(b)(1) enhancement based on intended loss instead of actual loss, I would vacate the sentence and remand for resentencing with instructions to the district court to calculate Boler's Guidelines range without considering intended loss for purposes of a § 2B1.1(b)(1) enhancement. I respectfully dissent.

I.

This appeal presents a simple question: does the word “loss” in § 2B1.1(b)(1) include tax refunds that Boler intended but failed to fraudulently obtain, or is it limited to the actual funds she fraudulently obtained from the government? The answer is also simple. “Loss” in § 2B1.1(b)(1) means only actual, not intended, loss.

A. Ordinary Meaning of the Text

We begin with the text of § 2B1.1(b)(1). See United States v. Skinner, 70 F.4th 219, 230 (4th Cir. 2023) (recognizing that “[c]ourts interpret the Sentencing Guidelines according to the ordinary rules of statutory construction” (internal quotations omitted)); Barnhart v. Sigmon Coal Co., 534 U.S. 438, 450, 122 S.Ct. 941, 151 L.Ed.2d 908 (2002) (“As in all statutory construction cases, we begin with the language of the statute.”). Section 2B1.1(b)(1) instructs courts to apply a sentencing enhancement “[i]f the loss exceeded $6,500,” followed by a table of enhancements corresponding to incrementally increasing loss amounts. U.S.S.G. § 2B1.1(b)(1) (emphasis added). The Guideline itself does not define “loss.” Even so, the ordinary meaning of the term is clear. “Loss” refers to actual losses, not purely intended ones.¹

To illustrate, imagine Jonathan overhears Julia telling her friends that she keeps $1,000 in cash in her closet for emergencies. Jonathan then breaks into Julia's home, intending to steal the $1,000 that he believes, based on what he overheard, Julia keeps in a closet. If Jonathan indeed finds $1,000 and flees with it, and the police ask Julia the amount of her loss, she will say $1,000. But what if while trying to gather up the $1,000, the sound of a security alarm startles Jonathon into fleeing with only $500, leaving the other $500 behind? If asked the amount of her loss, Julia would not say $1,000 just because Jonathan may have intended to steal that amount. She would say that her loss was the amount Jonathan actually stole.

1. Dictionaries

Dictionaries confirm this understanding. In 1987, when the Guidelines were enacted and first used “loss” in the context of sentencing enhancements for economic offenses, see U.S.S.G. §§ 2B1.1(b)(1), 2F1.1 (1987), Merriam-Webster defined “loss” as “the act of losing; that which is lost; defeat; diminution; bereavement; harm; waste by escape or leakage; [or] number of casualties suffered in war.” Loss, Webster's Dictionary (1987). Similarly, Collins Dictionary defined “loss” to include “bereavement, deprivation, disappearance, failure, forfeiture, losing, misfortune, mislaying, privation, squandering, [or] waste,” as well as “cost, damage, defeat, destruction, detriment, disadvantage, harm, hurt, impairment, injury, [or] ruin.” Loss, The New Collins Dictionary and Thesaurus in One (1987). All these definitions involve actual loss. None involve purely intended loss.

As the majority recognizes, the definitions of “loss” include a wide variety of categories. Maj. Op. at –––– – ––––. I agree that “loss” is a broad term. But breadth does not equal ambiguity. To explain, consider “vehicle.” Cambridge Dictionary defines “vehicle” as “a machine, usually with wheels and an engine, used for transporting people or goods, especially on land.”² Under that definition, cars, trucks, vans and buses are vehicles. Other dictionaries' definitions of “vehicle” are not limited to land. For example, Merriam-Webster defines “vehicle” as “a means of carrying or transporting something.”³ So, under that definition, “vehicle” may include planes, helicopters and boats. But the word is not ambiguous just because there are many different types of vehicles.

Even though the government insists “loss” is ambiguous, all the definitions that it—and the majority—cite are consistent. Each involves an event or state of being that has actually happened. Whether economic, physical or emotional, each definition concerns an actual, realized loss. None suggest that “loss” includes harm that was intended but never realized.

Resisting that conclusion, the government contends that three of the modern ⁴ dictionary definitions it cites cover intended loss. These three definitions are (1) “the partial or complete deterioration or absence of a physical capability or function”; (2) “the harm or privation resulting from losing or being separated from someone or something”; and (3) “a failure to gain, win, obtain, or utilize.” Resp. Br. at 14 (quoting Loss, Merriam-Webster (2023)). The government argues that the “absence of a physical capability,” such as the loss of the ability to play a sport, is not an actual, measurable loss. Id. And it contends that “the harm or privation resulting from losing or being separated from someone or something” is “a reflection of an unmeasurable, emotional reaction.” Id. As for the “failure to gain, win, obtain, or utilize,” the government insists that such a loss “would clearly be reflected by the IRS's inability to gain or obtain proper tax revenue.” Id.

I fail to see how those arguments help the government. The first two examples the government offers involve intangible losses. But intangible losses are still actual losses. Take, for instance, the “absence of a physical capability.” An accident that results in injuries that render a person unable to walk would fall under that definition. The inability to walk is an actual loss, as it actually impairs the person's physical capabilities. Such an impairment may decrease the person's quality of life. While that type of loss is intangible and difficult to measure with a monetary value, juries value such intangible losses all the time. The intangibility of such a loss does not make it any less real. It may be true that § 2B1.1(b)(1) concerns only tangible losses. But even so, the issue we confront is not whether “loss” includes intangible losses. Our question is whether “loss” includes purely intended losses. None of the definitions cited by the government encompass such losses.

At first blush, the government's third example—“a failure to gain, win, obtain, or utilize”—sounds the most like an intended loss. But the government's use of this definition views the failure to gain from the vantagepoint of the perpetrator. In contrast, the Guidelines view “loss” from the perspective of the victim. So, the government fails to show how this definition encompasses purely intended loss in the context of the Guidelines.

Last, the government offered a hypothetical at oral argument in an attempt to show that “loss” encompasses intended losses. The government asked us to imagine that an investor purchased $1,000 in stock, and the stock immediately decreased in value to $800. The government asserted that the investor might describe his stock purchase as a $200 loss even if he had not yet sold the stock and realized that loss. But there are at least two problems with this example. First, at the time the stock decreased in value to $800, the investor would have an actual loss. At that moment, the value of his stock portfolio decreased. Second, the investor did not intend any loss. So, the government's hypothetical does not help resolve the question here—whether “loss” includes harm that a wrongdoer sought to inflict but did not achieve.

2. Corpus Linguistics

a.

Corpus linguistics—another tool for interpreting the meaning of words in a statute (or other text)—confirms “loss” is not ambiguous. “Corpus linguistics is the study of language (linguistics) through systematic analysis of data derived from large databases of naturally occurring language (corpora, the plural of corpus, a body of language).” Thomas R. Lee & James C. Phillips, Data-Driven Originalism, 167 U. PA. L. REV. 261, 289 (2019). In layman's terms, corpus linguistics clarifies a term's meaning by studying the term's use across specified time periods and origins. “The uses can then be reviewed one by one, in their context, to determine” the term's meaning. United States v. Rice, 36 F.4th 578, 583 n.6 (4th Cir. 2022). When the uses are reviewed as a whole, “a broad picture of how a word or phrase was customarily used and understood during a specific time period can emerge.” Id.

To conduct a corpus linguistics analysis of “loss,” I consulted one of the online databases of naturally occurring language, the Corpus of Historical American English (“COHA”).⁵ COHA contains over 475 million words of text derived from fiction, popular magazines, newspapers, academic texts, and TV and movie subtitles from the 1820s to the 2010s. See Corpus of Historical American English, https://www.english-corpora.org/coha/ [https://perma.cc/HK78-6N95] (last visited July 1, 2024).

I began by searching the term “loss” in COHA. I limited my search results to uses of the term between 1984 and 1990, the years surrounding the 1987 enactment of the Guidelines, which used “loss” in the context of sentencing enhancements for economic offenses. See U.S.S.G. §§ 2B1.1(b)(1), 2F1.1 (1987). I obtained 1,434 hits for uses of “loss” during this time period. I then utilized COHA's random sample function to pull 100 hits from that total.⁶ I reviewed each of the 100 randomly selected hits and found that every instance of “loss” referred to an actual loss.

For example, the random sample included sixteen hits of newspaper or magazine sources using “loss” when reporting decreases relating to entities' revenues. Another twelve hits consisted of newspaper, magazine, fiction or television sources using “loss” in the context of a diminution in a physiological or psychological function, such as memory loss, or the removal of a body part, such as the loss of an ear. Ten hits represented newspaper, magazine, fiction, academic texts or television sources using “loss” as a synonym for death. And nine hits represented newspaper or magazine sources referring to “loss” when reporting the outcome of a given sporting event between two teams. Other hits within the sample consisted of sources using “loss” when referring to an emotional state, such as feeling at a loss, or when referring to deprivations or decreases in a variety of other contexts, such as animals' loss of habitat or a driver's loss of license. But none of the hits referred to intended but unrealized loss.⁷

This confirms what we knew at the start. If ordinary speakers of the English language say “loss,” they mean something that actually occurred. Given its plain and ordinary meaning, “loss” as used in § 2B1.1(b)(1) means only actual loss. That is how dictionaries use the term, and that is how people in the real world have used the term, as confirmed by a corpus linguistics analysis. No matter how you slice it, the ordinary meaning of “loss” does not encompass intended loss.

b.

Without challenging any specifics of my analysis, the majority dismisses corpus linguistics generally. It resists my statement that corpus linguistics is another tool for interpreting words and phrases, asserting that corpus linguistics “is not so benign.” Maj. Op. at –––– n.5. It adds that corpus linguistics “can be quite a flawed tool; a controversial tool at best.” Id. And it argues that “corpus linguistics largely ‘ignores the crucial contexts in which legal language is produced, interpreted, and deployed.’ ” Id. (quoting Anya Bernstein, Legal Corpus Linguistics and the Half-Empirical Attitude, 106 Cornell L. Rev. 1397, 1397 (2021)). Last, the majority concludes that “[a] survey of novels, magazines, newspapers, popular television shows and movies is no match for legal analysis of the text, structure, purpose and history of the actual regulation we are considering.” Id.

I recognize that not all academics and jurists have embraced corpus linguistics as an analytical tool. And I welcome debate on the topic, as healthy and respectful discussion about important ideas elevates our legal analysis. In that spirit, I respond with a few points.

First, it seems hard to dispute that corpus linguistics is gaining traction as an interpretive tool. Corpus linguistics finds support in opinions written by former Supreme Court Justice Breyer and current Justices Thomas and Barrett.⁸ Also, a number of state and federal courts around the country, including this court, have utilized corpus linguistics when discerning the meaning of disputed terms.⁹

But admittedly, popularity does not equate to efficacy. See Stinnie v. Holcomb, 77 F.4th 200, 230 (4th Cir. 2023) (Quattlebaum, J., dissenting) (noting that “as we learned as children, just because others are doing something, that does not make it right.”), cert. granted sub nom. Lackey v. Stinnie, ––– U.S. ––––, 144 S. Ct. 1390, 218 L.Ed.2d 678 (2024). So, as to the majority's specific critique about context, corpus linguistics does not ignore context; it embraces it. In fact, corpus linguistics, which surveys the various ways speakers and writers actually used the word or phrase at issue, provides real world examples that reveal how context can inform meaning. Assume a word or phrase has different meanings depending on the context. A corpus linguistics search should identify the different contexts in which the word or phrase is used. That should help a court compare the context of the statute, regulation or contract at issue with the different contexts identified in the corpus linguistics search.

Professors Lee ¹⁰ and Mouritsen have made this same point:

[C]orpus analysis does not take place in an acontextual vacuum. A corpus-based approach to ordinary meaning, as noted, does not simply evaluate which of two competing uses is the most common. Instead, the corpus allows us to examine the use of a word or phrase in a particular syntactic, semantic, or pragmatic context, in the speech or writing of a particular speech community or register, and at a particular point in time․ we are often able to find not only common ways to describe common real-world occurrences, but also the most common ways in which highly particularized and highly contextualized occurrences are described in a given speech community at a given point in time. If there are cases where it is natural to use a particular expression, but the circumstances do not arise often [cite] (internal citations omitted), an appropriately designed corpus search (performed in a sufficiently robust corpus) will help us identify these instances and make informed, evidence-based judgment about them.

Thomas R. Lee & Stephen C. Mouritsen, Judging Ordinary Meaning, 127 Yale L.J. 788, 874–75 (2018) (internal citations omitted). So understood, corpus linguistics does not ignore context; it is an additional tool to examine it.

To illustrate, take this case. I have provided the context for hits I found in my search of “loss.” If in certain contexts “loss” includes “intended loss,” one would expect that to show up in the way ordinary writers and speakers used the word. Here, however, out of the 100 randomly selected examples, none used “loss” to refer to “intended loss.” That convinces me that if a writer or speaker means intended loss, they pair “loss” with “intended,” “anticipated,” “expected” or some similar adjective. And since § 2B1.1(b) did not do that, “loss” means “actual loss.”

The majority states that “[a] survey of novels, magazines, newspapers, popular television shows and movies is no match for legal analysis of the text, structure, purpose and history of the actual regulation we are considering.” Maj. Op. at –––– n.5. But corpus linguistics is not a rival to text, structure, purpose or history. In fact, I considered those factors—in addition to the regulatory context—in my opinion. Instead, corpus linguistics compliments those considerations, much like dictionaries. The majority's criticisms of corpus linguistics could equally be lodged at dictionaries. Yet no one bats an eye at considering dictionaries to assist in interpreting language. In fact, the majority does so in its opinion. If our goal is to discern the ordinary public meaning of a word or phrase—as we all seem to agree it is—examining the ways the public actually used the words and phrases under consideration is worth a look.

B. Kisor

Perhaps recognizing the weakness of its position on ordinary meaning, the government contends that we should not confine our interpretation of loss to the plain meaning of the term. It argues that Kisor v. Wilkie, 588 U.S. 558, 139 S.Ct. 2400, 204 L.Ed.2d 841 (2019) instructs courts to “carefully consider the text, structure, history, and purpose of a regulation, in all the ways it would if it had no agency to fall back on.” Id. at 575, 139 S.Ct. 2400 (cleaned up). Considering the structure and purpose of § 2B1.1(b)(1) and the Guidelines as a whole, the government says “loss” is ambiguous. To the extent that Kisor applies (see infra Part I.C), I disagree.

In Kisor, the Supreme Court explained that “the possibility of deference can arise only if a regulation is genuinely ambiguous.” 588 U.S. at 573, 139 S.Ct. 2400 (emphasis added). A regulation is genuinely ambiguous only if uncertainty exists “even after a court has resorted to all the standard tools of interpretation.” Id. (emphasis added). “If uncertainty does not exist” as to the meaning of a regulation, “there is no plausible reason for deference” to an agency's interpretation of it. Id. at 574–75, 139 S.Ct. 2400. “The regulation then just means what it means,” and a court of law must give it effect. Id. That is what we have here. The meaning of the word “loss” is clear, as confirmed by the application of standard tools of interpretation. “Loss” means actual losses, not intended ones. Yet, the government asks us to scour the Guidelines to create an ambiguity that does not exist.

Considering the structure, purpose and history of the Guidelines in addition to the plain text, there is no genuine ambiguity as to the meaning of “loss” in § 2B1.1(b)(1). For instance, with respect to the structure of the Guidelines, the government contends that § 1B1.3—the Guideline that defines relevant conduct for all offenses—supports finding “loss” as used in § 2B1.1(b)(1) to be ambiguous. “Unless otherwise specified,” § 1B1.3 directs courts to determine a defendant's base offense level based on, among other things, “all harm that resulted from the acts and omissions ․ and all harm that was the object of such acts and omissions.” U.S.S.G. § 1B1.3(a)(3). Because “loss” in § 2B1.1(b)(1) is the “harm” for this criminal conduct, the government insists that § 1B1.3 directs that we consider both the “harm that resulted from” the criminal acts (actual loss) and “all harm that was the object” of the criminal acts (intended loss). Id.

But a look at other portions of the Guidelines undermines this position. Consider § 2T1.1, the Guideline for determining the base offense level for tax evasion. Section 2T1.1(c)(1) itself defines “tax loss” as “the total amount of loss that was the object of the offense (i.e., the loss that would have resulted had the offense been successfully completed).” If § 1B1.3's directive to consider the harm that was the object of an offense when determining base offense levels is understood to expand the meaning of “loss” in the Guidelines governing base offense levels, § 2T1.1(c)(1)'s definition of “tax loss” would be superfluous. There would be no need to define “tax loss” to include the loss that was the object of the offense because, if the government was right, § 1B1.3 would already require the consideration of that sum, unless otherwise specified. I would not interpret the Guidelines in a way that renders portions superfluous. See United States v. Haas, 986 F.3d 467, 479 (4th Cir. 2021) (applying canon against superfluity when interpreting Guidelines provision). So, in my view, § 1B1.3's directive regarding the consideration of harm does not expand the meaning of “loss” as used in § 2B1.1(b)(1).¹¹

As for the purpose of the Guidelines, the majority is correct that the Sentencing Commission promulgated the Guidelines to “avoid[ ] unwarranted sentencing disparities among defendants with similar records who have been found guilty of similar criminal conduct.” 28 U.S.C. § 991(b)(1)(B). But I disagree with the majority's view that limiting “loss” in § 2B1.1(b)(1) to actual losses would result in unwarranted sentencing disparities. Maj. Op. at ––––. In the majority's view, “defendants with equal culpability for the same criminal acts would receive longer sentences merely because, for one reason or another, one succeeded in carrying out their fraudulent scheme while the other did not.” Maj. Op. at ––––. Such sentencing disparities for basic economic offenses, however, are not unwarranted. Consider two individuals who each intend to obtain $1,000,000 in fraudulent bank transfers. One individual successfully obtains $1,000,000, while the other individual only obtains $1,000. Despite their identical intentions, they did not cause identical financial injuries. Both are guilty of committing the same substantive offense, but applying different sentencing enhancements under § 2B1.1(b)(1) based on the actual losses they caused does not result in an unwarranted sentencing disparity. Quite the opposite; the differing financial injuries is what warrants the application of different sentencing enhancements.

Also, the Sentencing Commission created the Guidelines to “provide certainty and fairness in meeting the purposes of sentencing.” 28 U.S.C. § 991(b)(1)(B). Basing the application of § 2B1.1(b)(1) enhancements on intended losses—but only if greater than actual losses—promotes neither certainty nor fairness. With respect to certainty, a defendant is left to wonder how a court will calculate his subjective intended loss. As for fairness, few would find it fair to apply the same loss enhancement to two individuals who caused vastly different financial injuries. So, like the structure of the Guidelines, the purpose of the Guidelines supports reading “loss” in § 2B1.1(b)(1) to mean actual loss.

Finally, the Guidelines' history supports this reading, as well. When the Guidelines were first enacted in 1987, the application of § 2B1.1(b)(1)'s loss enhancement was limited to offenses of “Larceny, Embezzlement, and Other Forms of Theft.” U.S.S.G. § 2B1.1(b)(1) (1987). Under this initial version of § 2B1.1(b)(1), “[i]f the value of the property taken exceeded $100,” a court was to increase a defendant's offense level based on the amount of “Loss.” Id. Separately, the 1987 version of § 2F1.1(b)(1) provided the loss enhancement for offenses involving “Fraud and Deceit.” U.S.S.G. § 2F1.1(b)(1) (1987). Contrary to § 2B1.1(b)(1), § 2F1.1(b)(1) explicitly instructed courts to apply sentencing enhancements for offenses involving fraud and deceit “[i]f the estimated, probable or intended loss exceeded $2,000.” Id. (emphasis added).

In 2001, the Sentencing Commission amended the Guidelines, deleting § 2F1.1 entirely and amending § 2B1.1 to encompass basic economic offenses beyond larceny, embezzlement and other forms of theft. See U.S.S.G. § 2B1.1 (2001). As such, § 2B1.1(b)(1)'s loss enhancement applied to what § 2F1.1(b)(1)'s loss enhancement once covered—offenses involving fraud and deceit. See id. Yet, in deleting § 2F1.1(b)(1) and amending § 2B1.1(b)(1), the Sentencing Commission did not include any “intended loss” language in § 2B1.1(b)(1). Instead, the amended version of § 2B1.1(b)(1) instructed courts, “[i]f the loss exceeded $5,000, increase the offense as follows ․” U.S.S.G. § 2B1.1(b)(1) (2001). Aside from increases in the monetary amount, this wording has lived on in subsequent versions of the Guidelines. Indeed, as previously noted, it is the language of the version at issue in this case and the present version. See U.S.S.G. § 2B1.1(b)(1) (2021) and id. (2023).

Since-deleted § 2F1.1(b)(1) demonstrates that the Sentencing Commission has previously specified in the text of a Guideline when “loss” encompasses intended loss. The Sentencing Commission's use of “intended loss” in § 2F1.1(b)(1) and only “loss” in § 2B1.1(b)(1) in early versions of the Guidelines should be presumed intentional. Cf. Russello v. United States, 464 U.S. 16, 23, 104 S.Ct. 296, 78 L.Ed.2d 17 (1983) (explaining that when “Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.” (internal quotations omitted)). Even after deleting § 2F1.1(b)(1) and expanding § 2B1.1(b)(1) to cover what § 2F1.1(b)(1) once did, the Sentencing Commission never amended the text of § 2B1.1(b)(1) to instruct courts to consider intended losses. This history of the Guidelines therefore also supports reading “loss” in § 2B1.1(b)(1) to encompass only actual losses.

C. Campbell and Moses

Last, we need not wade into the question of whether United States v. Campbell, 22 F.4th 438 (4th Cir. 2022) and United States v. Moses, 23 F.4th 347 (4th Cir. 2022) conflict with respect to how Kisor and Stinson v. United States, 508 U.S. 36, 113 S.Ct. 1913, 123 L.Ed.2d 598 (1993) apply to Guideline interpretation. That is because under either Kisor or Stinson, the commentary's interpretation of loss is not entitled to deference. Because “loss” as used in § 2B1.1(b)(1) is not genuinely ambiguous, Kisor would preclude deference to the commentary's interpretation of the term. See 588 U.S. at 573, 139 S.Ct. 2400 (holding that “the possibility of deference can arise only if a regulation is genuinely ambiguous”). And because the commentary's interpretation of “loss” to include intended loss is “plainly erroneous or inconsistent with” § 2B1.1(b)(1)'s use of the term, Stinson would likewise preclude deference. 508 U.S. at 45, 113 S.Ct. 1913; see also id. at 38, 113 S.Ct. 1913 (holding that commentary “that interprets or explains a guideline is authoritative unless it violates the Constitution or a federal statute, or is inconsistent with, or a plainly erroneous reading of, that guideline”). So, regardless of whether Stinson or Kisor governs our analysis, “loss” in § 2B1.1(b)(1) still means what it is commonly understood to mean—actual loss. It does not include losses a wrongdoer intended, but failed, to cause.

Of course, the Sentencing Commission has every right to define “loss” to include intended losses, notwithstanding the way the word is commonly understood. But it has only defined the term to include intended loss in the commentary to § 2B1.1(b)(1). Unless and until it does so in § 2B1.1(b)(1) itself, we should not defer to the commentary's definition. The term “loss” as used in § 2B1.1(b)(1) unambiguously means actual losses. It does not include purely intended losses.

II.

For the foregoing reasons, I would vacate Boler's sentence and remand for resentencing. I respectfully dissent.

DISSENT APPENDIX

FOOTNOTES

1.   In 2020, Congress passed the Coronavirus Aid, Relief, and Economic Security (“CARES”) Act, Pub. L. 116-136, § 1102, 134 Stat. 285, 286 (2020). Section 1102 of the CARES Act created the Paycheck Protection Program to grant forgivable loans to small business owners for certain expenses. Id.

2.   Citations to the “J.A.” refer to the Joint Appendix filed by the parties in this appeal.

3.   Bowles v. Seminole Rock & Sand Co., 325 U.S. 410, 414, 65 S.Ct. 1215, 89 L.Ed. 1700 (1945); Auer v. Robbins, 519 U.S. 452, 461, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997).

4.   The Supreme Court's recent ruling in Loper Bright Enterprises v. Raimondo, ––– U.S. ––––, 144 S. Ct. 2244, ––– L.Ed.2d –––– (2024), calls into question the viability of Auer deference. In Loper Bright, the Court rejected the long-standing notion that “ambiguities” in statutory language act as “implicit delegations to agencies.” Id. at 2265. Since Loper Bright dealt specifically with ambiguities in statutory directives to agencies and did not address the issue of agency interpretations of their own regulations, we will apply the Supreme Court's recent guidance in Kisor to address the issue before us today.

5.   We pause to address the use of corpus linguistics in the dissent's analysis. While the dissent views corpus linguistics as simply “another tool for interpreting the meaning of words in a statute (or other text),” it is not so benign. Post at ––––. Indeed, legal corpus linguistics can be quite a flawed tool; a controversial tool at best. See Anya Bernstein, What Counts as Data?, 86 Brook. L. Rev. 435 (2021); Anya Bernstein, Legal Corpus Linguistics and the Half-Empirical Attitude, 106 Cornell L. Rev. 1397 (2021) (hereinafter Bernstein, Legal Corpus Linguistics); Mark W. Smith & Dan M. Peterson, Big Data Comes for Textualism, 70 Drake L. Rev. 387 (2021); John S. Heretic, Against Corpus Linguistics, 108 The Geo. L.J. Online 51. The bottom line is that context matters and legal corpus linguistics largely “ignores the crucial contexts in which legal language is produced, interpreted, and deployed.” Berstein, Legal Corpus Linguistics, at 1397. A survey of novels, magazines, newspapers, popular television shows and movies is no match for legal analysis of the text, structure, purpose and history of the actual regulation we are considering.

6.   In fact, the Sentencing Commission has recently promulgated amendments to the Guidelines that will resolve this issue going forward. The amendments will incorporate the intended loss commentary, Guidelines § 2B1.1 cmt. 3(A), into the Guidelines' text. 89 Fed. Reg. 36,853, 36,855–36,857 (May 3, 2024) (effective Nov. 1, 2024, absent contrary action by Congress). Thus, the text of the Guideline itself will define “loss” as the greater of the actual loss or intended loss.

7.   The dissent ignores these sources, wherein, for example, the Office of General Counsel to the Commission clearly states that the Commission intended to retain the use of intended loss in the application of Guidelines § 2B1.1 when consolidating the theft and fraud Guidelines under the 2001 Amendment. While the dissent argues that the “intended loss” language previously contained in section 2F1.1 was intentionally removed by the 2001 Amendment, these sources indicate otherwise. Indeed, the Commission's Office of Education and Sentencing Practices highlights this point -- “[t]he revised definition of loss retains the rule that loss is the greater of actual and intended loss.” Highlights of Key Points [of the 2001 Amendment] 1 (emphasis supplied).

1.   Like the parties, the majority frames the interpretation of “loss” as seeking its ordinary meaning. Maj. Op. at –––– – ––––. But “[s]ometimes context indicates that a technical meaning applies. Every field of serious endeavor develops its own nomenclature—sometimes referred to as terms of art.” Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 73 (2012) (emphasis in original). There is an unsettled question of whether the Guidelines' primary audience is ordinary citizens, whose criminal sentences are affected by them, or lawyers and judges, who use and apply them. See United States v. Seefried, 639 F. Supp. 3d 8, 14 (D.D.C. 2022). Following the lead of the parties and the majority, I have evaluated “loss” in terms of its ordinary meaning. But even if “loss” is a legal term of art when used in the Guidelines, I get to the same result. Around the time the Guidelines first used “loss,” Black's Law Dictionary defined over 30 types of loss, including “capital loss,” “economic loss” and “passive loss.” See Black's Law Dictionary (6th ed. 1990). None of the defined types of loss refer to purely intended loss; they represent a broad variety of actual losses. Even “unrealized loss,” which might sound like it refers to a purely intended loss, does not cover intended losses when one looks to the definition of the term. See id. (defining “unrealized loss” as a “loss that has not yet materialized” (emphasis added)). So, regardless of whether “loss” is a legal term of art, it means only actual loss.

2.   Vehicle, Cambridge Dictionary, https://dictionary.cambridge.org/dictionary/english/vehicle [https://perma.cc/F6V26DZZ] (last visited July 1, 2024).

3.   Vehicle, Merriam-Webster, https://www.merriam-webster.com/dictionary/vehicle [https://perma.cc/W7X9-3XHE] (last visited July 1, 2024).

4.   The government relies on modern definitions of “loss.” However, our point of focus is the meaning of the term when it was first used in the Guidelines for sentencing enhancements for economic offenses. See United States v. Hill, 963 F.3d 528, 532 (6th Cir. 2020) (“And we look to dictionaries that defined the relevant words when the Sentencing Commission initially used them in 1987.”). This is the same way we interpret statutes. See Niz-Chavez v. Garland, 593 U.S. 155, 160, 141 S.Ct. 1474, 209 L.Ed.2d 433 (2021); Wis. Cent. Ltd. v. United States, 585 U.S. 274, 284, 138 S.Ct. 2067, 201 L.Ed.2d 490 (2018); Sandifer v. U.S. Steel Corp., 571 U.S. 220, 227–28, 134 S.Ct. 870, 187 L.Ed.2d 729 (2014). As explained above, “loss” was used in that context in the 1987 Guidelines. Still, we consider the government's modern definitions for argument's sake.

5.   Another valuable database is the Corpus of Contemporary American English (“COCA”). Encompassing sources of contemporary English originating between 1990 and 2019, COCA “contains over one billion words of text ․ from eight genres: spoken, fiction, popular magazines, newspapers, academic texts, TV and movie subtitles, blogs, and other web pages.” Corpus of Contemporary American English, https://www.english-corpora.org/coca/ [https://perma.cc/9JQP-9M42] (last visited Ju1y 1, 2024). But because the Guidelines' first use of “loss” in the context of sentencing enhancements for economic offenses predates this database's sources, I did not use COCA here.

6.   COHA allows a user to retrieve a random sample of 100, 200, 500 or 1,000 results.

7.   The random sample I retrieved from COHA is attached to this opinion as an appendix. See Dissent Appendix.

8.   See Muscarello v. United States, 524 U.S. 125, 129–30, 118 S.Ct. 1911, 141 L.Ed.2d 111 (1998); N.Y. State Rifle & Pistol Ass'n, Inc. v. Bruen, 597 U.S. 1, 109–10, 142 S.Ct. 2111, 213 L.Ed.2d 387 (2022) (Breyer, J., dissenting); Carpenter v. United States, 585 U.S. 296, 347 & n.4, 138 S.Ct. 2206, 201 L.Ed.2d 507 (2018) (Thomas, J., dissenting); Moore v. United States, ––– U.S. ––––, 144 S. Ct. 1680, 1701, 219 L.Ed.2d 275 (2024) (Barrett, J., dissenting).

9.   See, e.g., Rice, 36 F.4th at 583 n.6 (using corpus linguistics to determine the meaning of “strangulation”); Waetzig v. Halliburton Energy Servs., Inc., 82 F.4th 918, 927 n.12 (10th Cir. 2023) (“final proceeding”); Fulkerson v. Unum Life Ins. Co. of Am., 36 F.4th 678, 682–83 (6th Cir. 2022) (“reckless driving”); United States v. Seefried, 639 F. Supp. 3d 8, 13–16 (D.D.C. 2022) (“administration of justice”); Matthews v. Indus. Comm'n of Ariz., 254 Ariz. 157, 520 P.3d 168, 175–77 (2022) (“injury by accident”); In re Estate of Heater, 498 P.3d 883, 890–91 (Utah 2021) (“natural parent”); State v. Misch, 214 Vt. 309, 256 A.3d 519, 529–30 (2021) (“bear arms”); Wilson v. Safelite Grp., Inc., 930 F.3d 429, 439–45 (6th Cir. 2019) (Thapar, J., concurring in part and in the judgment) (“results in” and “extending to”); Richards v. Cox, 450 P.3d 1074, 1079–81 (Utah 2019) (“employment in”); People v. Harris, 499 Mich. 332, 885 N.W.2d 832, 839 (2016) (“information”).

10.   Professor Lee served on the Utah Supreme Court from 2010 to 2022.

11.   The majority reasons that § 1B1.3(a)'s “[u]nless otherwise specified” language leads to a different conclusion. According to the majority, § 1B1.3(a) instructs sentencing courts as to how to interpret “harm” as used throughout the Guidelines when the definition of harm is not “otherwise specified.” So, § 2T1.1(c)(1)'s definition of “tax loss” as “the total amount of loss that was the object of the offense” is simply an example of an otherwise specified situation. Thus, there is no superfluity. But “otherwise” suggests different. And § 2T1.1(c)(1)'s definition of “tax loss” is not different from the government's reading of § 1B1.3(a); it is the essentially same and, in my view, would not be needed—and thus is rendered superfluous—if the government's reading of § 1B1.3(a) supports its interpretation of § 2B1.1(b)(1).

THACKER, Circuit Judge:

Affirmed by published opinion. Judge Thacker wrote the opinion in which Judge Benjamin joined. Judge Quattlebaum wrote a dissenting opinion.