Loss amount in white collar matters plays center stage at federal sentencing. Unfortunately, over the years its framework, USSG § 2B1.1(b)(1), has undergone repeated amendments, continuing to balloon the outsized role of loss amount in sentencings for financial crimes. [1]
Section 2B1.1(b)(1) dominates the sentencing of financial crimes, assigning large, tiered enhancements based on loss amount. Its goliath role in sentencing is evident, where the highest tier enhancement alone places a defendant at an offense level of 36, a staggering 84 percent of the way down the sentencing table. [2]
The initial fraud Guideline was located in Section 2F1.1. As with other portions of the Guidelines, the Sentencing Commission (“Commission”) was tasked with collecting data on past sentencings to craft uniform advisory sentences that avoid disparities. [3] In preparing this initial loss table, the Commission reviewed presentence reports from 10,000 cases. [4] However, as to the fraud table, the Commission abandoned its standard practice of rooting sentencing ranges in empirical data. The Commission not only excluded fifty percent of the data at the outset—notably all probationary sentences—but also recommended more severe sentences than the mean of the remaining custodial sentences. [5] However, these initial deviations from the empirical data were modest in their draconian effect when compared to later amendments that leave us with the modern version of Section 2B1.1(b)(1). [6]
In 1989—a scant two years after the Guidelines went into effect—levels were increased for higher-end loss amounts. [7] Under the original version, the highest bracket was a loss over $5,000,000, garnering an increase of 11 levels. [8] In its first reformulation, loss over $5,000,000 bumped to a 14-level enhancement, with four additional brackets culminating with loss over $80,000,000 at an 18-level enhancement—a seven-level increase over the original. [9]
Then, in 2001, the brackets were amended a second time, moving further away from the empirical data. [10] Most importantly, Amendment 617 changed the tiered loss brackets from increasing enhancements by one level at each tier to increasing by two levels per tier. [11] At this stage, it is worth comparing the 2001 version of the table to its original counterpart (again, which itself departed from the empirical data):
| 1987 Version | 2001 Version | |||
| Loss | Increase | Loss | Increase | |
| $2,000 or less | No increase | $5,000 or less | No increase | |
| More than $2,000 | +1 | More than $5,000 | +2 | |
| More than $5,000 | +2 | More than $10,000 | +4 | |
| More than $10,000 | +3 | More than $30,000 | +6 | |
| More than $20,000 | +4 | More than $70,000 | +8 | |
| More than $50,000 | +5 | More than $120,000 | +10 | |
| More than $100,000 | +6 | More than $200,000 | +12 | |
| More than $200,000 | +7 | More than $400,000 | +14 | |
| More than $500,000 | +8 | More than $1,000,000 | +16 | |
| More than $1,000,000 | +9 | More than $2,500,000 | +18 | |
| More than $2,000,000 | +10 | More than $7,000,000 | +20 | |
| More than $5,000,000 | +11 | More than $20,000,000 | +22 | |
| More than $50,000,000 | +24 | |||
| More than $100,000,000 | +26 | |||
At first blush, one might posit that this increase reflects the need to adjust the chart to match inflation after 14 years. However, it was a later 2015 amendment that accounted for inflation, not prior amendments. [12] Contrarily, neither Amendment 154 (in 1989) nor Amendment 617 (in 2001) make any reference to inflation. The 2015 inflation-adjusted table is the version in effect today. This ballooning of the loss amount across earlier Amendments is not indicative of inflation moving at a glacial pace, rather these Amendments are a sea change—an unending deluge that began with the Commission’s initial departure from its traditional empirical approach.
To fully understand the current outsized impact of loss amount, it may be best understood by comparing the Guidelines of Section 2B1.1 to other crimes.
Here is a hypothetical financial fraud calculation: a loss amount over $25 million, with a four-level leadership enhancement, a one-level laundering enhancement, and full acceptance of responsibility—final offense level of 30.
This same offense level 30 is assigned to the following defendants (all assuming a three-point reduction for acceptance of responsibility):
- Conspiracy to commit murder, with no role enhancement (USSG § 2A1.5); [13]
- Drug trafficking conspiracy involving 1 kilogram of fentanyl, who was in possession of a firearm and engaged in laundering (USSG § 2D1.1); [14] and
- Kidnapper who abducted a victim for more than seven days (USSG § 2A4.1). [15]
The overbearing impact of loss amount produces similarly incongruous results when comparing a first-time offender range (in our hypothetical, 97 to 121 months) to the same range assigned to repeat offenders:
- Criminal History Category III defendant sentenced for a drug trafficking conspiracy involving 1 kilogram of fentanyl and money laundering (USSG § 2D1.1); [16] and
- Criminal History CategoryIIoffender sentenced for attempted murder, resulting in serious bodily injury, where offender selected the victim based on race (USSG § 2A2.1). [17]
Loss amount’s ponderous departure from the empirical data also implicates the sentencing factors under 18 U.S.C. § 3553(a). By treating first-time, non-violent offenders convicted of financial crimes with the same severity as repeat, violent offenders, the nature and circumstances of the crime are wildly misstated. Similarly, this apples-to-oranges comparison fails to properly account for the history and characteristics of these two vastly divergent categories of defendants. In one category, you have non-violent offenders at minimal risk of recidivism, with no criminal history; while in the other, you have offenders who have posed—and in all likelihood will continue to pose—a violent threat to society, regardless of prior sentences imposed.
On the need to impose just punishment, surely the framers of the Guidelines viewed crimes like armed fentanyl trafficking and murder conspiracies to be more insidious and worthy of heftier punishment than non-violent financial crimes—yet the Guidelines themselves do not.
On deterrence, first-time, financial crime offenders are much more easily deterred (generally and specifically) than repeat drug and violent offenders—but yet again, the Guidelines call for analogously harsh sentences.
Perhaps one of the greatest concerns within the 3553(a) factors arising from the loss table’s outsized role is sentencing disparities. There is a broad judicial consensus that the results of Section 2B1.1 yield sentences that are incongruent with the criminal conduct at issue. See, e.g., United States v. Corsey, 723 F.3d 366, 377–78 (2d Cir. 2013) (stating “the loss guideline is fundamentally flawed” and “sentences in high-loss cases will remain wildly divergent as some district judges apply the loss guideline unquestioningly while others essentially ignore it.”) (Underhill, J., concurring); see also United States v. Moody, No. 1:13-CR-00087, ECF 28 at 9, 2013 U.S. Dist. LEXIS 109506, at *14 (D. Colo. Aug. 5, 2013) (declining to impose a Guideline sentence based in part on the preceding Underhill concurrence in Corsey); United States v. Gupta, 904 F. Supp. 2d 349, 351 (S.D.N.Y. 2012) (“The Guidelines’ calculations for this offense are no longer tied to the mean of what federal judges had previously imposed for such crimes, but instead reflect an ever more draconian approach to white collar crime, unsupported by any empirical data. . . . Was such a crime really 500% worse in 2003 than it was in 1987?”); United States v. Parris, 573 F. Supp. 2d 744, 745, 747–48 (E.D.N.Y. 2008) (imposing a sentence of 60 months despite an advisory guidelines range of 360 months to life) (quoting United States v. Adelson, 441 F.Supp.2d 506, 515 and 510 (S.D.N.Y. 2006) (noting the Guidelines “have so run amok that they are patently absurd on their face,” due to the “kind of ‘piling-on’ of points for which the guidelines have frequently been criticized.”)). [18]
Moreover, this judicial view of white collar sentencing has been the consensus since before Section 2B1.1 was enacted, as the initial version of the table recommended sentences above the mean of the custodial sentences studied at its drafting. [19] This continued wide-ranging consensus bears out in nationwide sentencing data today, with the majority of Section 2B1.1 cases garnering a below Guidelines sentence. [20] These are by no means small variances. Referencing sequential Commission Sourcebooks from 2015 to 2019, the mean sentence was a 50 percent decrease off of the low-end Guidelines range, and the median received a 45 to 50 percent decrease off of the low end of the Guidelines range. [21] This degree of departures and variances was much more significant than for other criminal offenses. Over the same timeframe, all other crimes were sentenced between 35 and 37 percent below the low end of the Guidelines range. [22] Moreover, this trend has continued into the most recent data, with the mean downward variance for economic offenders at 58 percent below the Guidelines. [23]
Empirically, Section 2B1.1 has never reflected the actual sentencing practices for non-violent, economic offenses. Worse, it has steadily drifted further and further away from the actual sentencing data, until today recommending a sentence more in line with violent crimes and repeat offenders.
[1] For a thorough discussion of the amendments to USSG § 2B1.1 and its outsized impacts on fraud Guidelines, see Barry Boss and Kara Kapp, How the Economic Loss Guideline Lost its Way, and How to Save It, 18 Ohio St. J. of Crim. Law 605 (2021).
[2] With the lowest base offense level of 6 and a maximum loss enhancement of 30, such a defendant would be at an offense level of 36 out of 43 based on loss alone. Compare USSG § 2B1.1(a)(2) and (b)(1) with USSG, Ch. 5 Pt. A.
[3] See USSG, Ch. 1 Pt. A. (“In its initial set of guidelines, the Commission sought to . . . tak[e] an empirical approach that used as a starting point data estimating pre-guidelines sentencing practice. It analyzed data drawn from 10,000 presentence investigations, the differing elements of various crimes as distinguished in substantive criminal statutes, the United States Parole Commission’s guidelines and statistics, and data from other relevant sources in order to determine which distinctions were important in pre-guidelines practice.”).
[4] See id.; see also Boss and Kapp, supra note 1, at 608 (citations omitted).
[5] Boss and Kapp, supra note 1, at 609 (citing Mark Osler and Mark W. Bennett, A “Holocaust in Slow Motion?”: America’s Mass Incarceration and the Role of Discretion, 7 DePaul J. for Soc. Just. 117, 141 (2014)).
[6] Id. at 610.
[7] See USSG, App. C, Amend. 154 (Nov. 1, 1989).
[8] See id.
[9] See id.
[10] See USSG, App. C, Amend. 617 (Nov. 1, 2001).
[11] Id.
[12] See USSG, App. C, Amend. 791 (Nov. 1, 2015) (“Reason for Amendment: This amendment makes adjustments to the monetary tables in [] § 2B1.1 (Theft, Property, Destruction, and Fraud) . . . to account for inflation.”).
[13] This has a base offense level of 33 (USSG § 2A1.5(a)) and assumes a three-point reduction for acceptance of responsibility.
[14] The Drug Quantity Table sets a base offense level of 30 for offenses involving at least 400 grams but less than 1.2 kilograms of fentanyl (USSG § 2D1.1(c)), with a two-point enhancement for possession of a firearm (USSG § 2D1.1(b)(2)), and one point for laundering (USSG § 2S1.1(b)(2)(A)). This, again, also assumes a three-point reduction for acceptance of responsibility.
[15] This has a base offense level of 32 (USSG § 2A4.1(a)), with a one-point enhancement as the victim was not released within seven days (USSG § 2A4.1(b)(4)(B)). This, again, also assumes a three-point reduction for acceptance of responsibility.
[16] The Drug Quantity Table sets a base offense level of 30 for offenses involving at least 400 grams but less than 1.2 kilograms of fentanyl (USSG § 2D1.1(c)), with a one-point enhancement for laundering (USSG § 2S1.1(b)(2)(A)). This, again, also assumes a three-point reduction for acceptance of responsibility.
[17] The base offense is 27 (USSG § 2A2.1(a)(2)), with a two-level enhancement for serious bodily injury (USSG § 2A2.1(b)(1)(B)) and a three-level enhancement for a hate crime motivation (USSG § 3A1.1(a)). This, again, also assumes a three-point reduction for acceptance of responsibility.
[18] See also Boss and Kapp, supra note 1, at 618 n.64 (collecting and discussing the preceding cases).
[19] See supra note 5 and accompanying text; see also Gupta, 904 F. Supp. 2d at 351 (S.D.N.Y. 2012) (“The Guidelines’ calculations for this offense are no longer tied to the mean of what federal judges had previously imposed for such crimes, but instead reflect an ever more draconian approach to white collar crime, unsupported by any empirical data.”) (emphasis added).
[20] Boss and Kapp, supra note 1, at 621–22 (citing U.S. Sent’g Comm’n, 2015 Sourcebook of Federal Sentencing Statistics tbl.27A [hereinafter 2015 Sourcebook]; U.S. Sent’g Comm’n, 2016 Sourcebook of Federal Sentencing Statistics tbl.27A [hereinafter 2016 Sourcebook]; U.S. Sent’g Comm’n, 2017 Sourcebook of Federal Sentencing Statistics tbl.27A [hereinafter 2017 Sourcebook]; U.S. Sent’g Comm’n, 2018 Sourcebook of Federal Sentencing Statistics tbl.31 [hereinafter 2018 Sourcebook]; U.S. Sent’g Comm’n, 2019 Sourcebook of Federal Sentencing Statistics tbl.31 [hereinafter 2019 Sourcebook]).
[21] Boss and Kapp, supra note 20, at 622 (citing 2018 Sourcebook, supra note 20, at tbl.40; 2019 Sourcebook, supra note 20, at tbl.40).
[22] Id. (citing 2015 Sourcebook, supra note 20, at tbl.31C (36.8% decrease); 2016 Sourcebook, supra note 20, at tbl.31C (36.4% decrease); 2017 Sourcebook, supra note 20, at tbl.31C (34.9% decrease); 2018 Sourcebook, supra note 20, at tbl.40 (34.8% decrease); 2019 Sourcebook, supra note 20, at tbl.40 (34.8% decrease)).
[23] U.S. Sent’g Comm’n, 2022 Sourcebook of Federal Sentencing Statistics tbl.40 [hereinafter 2022 Sourcebook].