Vulnerable-Victim Sentencing in Mail Fraud: Exploited Financial Need Supplies the Nexus, and Fifteen Victims Suffice

A. The Governing Framework: Individualized Facts Within an Advisory Sentencing System

Section 3A1.1(b) operates in two stages. Subsection (b)(1) adds two offense levels when the defendant knew or should have known that a victim was vulnerable. If that subsection applies, subsection (b)(2) adds two more levels when the offense involved a large number of vulnerable victims. Neither defendant contested the initial enhancement in these appeals.

The Guidelines commentary defines a vulnerable victim as a victim of the offense or relevant conduct who is unusually vulnerable because of age, physical or mental condition, or otherwise particularly susceptible to the criminal conduct. The inquiry is therefore relational: the court must connect the individual’s circumstances to the particular offense.

Precedents establishing the method of review

• United States v. Rocha, 145 F.4th 1247 (10th Cir. 2025). This decision supplied the general framework for reviewing sentences: abuse of discretion overall, de novo review of legal conclusions about the Guidelines, and clear-error review of factual findings. It also established that reliance on an erroneous legal conclusion or clearly erroneous fact constitutes an abuse of discretion. Later in the opinion, it supported deference to the district court’s weighing of sentencing factors and rejection of any presumption that an outside-Guidelines sentence is unreasonable.

• United States v. Tony, 121 F.4th 56 (10th Cir. 2024). The court relied on this case for its approach to interpreting the Guidelines: ascertain the Sentencing Commission’s intent through the provision’s language and interpretive commentary. Under the rule stated there, commentary governs unless unconstitutional, contrary to statute, plainly erroneous, or inconsistent with the provision it interprets. This made the commentary’s examples central to the vulnerability analysis.

• United States v. Sumka, 81 F.4th 1153 (10th Cir. 2023). This precedent distinguished factual identification of unusually vulnerable victims from legal questions about whether established facts justify an enhancement. Following its approach, the panel declined to resolve the parties’ disagreement over the applicable standard because it would affirm under either clear-error or de novo review.

B. The Essential Nexus: Why Financial Need Could Facilitate This Fraud

The earlier appeal required more than a presumption about grief

The central starting point was United States v. Hess, 106 F.4th 1011 (10th Cir. 2024). In that earlier appeal, the court rejected two foundations for the enhancement. First, persons already deceased when the fraud occurred could not themselves be vulnerable victims under this provision. Second, the district court could not simply classify every grieving family member as unusually susceptible.

The earlier record showed that the defendants could perpetrate their scheme through positions of trust and unsupervised custody of bodies. It did not explain how grief made particular victims more susceptible to placing bodies in that custody. The remand therefore required individualized findings explaining how vulnerability enabled or facilitated the fraud. The earlier decision also required correction of the loss calculation and reconsideration of Koch’s sophisticated-means enhancement.

The new decision did not abandon that requirement. It held that the expanded record satisfied it. The district court received additional victim statements and testimony from Dr. Doug Shadel concerning the relationship between accumulated stressful life events and susceptibility to fraud. Importantly, the district court did not accept every expert-identified victim as qualifying: it found that four of the eleven discussed by the expert met the required nexus and identified additional qualifying victims from their statements.

Financial hardship affected the acquisition of bodies, not merely the opportunity to meet victims

The defendants’ fraudulent sales generated income that helped them offer unusually inexpensive cremations. They also promised free or discounted services in exchange for limited donations. Particular victims’ financial circumstances made these representations especially influential in selecting the funeral home or authorizing a donation.

That selection was integral to the fraud. By inducing financially constrained families to entrust remains to Sunset Mesa, the defendants obtained the unsupervised custody necessary to sell them. The vulnerability therefore operated at the procurement stage, even if the later unauthorized sale could have deceived a wealthier family equally effectively.

Similar individualized findings supported the enhancement for M.B., D.Mc., D.W., A.S.J., and R.B. Together with the five victims whose qualifying findings Hess did not contest—C.A.H., D.S., L.A.P., K.R., and C.B.—these findings established fifteen vulnerable victims.

Why inability to detect or prevent the fraud was not required

Hess sought to confine vulnerability to circumstances in which a victim was less able than others to detect or thwart the offense. The court rejected that formulation as a necessary condition. The commentary expressly treats marketing an ineffective cancer cure as an example of exploiting vulnerability. A cancer diagnosis need not diminish a person’s general ability to recognize deception; it makes the particular fraudulent offer especially compelling.

Likewise, the financially constrained victims here did not need to possess impaired judgment or unusual credulity. The relevant susceptibility was their heightened need for the very services falsely promised.

Precedents defining and limiting the nexus

• United States v. Proffit, 304 F.3d 1001 (10th Cir. 2002). A rancher’s cancer diagnosis led him to advertise his ranch for sale, allowing the defendant to approach him. The eventual fraud, however, involved a supposed investment in cattle futures, not the ranch sale. The court found the illness insufficiently connected to the investment decision, which reflected the ordinary attraction of a quick return. Critically, that opinion suggested a different result if the defendant had exploited the illness to defraud the victim of the ranch itself. The present panel used this distinction to show that vulnerability must connect directly to the fraud’s mechanism—not necessarily to reduced powers of detection. Here, unlike the investment fraud, the defendants’ promises addressed the victims’ identified needs.

• United States v. Fosher, 124 F.3d 52 (1st Cir. 1997). Hess relied on language asking whether a victim was less likely to detect or thwart the crime. The Tenth Circuit treated that First Circuit formulation as nonbinding and declined to adopt it as an exclusive test. The opinion does not establish that detection difficulties are irrelevant; it establishes that such difficulties are not indispensable where another direct nexus demonstrates unusual susceptibility.

• United States v. Brunson, 54 F.3d 673 (10th Cir. 1995). This case rejected an enhancement based on a successful fraud against a foreign company and the defendant’s belief that foreign businesses were easier to deceive. Successful victimization does not itself prove unusual vulnerability, and a defendant’s assumption about a group does not establish objective susceptibility. The present court distinguished that unsupported assumption from concrete evidence that particular families’ financial needs influenced their decisions. It also invoked the earlier case’s concern with heightened blameworthiness to explain why exploiting impoverished bereaved families warranted enhanced punishment.

• United States v. Creech, 913 F.2d 780 (10th Cir. 1990). The court had rejected a presumption that newlyweds receiving threats against their spouses were unusually vulnerable. Concern for a spouse is not confined to newly married people. This precedent illustrated the impermissible move from group membership to vulnerability without supporting evidence. The new sentencing findings avoided that defect by explaining each qualifying victim’s circumstances and their connection to the scheme.

• United States v. Tissnolthtos, 115 F.3d 759 (10th Cir. 1997). This decision required individualized findings rather than assuming that a seventy-one-year-old assault victim was unusually vulnerable because of age alone. It also recognized that some class characteristics may support an overwhelmingly strong inference of vulnerability. The present opinion used that distinction to preserve the general requirement of individualized proof without declaring every class-based inference impermissible.

• United States v. Drapeau, 110 F.3d 618 (8th Cir. 1997). Discussed through the preceding precedent, this case involved a one-year-old assault victim. The child’s inability to protect himself or report the abuse supplied an obvious age-related vulnerability. Its role was illustrative: some characteristics themselves powerfully establish susceptibility, whereas age seventy-one, bereavement, or poverty does not automatically do so in every case.

• United States v. Scott, 529 F.3d 1290 (10th Cir. 2008). This case supplied the principle that exploiting a particular victim’s vulnerability is more blameworthy and deserves greater punishment. Its application of the enhancement to exploitation of a minor’s small stature and unusual naivety in a trafficking offense reinforced the importance of connecting specific characteristics to the offense’s operation.

• United States v. Castaneda, 239 F.3d 978 (9th Cir. 2001). Quoted through the preceding case, this decision supplied the heightened-blameworthiness rationale. Its influence was explanatory rather than a separate factual analogy: the enhancement responds to the additional culpability of exploiting unusual susceptibility.

• United States v. Caballero, 277 F.3d 1235 (10th Cir. 2002). The defendants there exploited undocumented immigrants’ language barriers, unfamiliarity with American law, cultural backgrounds, and immigration status to facilitate immigration fraud. This precedent performed three functions here: it illustrated a valid vulnerability-offense nexus; showed that several victims may share a vulnerability without making individualized findings an impermissible class presumption; and established that sixteen vulnerable victims could satisfy the large-number requirement in a fraud case.

The resulting distinction is important: a shared characteristic does not invalidate individualized findings. The prohibition concerns unsupported presumptions, not evidentially established patterns of exploitation.

C. Fifteen Qualifying Victims Constituted a “Large Number”

Section 3A1.1(b)(2) does not assign a numerical definition to “large number.” The court approached the phrase in the context of the underlying offense rather than imposing a single number for every federal crime.

United States v. Kaufman, 546 F.3d 1242 (10th Cir. 2008), was the principal methodological precedent. It concerned exploitation and abuse of mentally ill residents in an unlicensed group home, including forced labor and involuntary servitude. The district court had declined the enhancement partly because the Guidelines did not clearly define a large number. The appellate court looked to the offense-specific involuntary-servitude guideline and concluded that ten victims was a large number in that setting, while remanding for particularized vulnerability findings.

United States v. Anderson, 349 F.3d 568 (8th Cir. 2003), appeared through that precedent as authority recognizing that the phrase was “largely uninterpreted.” It supplied background for the interpretive problem, not an independently controlling numerical threshold.

Applying the same contextual approach, the panel considered the fraud guideline, § 2B1.1(b)(2), which uses victim-related thresholds of five, ten, and twenty-five. It also relied on the sixteen-victim fraud decision in United States v. Caballero. These sources established that twenty-five was not a required minimum.

The court nevertheless left the lower boundary open. It did not decide whether five or ten should guide the inquiry in other fraud cases. Its holding was narrower: the fifteen qualifying victims established here were enough. It therefore had no need to review the six remaining vulnerability findings.

D. Koch’s Accountability Did Not Depend on Personal Contact with Every Victim

Koch argued that she lacked meaningful contact with most of the identified vulnerable victims and therefore could not have exploited their particular circumstances. The court rejected that argument by combining the enhancement’s language with the Guidelines’ relevant-conduct rules.

The commentary to § 3A1.1 includes victims of conduct for which a defendant is accountable under § 1B1.3. Section 1B1.3(a)(1)(A), in turn, includes acts the defendant committed, aided, abetted, or otherwise brought about within the specified relationship to the offense. Koch’s accountability therefore extended beyond her own conversations with customers to the fraudulent conduct she aided and abetted.

The court also distinguished the two enhancement stages. Subsection (b)(1) requires knowledge or constructive knowledge of at least one victim’s vulnerability. Subsection (b)(2) contains no additional requirement that the defendant know of every qualifying victim or personally appreciate the full number involved. Once the initial enhancement applies, the additional inquiry concerns the offense and its relevant conduct.

This reasoning does not make every participant automatically responsible for everything another person does. The necessary foundation remains accountability under the relevant-conduct rules. Here, that foundation was established by Koch’s participation in and aiding and abetting of the fraudulent scheme.

Precedents concerning offense context and the Government’s litigation position

• United States v. Boll, 3 F.4th 1099 (8th Cir. 2021). Koch invoked this case’s contextual approach, but the panel explained that “context” referred to the type of offense and its typical number of vulnerable victims, considering the conviction and relevant conduct. It did not restrict the inquiry to victims with whom the defendant personally interacted. Thus, the authority did not support Koch’s proposed personal-contact limitation.

• United States v. Ronquillo, 94 F.4th 1169 (10th Cir. 2024). This decision reviewed the legal scope of a search warrant despite the Government’s contrary concession below. It established that courts are not bound by parties’ stipulations on legal questions. The same principle permitted the sentencing court to apply the enhancement despite the Government’s plea-related position that it should not apply to Koch.

• Richison v. Ernest Grp., 634 F.3d 1123 (10th Cir. 2011). This case supported the distinction between seeking reversal on a new ground and defending an existing judgment. An appellee generally may defend the judgment on a ground supported by the record. Accordingly, the Government’s position below did not itself prevent it from supporting affirmance on appeal. The earlier decision in United States v. Hess had already explained that distinction to Koch.

• LaFevers v. Gibson, 182 F.3d 705 (10th Cir. 1999). Koch suggested that the Government might have breached her plea agreement by supplying evidence supporting the enhancement while formally opposing its application. Because she did not develop that contention, the court treated it as waived under this precedent’s rule concerning perfunctory appellate arguments.

The ruling must not be mistaken for a decision that the Government complied with every plea obligation. The court did not decide the merits of a breach claim. It separately held that the Government’s legal position did not bind the sentencing judge or establish appellate waiver.

E. Post-Sentencing Rehabilitation: A Limited Plain-Error Ruling

Hess submitted evidence of educational work, assistance to other inmates, participation in classes, suicide-watch certification, and numerous supportive letters. The district court acknowledged her productive conduct but characterized prison-rule compliance as both commendable and expected. It also noted the availability of good-time credit.

Hess argued that this reasoning improperly discounted rehabilitation. Because she had not objected during resentencing, however, she faced plain-error review rather than ordinary preserved-error review.

The controlling precedents

• United States v. Kee, 129 F.4th 1249 (10th Cir. 2025). This case supplied the four-part plain-error test: error, an error that is plain, an effect on substantial rights, and a serious effect on the fairness, integrity, or public reputation of judicial proceedings. Hess bore the burden of establishing each requirement.

• Pepper v. United States, 562 U.S. 476 (2011). The Supreme Court held that a resentencing court may consider post-sentencing rehabilitation and that such evidence may support a downward variance. It rejected categorical barriers to considering that evidence, including a rationale treating good-time credits as an adequate substitute. But it expressly did not require a sentence reduction whenever rehabilitation is shown. The Tenth Circuit distinguished a prohibition on consideration from a sentencing judge’s assessment of the evidence’s weight.

• United States v. Booker, 63 F.4th 1254 (10th Cir. 2023). This Tenth Circuit decision supplied the requirement that a plain error be obvious under current well-settled law. It also required a reasonable probability of a different outcome to establish an effect on substantial rights. Those principles independently defeated Hess’s claim.

The panel assumed error without deciding whether one occurred. It held that any error was not obvious: Pepper v. United States authorized consideration of rehabilitation but did not mandate a reduction or declare good-time credit irrelevant to the weight assigned to that evidence.

Independently, Hess failed to establish prejudice. The district court expressly found that the offense’s extraordinary seriousness outweighed mitigating personal circumstances. It also found continued minimization of the victims’ suffering and an absence of remorse. Given those findings, Hess had not demonstrated a reasonable probability that a different treatment of rehabilitation would have produced a lower sentence.

The holding is therefore limited. It does not establish that rehabilitation evidence may categorically be dismissed because compliant prisoners receive good-time credits. Nor does it conclusively approve every aspect of the district court’s reasoning. It establishes that this unpreserved challenge failed both the obviousness and prejudice requirements.

F. Koch’s Upward Variance: Extraordinary Harm Beyond Ordinary Financial Fraud

Koch emphasized her age of seventy-one, serious health problems, limited criminal history, and lesser role. She characterized the fifteen-year sentence as effectively a death sentence and argued that such a substantial variance was unwarranted in a white-collar fraud prosecution.

The panel’s response was not that these considerations lacked relevance. The district court had considered them but assigned greater weight to the offense’s seriousness and enduring harm. It also noted an earlier fraud-related felony that received no criminal-history points and found virtually no remorse.

The court rejected the characterization of Koch’s participation as merely passive or ancillary. Although less culpable than Hess, she personally participated in the scheme and in dismembering remains. The district court accounted for the difference in culpability by imposing a sentence five years shorter than Hess’s.

Most importantly, the district court concluded that ordinary fraud calculations did not adequately capture the betrayal of families, desecration of remains, and lasting emotional injury. The appellate court deferred to that reasoned assessment.

Precedents supporting deferential review and substantial variances

• United States v. Peña, 963 F.3d 1016 (10th Cir. 2020). This case supported the demanding standard for overturning a substantively challenged sentence: the decision must be arbitrary, capricious, whimsical, or manifestly unreasonable. It underscored that appellate disagreement with the sentencing weight assigned to mitigating facts is not enough.

• United States v. Holt, 161 F.4th 1253 (10th Cir. 2025). The court used this decision’s formulation that a sentence stands unless it exceeds the bounds of permissible choice under the facts and applicable law. That framing allowed a range of reasonable outcomes rather than a single appellate-preferred sentence.

• United States v. Doty, 150 F.4th 1351 (10th Cir. 2025). The court cited its approval of a variance from 33–41 months to 96 months because the offense was atypical and extraordinarily harmful. It demonstrated that substantial departures from the advisory range may be justified by harm inadequately reflected in the calculation.

• United States v. Rocha, 145 F.4th 1247 (10th Cir. 2025). Beyond providing the general review framework, this decision upheld a variance from 27–33 months to 60 months for extraordinary recklessness not fully captured by the Guidelines. It also required deference to the sentencing court’s assessment of the weight of supporting facts.

• United States v. Lawless, 979 F.3d 849 (10th Cir. 2020). This case affirmed a variance from 60 months to 144 months where the arson was far from typical and the conduct particularly disturbing and serious. It also supplied the principle that appellate courts may not simply reweigh the § 3553(a) factors. Koch’s argument, the panel concluded, largely sought precisely that reweighing.

• United States v. Walker, 74 F.4th 1163 (10th Cir. 2023). Its approval of a variance from 84–105 months to 300 months, based partly on severe mental and physical harm, reinforced the legitimacy of considering injuries that a numerical range may inadequately represent.

• United States v. Vazquez-Garcia, 130 F.4th 891 (10th Cir. 2025). This decision upheld a variance from 24–30 months to 48 months based partly on an absence of remorse for child sexual abuse. It supported treating Koch’s lack of remorse as one permissible consideration within the broader sentencing assessment.

These precedents were not numerical formulas authorizing a particular percentage increase. Their common principle was that a substantial variance can be reasonable when supported by individualized findings explaining why the advisory range understates the offense’s seriousness.

The decision also preserves an important distinction between Guidelines loss calculations and sentencing discretion. The court did not hold that emotional injury must be added to the monetary loss amount. Instead, it upheld consideration of nonfinancial harm under § 3553(a) after the advisory calculation had been corrected.