In the post-Booker1 era, certain aspects of federal sentencing law have been solidified. It is, of course, now understood that the federal Sentencing Guidelines are advisory, not mandatory. In the current regime, when imposing a sentence, district courts must consider the applicable Sentencing Guidelines range, along with the other factors itemized in 18 U.S.C. §3553(a). Appellate courts can review sentences for procedural error and “substantive reasonableness.”2 But it is not yet clear precisely how much leeway appellate courts will give district courts when faced with government challenges to the substantive reasonableness of sentences that are far below the applicable Guidelines range. The recent decision by the U.S. Court of Appeals for the First Circuit in United States v. Prosperi, 686 F.3d 32 (1st Cir. 2012), affords great discretion to sentencing courts to deviate from the Sentencing Guidelines, despite expressing palpable discomfort with the extent of deviation at issue in this particular case. For this reason, the opinion is likely to be cited often in the First Circuit and elsewhere, and its analysis and approach warrants examination.


In white-collar cases, the question of how much room sentencing courts have to impose below-Guidelines sentences (and how stringent appellate review of such sentences will be) looms large in light of the general trend in post-Booker sentencing. Although the Guidelines have been rendered advisory, the government still more often than not takes the position that the appropriate sentence in a white-collar case is one within the Guidelines range. This is so, notwithstanding that the Guidelines ranges for white-collar cases are often extremely high. In particular, because the Guidelines ranges for white-collar cases are subject to large enhancements driven by the amount of money and number of victims involved, the ranges often dictate lengthy prison sentences, even for garden variety frauds.

Against this backdrop, defense attorneys often argue that the Sentencing Guidelines in fraud cases overstate the seriousness of the offense, and urge district courts to impose sentences below the applicable Guidelines range. For their part, prosecutors resist such arguments, and usually argue that a sentence within the applicable Sentencing Guidelines range is appropriate. District courts, attempting to discharge their responsibilities to balance the factors set forth in 18 U.S.C. §3553(a), have evinced a willingness to impose below-Guidelines sentences.3 This willingness has not gone unnoticed by the government. In fact, the government protests tendencies by district court judges, in executing their obligation to consider all of the §3553(a) factors, to impose sentences below the applicable Sentencing Guidelines range.4

District Court’s Decision in ‘Prosperi’

This tension is highlighted by the First Circuit’s recent decision in Prosperi. The case arose out of Boston’s Central Artery/Tunnel project (the “Big Dig”), one of the largest public works projects in U.S. history at the time of its completion. The project lasted for many years (from 1991 to 2007), involved numerous private contractors and subcontractors, and ultimately cost approximately $14 billion (funded jointly by the Commonwealth of Massachusetts and the federal government).5 The Big Dig required massive amounts of concrete, approximately 60 percent of which was provided by Aggregate Industries NE, a subcontractor. The government charged that over the course of nine years, Aggregate knowingly provided concrete that did not meet the project specifications, and concealed that failure by creating false documentation purporting to show that the concrete did satisfy the relevant specifications. Several employees of Aggregate, including Prosperi and Stevenson (whose sentences are at issue in the opinion), were convicted of various fraud and false statement offenses in connection with this scheme.6

In particular, the government alleged that Aggregate supplied concrete to the Big Dig that failed to meet contract specifications, and topped-off loads of leftover concrete (sometimes of a different mix design) with fresh concrete meeting specifications, representing that the entire load was fresh concrete consistent with project specifications.7 To conceal this practice, Aggregate employees falsified paperwork and took other actions designed to elude detection by inspectors.8

At sentencing, the issue of the proper “loss amount” to be used for calculating the offense level under the fraud Guidelines was hotly contested. The government argued that the loss amount should be based on the total amount paid for the concrete that failed to meet specifications (without credit for the value of the services/concrete that were actually provided). To that end, the government proffered a loss amount based on an estimate of that value, and argued that it was “conservative” because it omitted certain non-compliant concrete as well as any amount for reasonably foreseeable repairs that would be necessitated due to the use of non-specification concrete.9 Defendants, however, maintained that there was no loss attributable to them because the necessary amount of concrete was actually delivered (and, in particular, the Massachusetts Turnpike Authority had certified to federal authorities after post-construction testing—and after the fraudulent scheme had been uncovered—that the materials conformed with applicable plans and specifications).10 Thus, the defendants argued that the finished product met applicable safety standards and required no repairs, and therefore that there was no monetary harm on which to base a sentence enhancement.

Ultimately, prior to sentencing, the district court adopted the government’s multimillion dollar loss figure. The court reasoned that although neither approach was “precise,” the government’s method seemed “probably as good as any.” But, the court observed that the defendants had not intended to enrich themselves personally, nor had they intended to harm the Big Dig project or the public in any specific respect. And, the district court warned the parties in no uncertain terms that, given the nature of the case, it did not believe the loss amount was of “pivotal significance” in fashioning the appropriate sentence.11

As is typical in fraud cases, the district court’s adoption of the government’s view of the loss amount attributable to the defendants had a pronounced effect on the calculation of the Sentencing Guidelines range (resulting in an 18-level enhancement). With other applicable enhancements, the district court calculated the applicable Guidelines range as 87-108 months’ imprisonment for both Prosperi and Stevenson. As harbingered by its earlier memorandum regarding the loss amount, the district court emphasized at the sentencing hearing that it did not believe the Guidelines range, which was driven by the loss amount, accurately reflected the defendants’ culpability. Id. at 39. After hearing from the parties, the district court stated that after extensive reflection, it had concluded that a non-incarcerative sentence was appropriate for both defendants. Accordingly, each defendant was sentenced to a term of probation, with some requirements of home confinement, community service, and a fine. In its statement of reasons, the court explained that neither defendant acted to enrich himself, but rather, acted as part of a corporate culture to benefit his employer, and that the sentences imposed would serve as adequate deterrents and just punishment, and also permit the defendants to meet their respective family obligations.12

The First Circuit Reviews

The government appealed the sentences as substantively unreasonable. The First Circuit observed that sentencing courts have “very broad” discretion in fashioning sentences, and that on appellate review, the district court’s selected sentence will not be disturbed as long as there is a “plausible explanation” for the sentence and the overall result is “defensible.”13 In essence, the government argued that the probationary sentences represented an 87-month variance (100 percent) from the bottom of the applicable Sentencing Guidelines range, and that such a drastic variance was not justified by the sentencing court’s rationale (although acknowledging—as it had in the lower court—that a more modest variance from the Guidelines range would be appropriate).14 Because the loss amount was pivotal in setting the 87-month floor for the Guidelines range, the government also criticized heavily the district court’s refusal to let the loss amount drive the sentences imposed.15

Accordingly, in reviewing the substantive reasonableness of the sentences, the First Circuit initially focused on whether the district court had offered a plausible explanation for minimizing the impact of the loss amount. The court reviewed the reasons articulated by the district court in detail, and concluded that several factors underpinned the challenged decision: The uncertain, imprecise estimate of the amount paid for the noncompliant concrete; uncertainty regarding whether any monetary loss existed in light of the Massachusetts Turnpike Authority’s certification (suggesting that at least some value had been delivered); the fact that independent testing showed the project concrete met or exceeded concrete standards; the lack of any non-speculative proof that defendants’ conduct would cause any future failure of the concrete; and the absence of any personal benefit flowing to the defendants.16 Overall, the First Circuit found that these findings and conclusions constituted “plausible” explanations for the district court’s refusal to give significant weight to the loss amount it calculated pursuant to the Sentencing Guidelines.17


The relaxed review applied by the First Circuit to this aspect of the district court’s rationale is significant. As the court recognized, the strength of the justification required to support a variance from a Sentencing Guidelines range fluctuates with the degree of that variance; the greater the deviance from the applicable Sentencing Guidelines range, the more significant the justification required to support it.18 Here, the government’s principal complaint boiled down to the huge extent of the variance—from a more than seven-year sentence to probation. By accepting the district court’s decision not to give the loss amount much weight, the First Circuit essentially approved a reduction in the spread; once the loss amount is removed as the pivotal factor driving the sentence, the government’s argument that the breadth of the variance between zero and 87 months is unjustifiable loses traction.

The balance of the First Circuit’s analysis of the district court’s rationale reflects its acceptance of its key tenet—the disregard of the loss amount as the determinative factor. The court reviewed the government’s other objections to the district court’s proffered justification—including its arguments that the court improperly ascribed blame for the offense to corporate culture (thereby absolving the defendants); gave undue weight to the lack of any intent by the defendants to harm the Big Dig or the public; and gave insufficient weight to the need for deterrence—and swiftly rejected them.19

Tellingly, the First Circuit went out of its way to question the wisdom of the sentences imposed. The court noted that the sheer extent of the variance gave it pause, stating: “We are mindful of how rare it is to encounter a variance of this magnitude.”20 The court also observed:

One can easily argue that home confinement remains an unreasonably shallow sentence for a serious and deliberate crime which had the potential to cause large monetary loss and even physical harm to others. Many judges would have imposed prison sentences in this case….21

The court also focused on a “less tangible” harm caused by the offense—that the fraud undermined public confidence in the safety of the Big Dig and the ability of the government to discharge its function—emphasizing that it “adds to our uneasiness” and “should have been addressed by the court in its explanation of the sentences.”22 Notwithstanding these strong signals that it had great qualms about the appropriateness of the probationary sentences, the First Circuit hewed carefully to the deferential abuse-of-discretion standard of review. It concluded that the district court had carefully explained its reasoning and offered a “plausible explanation,” and that the “overall result is defensible.”23 Accordingly, the court affirmed the sentences.

Like most sentencing decisions, Prosperi is highly dependent on its facts, yet the opinion is likely to reverberate in white-collar sentencing jurisprudence. The willingness of the district court not just to mitigate the impact of the loss amount on the sentence, but essentially to disregard its effect entirely, will be an attractive precedent to defendants facing staggering sentences driven largely by loss amounts. And while the government will surely strive to limit Prosperi to its facts, it will not be difficult for defense lawyers to analogize other fraud cases to at least some of the factors present in Prosperi. Fundamentally, the Prosperi opinion also signals to district courts that, at least in the First Circuit, there are few restraints on their discretion to impose sentences far below the applicable Guidelines range in fraud cases; as long as they explain why they did so, citing lawful considerations, the sentences will not be disturbed on appeal even when the Court of Appeals plainly disagrees with the result. If embraced by district courts, this may galvanize a trend away from the uniformity that the Guidelines seek to impose, particularly in white-collar cases, and toward a return to the flexibility and discrepancy in sentencing often associated with the pre-Guidelines era.

Laura Grossfield Birger is a partner in the business litigation practice group at Cooley in New York.


1. Booker v. United States, 543 U.S. 220 (2005).

2. Gall v. United States, 552 U.S. 38 (2007).

3. See, e.g., United States v. Ovid, 09 Cr. 216 (E.D.N.Y. Oct. 1, 2010) (Gleeson, J.).

4. See Letter from Lanny A. Breuer, Assistant Attorney General, and Jonathan J. Wroblewski, Director, Office of Policy and Litigation, to Honorable Patti B. Saris, U.S. Sentencing Commission (July 23, 2012), at 6 (noting concern that federal sentencing practice has trended away from guideline sentencing, resulting in erosion of the guidelines and unwarranted sentence disparities both within and among courthouses).

5. Prosperi, 686 F.3d. at 34-35.

6. Id. at 34-35, 37.

7. Id. at 35.

8. Id. at 36.

9. Id. at 37.

10. Id. at 38.

11. Id. at 38-39.

12. Id. at 41-42.

13. Id. at 42-43 (citations omitted).

14. Id. at 43; see Brief for the United States, United States v. Prosperi and Stevenson, Nos. 10-1739, 1741 (1st Cir.) at 24, 34.

15. Prosperi, 686 F.3d at 43.

16. Id. at 43-44.

17. Id. at 44.

18. Id. at 42.

19. Id. at 45-48.

20. Id. at 49.

21. Id.

22. Id. at 46 n.11.

23. Id. at 50 (internal citations omitted).