18 U.S.C. § 641 Embezzlement Is Not a “Continuing Offense” Under Toussie (First Circuit)

I. Introduction

United States v. Pontz arises from an alleged long-running public-benefits fraud in which Kenneth Pontz was convicted by a jury of theft of government money under 18 U.S.C. § 641. The government’s trial theory was that Pontz misrepresented his living arrangements to obtain Supplemental Security Income (SSI) to which he was not entitled, while his wife received Social Security Disability Insurance (SSDI). The indictment charged a single count covering a broad time range—May 2014 through June 2020—despite the five-year limitations period for noncapital federal crimes, 18 U.S.C. § 3282(a).

The appeal presented two main issues:

1. Statute of limitations / continuing-offense doctrine (question of first impression in the First Circuit): whether § 641 embezzlement is a “continuing offense” such that an indictment filed in 2022 could lawfully charge embezzlement conduct beginning in 2014.
2. Evidentiary ruling: whether the district court improperly admitted “lay opinion” testimony under Federal Rule of Evidence 701 from an SSA “claims technical expert” (Aguayo) on matters that Pontz argued required expert qualification under Rule 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc..

The First Circuit (Rikelman, J.) joined the majority of circuits on the limitations issue, rejecting the district court’s continuing-offense ruling, while affirming the challenged evidentiary decisions.

II. Summary of the Opinion

The First Circuit held that embezzlement under § 641 is not a continuing offense under the two-prong test of Toussie v. United States: Congress did not expressly make it continuing, and embezzlement is not continuing “by its nature” because once the elements are complete there is no renewed daily “threat of . . . substantive evil.” As a result, the government could not charge Pontz with a § 641 crime that “occurred” more than five years before the indictment (i.e., before June 16, 2017).

For relief, the court:

• Vacated restitution and forfeiture because they included losses outside the limitations window.
• Remanded for the district court to determine the appropriate remedy concerning the conviction itself (the parties had not adequately briefed whether the conviction must be reversed, can stand with a narrower time window, or requires other corrective action).
• Affirmed the admission of the SSA witness’s testimony as lay opinion to the limited extent allowed, and found any possible hearsay error harmless.

III. Analysis

A. Precedents Cited

1. The controlling framework: Toussie v. United States

The decision is anchored in Toussie v. United States, which sharply limits the continuing-offense doctrine to two circumstances: (1) where the statute’s explicit language compels a continuing-offense reading, or (2) where the nature of the crime is such that Congress “must assuredly have intended” it to be treated as continuing. Toussie also emphasizes that statutes of limitation should be “liberally interpreted in favor of repose.”

Applying Toussie, the First Circuit treated “continuing offense” as a term of art, not a synonym for a course of conduct that happens to extend over time. In this respect, the court’s reasoning tracks the approach articulated by the Seventh Circuit in United States v. Yashar.

2. Distinguishing “scheme” cases from “continuing offense” cases: United States v. Yashar and United States v. Jaynes

The court relied on United States v. Yashar to underscore the key conceptual separation: a multi-act “scheme” is not necessarily a continuing offense for limitations purposes. Yashar warned that focusing on the indictment’s time span or a prosecutor’s “course of conduct” framing would effectively add a third prong to Toussie and weaken limitations protections. Similarly, United States v. Jaynes is cited for the idea that “continuing offense” is not the same as a pattern of illegal conduct.

3. The court’s sister-circuit alignment: United States v. Green, United States v. Askia, and the rejection of United States v. Smith

The First Circuit expressly joined the “majority of our sister circuits,” citing:

• United States v. Green (2d Cir.) (addressing § 641 embezzlement under Toussie),
• United States v. Askia (8th Cir.) (addressing embezzlement under § 666 and treating it as closely related), and
• United States v. Yashar (7th Cir.) (also § 666; government concession there notwithstanding).

Against these, the court confronted the Fourth Circuit’s outlier view in United States v. Smith, which treated certain “recurring, automatic scheme[s]” of § 641 embezzlement as continuing offenses. The First Circuit rejected Smith for reasons tied directly to Toussie: Smith’s “automatic vs. affirmative acts” distinction depends on the defendant’s manner of committing the crime, whereas Toussie directs courts to look to statutory language and the nature of the offense categorically, not to fact-specific charging characterizations. The court also cited United States v. Tavarez-Levario for the proposition that the inquiry examines “the offense itself, not the defendant’s particular conduct.”

4. Embezzlement’s definition and elements: Moore v. United States, United States v. Torres Santiago, and United States v. Lee

The court grounded “embezzlement” in the classic definition from Moore v. United States: the fraudulent appropriation of property by someone to whom it was entrusted or who lawfully possessed it. For § 641’s elements, it cited United States v. Torres Santiago and treated the $1,000 threshold as an element of felony § 641 embezzlement, consistent with United States v. Lee. The panel’s footnote discussion matters operationally: in multi-transaction cases, the felony is complete when “the first cent after $1,000 is embezzled.”

5. Aggregation and “single count” charging vs. limitations: United States v. Daley and duplicity doctrine

The government leaned on language from United States v. Daley that embezzlement may be a “single, continuing scheme” and can be charged in a single count. The First Circuit sharply limited Daley to its context: duplicity concerns (as later framed in United States v. Prieto) differ from whether an offense “continues” for limitations. Put differently, the permissibility of aggregating multiple transactions into one count does not imply tolling the statute of limitations after the offense’s elements are satisfied.

6. Related limitations-policy and interpretive references

In reinforcing limitations policy, the court echoed Toussie’s rationale (fairness, staleness, prompt investigation incentives) and referenced the admonition from Pendergast v. United States that limitations rules may allow some wrongdoers to escape, but that consequence is inherent in repose-based legislative judgments. The court also cited larceny-related authorities (e.g., United States v. McGoff) and the Eighth Circuit’s statement in United States v. Askia that theft/larceny are completed offenses, to rebut the idea that § 641’s penalty “aggregation” language signals continuity.

7. Evidence law precedents: Rule 701 boundaries and harmless error

On lay opinion, the court synthesized its line of cases: United States v. Maher (lay expertise from job experience; expert disclosure and Daubert when technical), United States v. Vega (must be reasoning familiar to average jurors), United States v. Galatis (industry terms without instructing on the law), United States v. Valdivia (personal-knowledge-based opinions), and contrasted United States v. O'Donovan (inadmissible where not grounded in witness perception/experience and not meaningfully testable). For hearsay and harmlessness, it invoked United States v. Maldonado-Peña and United States v. Laureano-Pérez.

B. Legal Reasoning

1. Why § 641 embezzlement fails Toussie prong one (explicit statutory language)

The panel quickly disposed of Toussie’s first prong: § 641 contains no express “continuing offense” language of the type Congress has used elsewhere. That absence put the full weight of the dispute on prong two.

2. Why § 641 embezzlement fails Toussie prong two (nature of the crime and assured congressional intent)

The heart of the opinion is its categorical assessment of embezzlement’s “nature.” The court reasoned that embezzlement is completed at a discrete moment—when the defendant knowingly converts government money to his own use with improper intent. Even when a defendant repeats that act over time, the harm is “episodic” (each conversion), not a single persisting condition that continues to generate a renewed daily threat after completion.

This is the opinion’s central conceptual move: it distinguishes crimes like conspiracy and kidnapping—which, once begun, persist as ongoing conditions until termination—from embezzlement, where each conversion is a completed wrong. The court drew on the “renewed threat” language from Toussie, and by analogy to kidnapping and escape cases (e.g., United States v. Rodriguez-Moreno; United States v. Bailey) to illustrate what “continuing by nature” looks like.

3. The court’s treatment of statutory structure, amendments, and legislative history

The government argued that § 641’s penalty provision—allowing aggregation “combining amounts from all the counts” to meet the $1,000 felony threshold—signals an intent to treat embezzlement as continuing. The court rejected that inference:

• The aggregation clause addresses valuation for felony punishment, not limitations tolling once the offense is complete.
• Reading aggregation as a continuing-offense signal would logically apply to all § 641 modalities (steals, purloins, converts, sells), which clashes with well-established treatment of theft/larceny as completed offenses.
• The 2004 amendment history (Identity Theft Penalty Enhancement Act) reflects a desire to resolve a felony-aggregation split—especially for “improper receipt of monthly Federal benefits” under $1,000— but does not speak to limitations rules. The court characterized this as Congress “simply not thinking about” extending limitations, borrowing the analytic posture from United States v. Figueroa-Cartagena.

4. The “$1,000 element” and when limitations begins

A practically significant clarification is the panel’s timing rule for multi-transaction felony embezzlement: because the $1,000 threshold is an element, the limitations period for a felony begins when that element is met—i.e., when the embezzled total crosses $1,000. This frames the limitations question around when the felony became complete, not when the last payment was taken.

The court also rejected any suggestion (attributed to readings of Yashar) that if the government could have charged a felony earlier, it is forever barred from charging later felonies tied to a continuing scheme. It aligned instead with Askia: the government may charge violations committed within the limitations period even if earlier conduct is time-barred.

5. Remedy: vacatur of restitution/forfeiture; remand on conviction

Having found a limitations error in the charging period, the court’s remedial steps were cautious and procedure-focused. It vacated restitution and forfeiture under the principle (from United States v. Cutter quoting Hughey v. United States) that restitution must be tied to the “specific conduct that is the basis of the offense of conviction.”

As to whether the conviction itself must be reversed, the panel emphasized that neither party adequately developed the remedy argument (citing the briefing-duty principle from United States v. Zannino), and remanded for the district court to determine the appropriate relief in the first instance. In previewing one possible analytic route, the court pointed to variance doctrine (as discussed in United States v. Katana and earlier in United States v. Mubayyid quoting United States v. Miller), suggesting that proving a narrower scheme within the limitations period might not necessarily add “anything new” to what the grand jury charged, depending on the record and prejudice analysis.

C. Impact

• First Circuit alignment with the majority rule: Prosecutors in the First Circuit can no longer rely on a “single continuing scheme” theory to reach back beyond five years for § 641 embezzlement by labeling it a continuing offense. This brings the circuit in line with decisions like United States v. Green and (by analogy) § 666 decisions like Yashar and Askia.
• Charging strategy consequences: The government must either (a) charge only conduct within five years, or (b) ensure that the offense of conviction is complete within five years (e.g., that the felony elements, including the $1,000 threshold, were met within the window), while accepting that earlier transactions may be time-barred as substantive bases for conviction.
• Restitution/forfeiture discipline: Even when long-term benefits fraud is proved, monetary penalties must track the legally valid offense period; the opinion signals heightened scrutiny where restitution is calculated over an overbroad date range.
• Rule 701 boundary reinforcement: The court reaffirmed that agency employees may offer limited lay opinion grounded in their day-to-day experience, but courts must police the line to prevent backdoor expert testimony—an issue likely to recur in benefits-fraud and white-collar prosecutions involving administrative codes, eligibility systems, and complex calculations.

IV. Complex Concepts Simplified

1. “Continuing offense” vs. “continuing scheme”

A continuing scheme is a factual description: repeated acts over time (e.g., monthly improper benefits). A continuing offense is a legal category: the crime is treated as persisting day-to-day so the statute of limitations starts only when the offense ends. Pontz holds that § 641 embezzlement may be a “scheme,” but it is not a “continuing offense” under Toussie.

2. Why the statute of limitations matters here

Federal prosecutors typically must indict within five years of when the crime is complete. If an offense were “continuing,” prosecutors could reach back further as long as the offense continued into the limitations window. The First Circuit rejected that extension for § 641 embezzlement.

3. Rule 701 lay opinion vs. Rule 702 expert testimony

Lay opinion (Rule 701) is allowed when it is based on a witness’s own perceptions and uses reasoning ordinary jurors can follow (including some on-the-job experience). Expert opinion (Rule 702) is required when testimony depends on specialized technical or scientific knowledge and must satisfy disclosure and reliability requirements (including Daubert). In Pontz, the court allowed limited SSA explanations and high-level impact opinions but disallowed (and the trial court policed) technical loss-calculation testimony as lay opinion.

V. Conclusion

United States v. Pontz establishes a clear First Circuit rule: § 641 embezzlement is not a continuing offense under Toussie. The statute of limitations therefore bars charging a § 641 embezzlement crime that was complete more than five years before indictment, even if the government frames the conduct as a long-running scheme. The court’s holding reshapes charging and restitution/forfeiture practices in extended benefits-fraud cases, while also reaffirming careful limits on admitting agency “technical” testimony as lay opinion.