Clarifying Intended-Loss Calculations and Distinct Enhancements for Overseas Crypto Laundering; MVRA Restitution Without Express Request — United States v. Mitan

Introduction

In United States v. Adrian Mitan, the Sixth Circuit affirmed a 140‑month sentence and restitution orders arising from three interrelated international fraud and money laundering conspiracies. Mitan, a Romanian national, pleaded guilty in a global resolution to conspiracy to commit bank fraud, money laundering, and a RICO conspiracy tied to vishing, brute-force phishing of payment cards, and a large-scale online auction scam. On appeal, he challenged (1) the district court’s use and calculation of “intended loss” under U.S.S.G. § 2B1.1; (2) the simultaneous imposition of enhancements for sophisticated money laundering and substantial non‑U.S. conduct as impermissible double counting; (3) the substantive reasonableness of a below‑Guidelines sentence; and (4) restitution, including a claim that the court awarded restitution on one count without notice and that RICO restitution exceeded his personal gain.

The Sixth Circuit rejected all arguments. The opinion is unpublished but notable for three clarifications: (a) reaffirming that “intended loss” remains the lodestar for § 2B1.1 in the Sixth Circuit for pre‑Amendment 827 sentencings and approving an empirically supported per‑card extrapolation; (b) articulating a “where versus how” framing that permits stacking the § 2B1.1(b)(10)(B) overseas-conduct enhancement with the § 2S1.1(b)(3) sophisticated-laundering enhancement when they address distinct harms; and (c) holding that MVRA restitution may be imposed at sentencing for a count not explicitly included in the government’s advance restitution briefing when the defendant had both general and specific notice and an opportunity to contest.

Summary of the Opinion

  • Procedural reasonableness: The district court properly used intended loss under § 2B1.1, reasonably estimated loss at approximately $500 per compromised card based on financial institutions’ data, and permissibly applied both the “substantial part outside the United States” enhancement (§ 2B1.1(b)(10)(B)) and the sophisticated laundering enhancement (§ 2S1.1(b)(3)) without double counting because they punished distinct aspects of the conduct.
  • Substantive reasonableness: A downward-variance sentence of 140 months (from a 151–188 month range) was not greater than necessary under § 3553(a). The court reasonably rejected a co‑defendant disparity argument in light of differences in criminal histories and culpability.
  • Restitution: The court did not err in ordering $75,220.51 restitution for the brute‑force scheme despite the government’s failure to expressly request it prehearing, because the MVRA made restitution mandatory, the global plea provided notice, the government disclosed victim loss figures before sentencing, and Mitan cross‑examined the witness. The $675,000 RICO restitution, apportioned by months of participation (15 months × $45,000 average monthly loss), was also affirmed; restitution compensates victims’ losses, not the defendant’s personal gain.

Analysis

Precedents Cited and Their Role

  • Standards of review and reasonableness:
    • Gall v. United States (procedural and substantive reasonableness; abuse of discretion review).
    • United States v. Gates; Perez‑Rodriguez; Xu; Vowell; Nunley (defining procedural errors; meaningful consideration of § 3553(a); appellate presumptions and burdens, especially for below‑Guidelines sentences).
    • Rita v. United States (adequacy of explanation; showing the court listened, considered evidence, and applied § 3553(a)).
  • Guidelines methodology and loss:
    • United States v. Agrawal (methodology de novo; factual findings clear error).
    • United States v. You; Smith; Tellez; Kennert (Sixth Circuit’s adoption of the commentary’s intended‑loss definition for § 2B1.1 before Amendment 827).
    • United States v. Riccardi (caution against mechanically importing a per‑unit loss from commentary without evidentiary support; distinguished here because the government supplied institution‑based data).
    • United States v. Ellis; Estrada‑Gonzalez (reasonable estimate suffices; acceptance if plausible on the record as a whole).
    • United States v. Nicolescu; Jackson (heavy burden to show loss findings are outside permissible computations).
    • Peugh v. United States (temporal use of Guidelines in effect at offense/sentencing consistent with ex post facto limitations).
    • United States v. Moody (panel bound by prior Sixth Circuit precedent absent en banc or Supreme Court change).
  • Enhancements and double counting:
    • U.S.S.G. §§ 2B1.1(b)(10)(B) (substantial part outside U.S.); 2S1.1(b)(3) and cmt. n.5 (sophisticated laundering, layered transactions).
    • United States v. Myers (multiple enhancements permissible when triggered by different conduct/harms) versus United States v. Mehmood (admonition where the same conduct triggers both sophisticated means and sophisticated laundering).
    • United States v. Hubbard; Vela‑Salinas (examples upholding sophisticated-laundering for multi‑step obfuscation).
  • Restitution (MVRA) and procedures:
    • 18 U.S.C. §§ 3663A (MVRA), 3664 (procedures; PSR content; late ascertainment options), and § 3664(h) (joint and several or apportionment).
    • United States v. Hills (scope and amount review standards).
    • United States v. Sawyer (due process satisfied by notice and opportunity to be heard).
    • United States v. Bogart (timing and procedure when losses are not ascertainable pre‑sentencing).
    • United States v. Williams (joint and several liability for conspiracy losses).
    • United States v. Kilpatrick (restitution compensates victim loss, not capped by defendant’s profit).

Legal Reasoning

1) Intended Loss Under § 2B1.1

The panel affirmed the district court’s reliance on “intended loss” as the measure of “loss” for § 2B1.1 because, at the time of sentencing (August 2021), the Guidelines’ main text did not define “loss,” but the commentary did, and Sixth Circuit precedent (You; Smith; Tellez; Kennert) embraced that definition: loss is the greater of actual or intended loss, and intended loss includes harm that was impossible or unlikely to occur. The court noted the 2024 Amendment 827 that later moved the loss definition into the text, but under Peugh, the court correctly applied the then‑governing framework and was bound by circuit precedent (Moody).

2) Estimating Intended Loss: Per‑Card Extrapolation

On the contested computation, the district court reasonably extrapolated a per‑card intended loss of about $500 across approximately 13,000 compromised cards (reflecting a 25% discount from ~16,000 to account for invalid numbers). Unlike the “arbitrary” per‑unit figure criticized in Riccardi, the government here supported the $500 figure with empirical data from affected financial institutions showing average losses ranging from roughly $500 to $648 per card. The district court explicitly treated the estimate as conservative and observed that even assuming more than half of the numbers were unusable, the total still exceeded the $3.5 million threshold for the 18‑level increase under § 2B1.1(b)(1)(J). Given the deferential clear‑error standard for factual loss findings and the permissibility of reasonable estimates in fraud cases (Ellis; Estrada‑Gonzalez), the calculation was “plausible on the record as a whole.”

3) Enhancements: No Double Counting Where the Enhancements Target Distinct Harms

Mitan argued that applying both § 2B1.1(b)(10)(B) (substantial part outside the United States) and § 2S1.1(b)(3) (sophisticated laundering) punished the same conduct. The court disagreed and adopted a clarifying “where versus how” rubric:

  • Where (Geographic locus): § 2B1.1(b)(10)(B) addressed the transnational nature of the fraud — operation from Romania, coordination with foreign actors, and routing funds through international accounts.
  • How (Method/complexity): § 2S1.1(b)(3) addressed the laundering architecture — converting proceeds to Bitcoin, cycling through multiple wallets, and using foreign intermediaries to layer and obfuscate transactions.

This analytical separation fits Sixth Circuit guidance permitting multiple enhancements when they address different aspects of culpability (Myers) and is consistent with cases recognizing sophisticated laundering in multi‑step financial obfuscation (Hubbard; Vela‑Salinas). The panel distinguished the caution in Mehmood, where the same conduct supported multiple “sophistication” enhancements, by showing that the enhancements here addressed different dimensions of the conduct.

4) Substantive Reasonableness

The district court varied downward to 140 months after considering the § 3553(a) factors. The panel rejected Mitan’s argument that the loss calculation overstated culpability (already addressed in the procedural review) and his claim of unwarranted disparity with co‑defendants. Citing Conatser, the court emphasized that co‑defendant parity is not required where defendants are not similarly situated. Here, other participants had Criminal History Category I, while Mitan had Category III and played a sustained, sophisticated, and international role warranting a deterrence‑oriented sentence. Given these factors and the below‑Guidelines sentence, Mitan’s heavy burden to show substantive unreasonableness (Nunley) was not met.

5) Restitution: MVRA, Notice, and Apportionment

The court affirmed two distinct restitution rulings:

  • Brute‑force scheme ($75,220.51): Although the government’s pre‑sentencing brief expressly sought approximately $2.7 million restitution for the RICO conspiracy and was late under the parties’ schedule, the MVRA made restitution mandatory for both fraud and the specified Title 18 offenses; the global plea agreement put Mitan on express notice that restitution would be “determined at sentencing” for all consolidated cases; and the government’s sentencing memorandum filed eleven days before the hearing disclosed specific brute‑force victim losses (e.g., FTFCU and U.S. Bank). At the hearing, the government presented testimony on brute‑force losses and Mitan cross‑examined the witness; he did not seek a continuance. Under Sawyer and § 3664, this sequence provided both general and specific notice and an opportunity to be heard, satisfying due process notwithstanding the missed filing deadline (Bogart). The district court did not abuse its discretion in relying on the detailed data.
  • RICO scheme ($675,000): The panel rejected Mitan’s argument to cap restitution at his personal gain ($10,801). Restitution compensates victims’ losses rather than the defendant’s profits (Kilpatrick), and joint and several liability is permitted for conspiracy losses (Williams; § 3664(h)). Notably, the district court apportioned rather than imposed the full $2.7 million, tying Mitan’s responsibility to his proven period of participation: the conspiracy’s average monthly loss ($45,000) multiplied by 15 months. This fact‑bound apportionment rested on evidence of continued involvement and was affirmed.

Impact and Practical Implications

  • Intended loss remains robust in the Sixth Circuit for pre‑Amendment 827 sentencings: Mitan reinforces that intended loss governs § 2B1.1 for sentencings that predate the 2024 amendment moving “loss” into the text. More broadly, it signals the court’s continued comfort with intended‑loss analysis even post‑Amendment, given the principled reliance on circuit precedent.
  • Empirical per‑card extrapolations are viable when evidence‑based: The court distinguished Riccardi by crediting institution‑level data to support a $500 average loss per card, and recognized conservative discounts (e.g., 25% reduction for invalid numbers). Prosecutors should present concrete loss metrics from victim institutions; defense counsel should challenge the nexus between sampled losses and the card universe or propose credible alternative averages with record support.
  • “Where vs. how” clarifies double‑counting boundaries: The opinion’s framing provides a clean template for stacking enhancements in transnational cyber‑laundering: § 2B1.1(b)(10)(B) captures the scheme’s overseas locus, while § 2S1.1(b)(3) captures the complexity of laundering (layering, crypto conversion, foreign intermediaries). Defendants arguing double counting must show the same conduct, not merely related conduct, triggers both enhancements.
  • Crypto‑based laundering comfortably fits “sophisticated laundering”: Converting proceeds to Bitcoin, cycling through multiple wallets, and cashing out via foreign intermediaries remains squarely within § 2S1.1(b)(3)’s ambit, consistent with prior Sixth Circuit approvals of layered obfuscation.
  • Restitution practice under the MVRA:
    • Courts may determine restitution at sentencing even if the government’s filing is late or omits a count, provided the defendant had notice and an opportunity to contest.
    • Plea agreements that expressly contemplate restitution for all counts, combined with pre‑hearing disclosure of victims and amounts, will typically satisfy due process.
    • Judges may apportion conspiracy restitution by duration of involvement using reliable average‑loss metrics — a pragmatic middle path between full joint-and-several liability and profit‑only caps.
  • Sentencing disparities: Mitan underscores that disparity arguments falter where co‑defendants differ in criminal history category and individualized § 3553(a) factors; below‑Guidelines sentences remain difficult to disturb on appeal.
  • Grouping in consolidated cyber‑fraud cases: Although not central to the appeal, the district court’s decision to group all three offenses and anchor the base offense level in the money‑laundering count illustrates a practical approach in multi‑scheme resolutions where money laundering is the most serious guideline driver.

Complex Concepts Simplified

  • Intended loss vs. actual loss: Actual loss is what victims actually lost. Intended loss is what the defendant sought to cause, even if the plan failed or was impossible. The Sixth Circuit applies the greater of the two for § 2B1.1 (for pre‑Amendment 827 sentencings).
  • Reasonable estimate of loss: Courts need not calculate loss with precision in fraud cases; they may reasonably estimate based on a preponderance of evidence (e.g., sampling, averages, extrapolation).
  • Double counting: Impermissible when two enhancements punish the same conduct for the same harm. Permissible when each enhancement targets a distinct aspect (e.g., geography vs. laundering complexity).
  • Sophisticated laundering: Multi‑step financial transactions designed to hide the source of criminal proceeds — including converting funds to cryptocurrency, using multiple wallets, layering transfers, and employing foreign intermediaries.
  • MVRA restitution: Mandatory for specified offenses, including many frauds, and designed to make victims whole. Courts can impose it at sentencing if the defendant had notice and a chance to contest, even if the government’s briefing was imperfect.
  • Joint and several liability vs. apportionment: Courts can hold each conspirator liable for the full loss or apportion responsibility based on participation. Restitution focuses on the victims’ losses, not the defendant’s gain.
  • Standards of review: Methodology for Guidelines calculations is reviewed de novo; underlying factual findings (including loss amounts) are reviewed for clear error. Overall sentence reasonableness is reviewed for abuse of discretion.
  • Unpublished decision: “Not recommended for publication” means the opinion is not binding precedent within the circuit; however, it is informative of how the court is likely to address similar issues.

Conclusion

United States v. Mitan reinforces three practical pillars in federal fraud and money‑laundering sentencing within the Sixth Circuit. First, intended loss remains an appropriate and often controlling metric for § 2B1.1 where supported by record evidence; empirically grounded per‑card averages are a permissible basis for extrapolation. Second, the court’s “where versus how” articulation clarifies that enhancements for substantial non‑U.S. conduct and sophisticated laundering may both apply without double counting when they target different facets of wrongdoing — an especially salient point for schemes using cryptocurrency and cross‑border layering. Third, MVRA restitution may be awarded at sentencing even absent an express prehearing request for a particular count when the defendant had adequate notice and an opportunity to be heard; courts may apportion conspiracy restitution by duration and average loss rather than defaulting to full joint‑and‑several liability or limiting restitution to the defendant’s personal gain.

Although unpublished, the decision offers clear guidance for future cyber‑enabled fraud prosecutions and sentencings: build an evidentiary record for loss estimates, separate geographic scope from laundering sophistication to avoid double counting, and provide timely (or at least adequate) notice for restitution, with itemized support. For defense counsel, Mitan highlights the importance of contesting the representativeness of loss samples, demonstrating overlap when claiming double counting, and seeking continuances when late disclosures threaten meaningful restitution challenges. The Sixth Circuit affirmed across the board, signaling continued deference to well‑reasoned district court findings grounded in concrete data and tailored § 3553(a) explanations.

Key Takeaways

  • Intended loss under § 2B1.1 — grounded in record evidence — remains the benchmark; empirical per‑card averages are acceptable.
  • Stacking § 2B1.1(b)(10)(B) and § 2S1.1(b)(3) is proper where they address different harms (geography vs. laundering complexity), particularly in crypto‑based schemes.
  • MVRA restitution can be imposed at sentencing on counts not expressly briefed if the defendant had notice and an opportunity to contest; apportionment by participation duration is permissible.
  • Below‑Guidelines sentences are difficult to overturn absent clear misapplication of § 3553(a) or arbitrary reasoning; co‑defendant disparity claims require truly comparable circumstances.