UNITED STATES of America, Plaintiff–Appellee, v. Collins Max CHRISTENSEN, aka Collie Christensen, Defendant–Appellant.
Argued and Submitted June 10, 2013. -- October 11, 2013
Courtney J. Linn (argued) and McGregor W. Scott, Orrick Herrington & Sutcliffe LLP, Sacramento, CA, for Defendant–Appellant.Camil A. Skipper (argued), Assistant United States Attorney, Benjamin B. Wagner, United States Attorney, Russell L. Carlberg, Assistant United States Attorney, Sacramento, CA, for Plaintiff–Appellee.
Collins Max Christensen appeals the sentence imposed by the district court following his pre-indictment guilty plea to one count of wire fraud in violation of 18 U.S.C. § 1343. Christensen's sentence–60 months in prison-was 19 months above the high end of the applicable advisory guideline range and 27 months above the recommended sentence set out in the plea agreement. We have jurisdiction under 28 U.S.C. § 1291, and we affirm.
Christensen waived indictment by a grand jury and pleaded guilty to a one-count information charging wire fraud. The plea agreement revealed that, between 2006 and 2008, Christensen managed the operations of at least six land development companies, which—in turn—managed multiple real estate projects spearheaded by Christensen. By soliciting individual investors, Christensen received a total of approximately $2,385,959 from fourteen individuals who agreed to invest in one or more of Christensen's projects. Over time, Christensen diverted a significant amount of the investors' funds, using the diverted funds for undisclosed purposes. Indeed, Christensen admitted that he misappropriated approximately $985,994 of investors' funds using the interstate wires to further his criminal scheme. Christensen further admitted that he diverted investors' funds not only for use on undisclosed real estate projects but also for his own personal use.
In the Presentence Investigation Report (“PSR”), the probation officer summarized the losses sustained by the various “victims” of Christensen's offense of conviction. Only those persons who had some or all of their investment funds unlawfully diverted by Christensen were listed as “victims” of Christensen's offense. Consistent with the plea agreement, the total “victim” loss reported by the probation officer was $985,994.
Given the loss amount reported in the PSR and admitted by Christensen, the probation officer calculated a total offense level of 20, including a three-level reduction for acceptance of responsibility. With a criminal history category of I, the recommended sentencing range under the Guidelines was 33 to 41 months. Christensen agreed with the probation officer's Guidelines calculations.
The probation officer included in the PSR a sampling of the written statements that Christensen's “victims” submitted to the court. In his informal objections to the PSR, which were submitted to the probation officer before sentencing, Christensen complained that one of the victim impact statements was “misleading.” He did not provide specifics, and he did not otherwise object to the content of any of the victim impact statements quoted in the PSR. After the probation officer amended the PSR in response to Christensen's informal objections, Christensen did not again complain about the victim impact statements.
One day before sentencing was scheduled to begin, the district court advised counsel that-for “a number of reasons”—it was considering an upward variance to Christensen's sentence. To that end, the district judge requested that the government provide information concerning the amount of investor monies that Christensen diverted to his own personal use. Given the last-minute notice concerning a possible upward variance, Christensen's counsel requested and obtained a continuance of the sentencing hearing.
As directed by the district court, the government filed a spreadsheet itemizing Christensen's use of $507,805 for personal expenses. Among other things, the spreadsheet revealed that Christensen used investor monies for personal investment properties held in his wife's name, to pay his more–than–$13,000–per–month home mortgage, to make payments to his ex-wife, to make a car payment for his daughter, to pay his daughter's college tuition, and to gamble at a casino in Biloxi, Mississippi. According to the FBI, $507,805 was a “conservative figure” for the amount of diverted funds used for Christensen's personal expenses.
Christensen did not contest the accuracy of the personaluse transactions listed in the government's spreadsheet. Christensen instead argued—both in a written sentencing memorandum and in open court at sentencing—that an upward variance based on his personal use of diverted funds would constitute improper double-counting because the total loss amount—$985,994—had already resulted in a 14–level increase in Christensen's offense level pursuant to U.S.S.G. § 2B1.1(b)(1)(H). The district court was unpersuaded by Christensen's double-counting argument.
At the first of two sentencing hearings, the district court overruled Christensen's informal objections to the PSR and, on the record, adopted the findings in the amended PSR, determining them to be “true and correct as modified.” Included in the findings adopted by the district court was the total loss—$985,994—resulting from Christensen's criminal offense. Although a number of Christensen's victims spoke at that first hearing, Christensen failed to call the district court's attention to any false or misleading information provided by those victims.1 Indeed, Christensen offered no objection at all to the victims' statements.
At the second sentencing hearing, after hearing from both counsel and Christensen himself, the district court explained its reasons for an upward variance as follows:
The sentencing factors require the Court to impose a sentence that is sufficient, but not greater than necessary to comply with the purposes set forth in 18 U.S.C. Section 3553(a)(2). And if the Court is going to vary, the Court has to state its reasons for varying upward.
And focusing on the nature and circumstances of the offense, and the nature and characteristics of the defendant, ․ it concerns the Court ․ that in effect [Christensen] used his positions to influence innocent victims to invest at a time when he knew that he was not going to use the money for those purposes, without disclosing that to them. He diverted $985,994, which has been the agreed upon sum in terms of loss in this case.
After adding that (1) Christensen apparently learned nothing from a 28–year–old felony conviction for obtaining money by false pretenses and (2) Christensen's numerous victims included “vulnerable, retirement age victims, victims that trusted Mr. Christensen,” the district court continued:
Mr. Christensen indicated that he did not set out to bilk his investors. I accept that. But, again, that's only half the story. At some point he did consciously and intentionally decide to bilk his investors, and that conduct continued for a period of years, and it involved a number of different investors in the same type of conduct.
And I found interesting Mr. Christensen's statement to the Court in which he explained, at least attempted to explain why he did what he did. And he wrote in there that he's the type of person in which he believes that failure is not an option. I've heard that phrase over and over again by many people. And in this case, it clearly was an option. It's clearly what you should have done. You should have accepted failure. Sometimes you learn more from your failures than you do from your successes. But it certainly would have saved not only your family, but all of these victims who have made statements, submitted loss statements to the Court, from having to postpone their retirement, have their marriages destroyed, lose their jobs or lose their homes.
The Court believes, therefore, for those reasons, and specifically given the—as in Schlueter, the fact that I don't think the sentencing guideline range adequately accounts for the harm that Mr. Christensen's fraud caused his victims, I believe that given the egregiousness of his conduct, including lying, covering up, using funds for personal purposes, destroying the victims' lives, cheating victims out of significant sums of money that they needed, and taking advantage of personal relationships, that the guideline range doesn't adequately account for the harm that his conduct has caused. There is a need—despite the fact that there has been almost 28 years between his last conviction, there still is a need I believe in this case to protect the public from further crimes.
For all these reasons, the Court believes an upward variance is appropriate and, given the loss in this case, that a sentence of 60 months is appropriate and sufficient, but not greater than necessary to satisfy the sentencing requirements imposed by the Court.
At the conclusion of the sentencing hearing, Christensen offered only two objections to the sentence imposed. Christensen first objected to the purported lack of notice regarding the basis for an upward variance. Specifically, Christensen asserted that he had not been given notice that the court might base an upward variance on the victims' statements. He did not assert that the district court committed procedural error by relying on clearly erroneous facts, speculative evidence, and unreliable victim statements. Second, Christensen objected to the district court's alleged “discounting” of his acceptance of responsibility. He offered no other objections to the sentence imposed.
Christensen raises a number of issues, only two of which were properly preserved for appeal—namely, whether the district court committed procedural error by imposing a nonGuidelines sentence based on factors already incorporated into the Guidelines and whether the district court improperly “discounted” Christensen's acceptance of responsibility. We find no merit to either issue.
We review a sentence for reasonableness; “only a procedurally erroneous or substantively unreasonable sentence will be set aside.” United States v. Carty, 520 F.3d 984, 993 (9th Cir.2008) (en banc). “Procedural errors include, but are not limited to, incorrectly calculating the Guidelines range, treating the Guidelines as mandatory, failing to properly consider the [18 U.S.C.] § 3553(a) factors, using clearly erroneous facts when calculating the Guidelines range or determining the sentence, and failing to provide an adequate explanation for the sentence imposed.” United States v. Armstead, 552 F.3d 769, 776 (9th Cir.2008). Christensen has only challenged the procedural reasonableness of his sentence.
We review the district court's construction and interpretation of the Guidelines de novo, United States v. Nielsen, 694 F.3d 1032, 1034 (9th Cir.2012), and the district court's application of the Guidelines for abuse of discretion, United States v. Holt, 510 F .3d 1007, 1010 (9th Cir.2007). We review the district court's factual determinations for clear error. United States v. Tulaner, 512 F.3d 576, 578 (9th Cir.2008).
We review the first of Christensen's preserved issues—whether the district court engaged in impermissible double-counting—for abuse of discretion. See Holt, 510 F.3d at 1010. As the Fifth Circuit correctly noted in United States v. Williams, 517 F.3d 801 (5th Cir.2008), “[t]he Supreme Court's decision in [United States v. Booker, 543 U.S. 220 (2005),] implicitly rejected the position that no additional weight could be given to factors included in calculating the applicable advisory Guidelines range, since to do otherwise would essentially render the Guidelines mandatory.” Id. at 809 (internal footnote omitted). “This necessarily means that the sentencing court is free to conclude that the applicable Guidelines range gives too much or too little weight to one or more factors, either as applied in a particular case or as a matter of policy.” Id. Here, when selecting an above Guidelines sentence for Christensen, the district court was not prohibited from considering the extent to which the Guidelines did not sufficiently account for the nature and circumstances of Christensen's offense, including the amount of the loss, the number of victims, or the harm to the victims, even though the Guidelines account for these factors either implicitly or explicitly, to some extent. The district court did not err by concluding that the Guidelines did not properly take into account the harm caused by “the egregiousness of [Christensen's] conduct, including lying, covering up, using funds for personal purposes, destroying the victims' lives, cheating victims out of significant sums of money that they needed, and taking advantage of personal relationships.” The district court's conclusion is supported by the record and did not constitute impermissible double-counting.
We review Christensen's second preserved issue—whether the district improperly discounted his acceptance of responsibility-for clear error. United States v. Cortes, 299 F.3d 1030, 1037 (9th Cir.2002). Christensen was, in fact, awarded a 3–level reduction for acceptance of responsibility; the district court expressly denied that it discounted “in any way the fact that [Christensen] accepted responsibility;” and the record otherwise belies Christensen's suggestion that the district court “turned his acceptance of responsibility on its head by using it as a justification for the upward variance.” To determine that the district court clearly erred, we must be “left with the definite and firm conviction that a mistake has been committed.” Easley v.. Cromartie, 532 U.S. 234, 242 (2001). Mere evidence that Christensen received a 3–level reduction, but ended up with an above Guidelines sentence is insufficient, in this case, to leave us with a “definite and firm conviction” that the district court erred. Rather, the district court acknowledged that Christensen was entitled to an acceptance of responsibility reduction and based the upward variance on Christensen's behavior wholly independent of his acts constituting acceptance of responsibility. We have no basis for questioning whether the district court did, both in word and in fact, award Christensen a 3–level reduction for acceptance of responsibility.
Christensen contends that the district court committed procedural error by, among other things, (1) failing to resolve factual conflicts in the PSR regarding victim impact and loss amounts as required by Federal Rule of Criminal Procedure 32(i)(3)(B); (2) failing to provide advance notice of the precise grounds for the 19–month upward variance in his sentence in violation of the Due Process Clause; and (3) improperly relying on clearly erroneous facts, speculative evidence, and unreliable victim statements, also in violation of the Due Process Clause. Because Christensen failed to raise these alleged procedural errors before the district court, we review for plain error. United States v. Burgum, 633 F.3d 810, 812 (9th Cir.2011).
To secure reversal under the plain error standard, Christensen must show that: (1) there was error; (2) the error was plain; and (3) the error affected Christensen's substantial rights. United States v. Joseph, 716 F.3d 1273, 1277 (9th Cir.2013). “If these three conditions [are satisfied], [we] may exercise [our] discretion to notice a forfeited error that ․ seriously affects the fairness, integrity, or public reputation of judicial proceedings.” Id. (internal quotation marks omitted) (quoting United States v. Ameline, 409 F.3d 1073, 1078 (9th Cir.2005) (en banc)). A sentencing error prejudices the substantial rights of a defendant when there is a reasonable probability that he would have received a different sentence had the district court not erred. Id. at 1280. The defendant bears the burden of showing a reasonable probability that he would have received a different sentence absent the error. Id.
When a defendant fails to make specific allegations of factual inaccuracy in a PSR, a district court has no obligation under Rule 32(i)(3)(B). United States v. Petri, No. 11–30337, 2013 WL 1490604, at *7 (9th Cir. Apr. 12, 2013). Because Christensen never made specific factual objections to the PSR regarding victim impact and loss amounts, Rule 32 was never triggered. Id. We find that the district court committed no error, much less plain error, under Rule 32(i)(3)(B).
We likewise find that the district court committed no error, much less plain error, by failing to provide advance notice of the precise grounds upon which the the 19–month upward variance to Christensen's sentence was based. A district court is not required—either by the Federal Rules of Criminal Procedure or by the Due Process Clause—to give advance notice of its intent to impose a sentence outside the advisory Guidelines range. Irizarry v. United States, 553 U.S. 708, 713–16 (2008). It follows that, if a district court is not required to give advance notice of its intent to vary upward, it is also not required to give advance notice of the precise grounds upon which it might base an upward variance.
To the extent Christensen complains, in general, that the district court erred by taking into account the “uncorroborated,” “unsworn,” and “untested” statements of victims, his claim of error is without merit. The Federal Rules of Evidence do not apply at a sentencing hearing. Fed.R.Evid. 1101(d)(3). Indeed, “a sentencing judge may appropriately conduct an inquiry broad in scope, largely unlimited either as to the kind of information he may consider, or the source from which it may come.” Nichols v. United States, 511 U.S. 738, 747 (1994) (internal quotation marks omitted). By statute, there is “[n]o limitation ․ on the information concerning the background, character, and conduct of a person convicted of an offense which a court of the United States may receive and consider for the purpose of imposing an appropriate sentence.” 18 U.S.C. § 3661. The Federal Rules of Criminal Procedure, moreover, provide that the presentence report must contain “information that assesses any financial, social, psychological, and medical impact on any victim.” Fed.R.Crim.P. 32(d)(2)(B). As a general matter, then, a district court may consider victim impact statements, whether sworn or not, at sentencing. United States v. Santana, 908 F.2d 506, 507 (9th Cir.1990) (per curiam).
To the extent Christensen complains that the district court procedurally erred and his due process rights were violated by the district court's purported reliance on “erroneous facts” provided by the victims, we are presented with a closer question. Specifically, Christensen points to statements made by the district court at sentencing regarding victim losses and impacts. The district court referred, for example, to the statement of one victim (Jennifer R.) who said she invested her entire life savings ($330,000) in one of Christensen's projects, only to lose it all and her 19–year marriage to boot. In the PSR, that particular victim's loss—or, more precisely, the portion of her overall investment that was attributed to Christensen's diversion of investor funds ($90,000) that was lost—was reported to be only $23,017. The district court noted that another investor had been unable to retire as planned, although the PSR stated that the portion of that victim's overall investment that was subject to misappropriation ($20,000) only resulted in a loss of $5,496. The district court also referred to the statement of another individual (W.H.Mehr), who said that he would not have invested his “$128,000 plus dollars” had Christensen been truthful with him. Mehr was not listed as a “victim” in the PSR at all, and his losses were not included in the $985,994 loss figure reported in the PSR, apparently because his investment dollars were not diverted by Christensen. Finally, the district court repeated the claims of various victims regarding the impacts they suffered—foregone vacations, delayed retirement, a ruined marriage, sale of the family homes, lost savings—without separating (perhaps an impossible task) the impacts attributable to Christensen's criminal behavior from the impacts attributable to non-criminal investment losses.
Christensen bears the burden of showing that the district court relied on clearly erroneous facts, affecting his substantial rights, when either (1) calculating the Guidelines range or (2) determining his sentence. See Armstead, 552 F.3d at 776. Christensen does not argue that the district court relied on any clearly erroneous fact in calculating the Guidelines range. In fact, Christensen agreed with the calculated Guidelines range, including the loss amount that the district court accepted, and reiterated numerous times as the basis for the loss calculation in the case, in arriving at that calculation—$985,994. Thus, the only question is whether the district court relied on any clearly erroneous fact in determining the sentence. Again, to find clear error, we must be “left with the definite and firm conviction that a mistake has been committed.” Easley, 532 U.S. at 242. We are left with no such conviction that a mistake has been committed in this case.
It is clear that, at sentencing, the district court expressly adopted the findings reported in the PSR, including the $985,994 loss suffered by Christensen's “victims.” During its recitation of reasons for varying upward, moreover, the district court expressly stated that (1) Christensen “diverted $985,994, which has been the agreed upon sum in terms of loss in this case;” (2) “this case ․ involved almost a million dollars;” and (3) “the Court believes an upward variance is appropriate ․ given the loss in this case.” It thus seems clear that the district court based its variance on “the agreed upon sum in terms of loss in this case,” a sum—$985,994—that included only victim losses attributable to Christensen's criminal conduct.
In addition to explaining that the upward variance was based on “the agreed upon sum in terms of loss in this case,” the district court iterated the following additional reasons for its upward variance: (1) Christensen took advantage of personal relationships to solicit large sums of investment monies from people who trusted him; (2) Christensen used over $500,000 of investor monies to enrich himself—to invest in properties held in his wife's name, to pay his home mortgage, to make payments to his ex-wife, to pay his daughter's college tuition, and to gamble at a casino in Biloxi, Mississippi; (3) Christensen repeatedly and continually lied to his victims, not only about the true state of project affairs but also about the uses to which he was putting their investment dollars, enabling him to keep his unlawful scheme going for a period of years; (4) Christensen's unlawful scheme caused his victims to lose sums of money they could ill afford to lose; and (5) Christensen apparently learned nothing from a 28–year–old unscored conviction for obtaining money by false pretenses. Christensen did not and does not dispute the factual basis for the above reasons iterated by the district court. Each of those reasons, moreover, constitutes a proper sentencing factor under 18 U.S.C. § 3553(a), and together they provide sufficient support for the district court's decision to vary upward.
The district court included another reason for the upward variance when it referred to the “life-destroying impacts” described by Christensen's victims. As noted by the district court, these victim-reported impacts included loss of homes, a failed marriage, postponement of retirement, foregone vacations, loss of life-savings, and loss of the ability to pay for children's private schooling. The victims reported the impacts they suffered from having lost all of their investment dollars, without differentiating losses that were solely attributable to Christensen's diversion of investor funds. These “life-destroying impacts” undoubtedly went beyond the stipulated losses to investors based on Christensen's diversion of funds.
These “life-destroying impacts,” however, were proper for the district court to consider even if not tied to the loss Christensen caused by misappropriating investor funds. There is “[n]o limitation ․ on the information concerning the background, character, and conduct of a person convicted of an offense which a court ․ may receive and consider for the purpose of imposing an appropriate sentence.”2 18 U.S.C. § 3661; see also Pepper v. United States, 131 S.Ct. 1229, 1240 (2011) (“Permitting sentencing courts to consider the widest possible breadth of information about a defendant ‘ensures that the punishment will suit not merely the offense but the individual defendant.’ ”) (quoting Wasman v. United States, 468 U.S. 559, 564 (1984)). These “life-destroying impacts,” supported by victim statements, provide greater insight into Christensen's “background, character, and conduct” that the district court was entitled to rely on in determining that for a specified loss resulting from criminal conduct, the Guidelines did not adequately account for the seriousness of Christensen's offense, provide adequate deterrence, or sufficiently protect the public and innocent investors from the infliction of further harm at the hands of Christensen.3 See 18 U.S.C. § 3553(a)(2). Moreover, it was not the dollar amount of these victims' losses that the district court relied on in imposing this upward variance; rather, it was the intangible nature of Christensen's conduct. As the district court stated:
I believe that given the egregiousness of his conduct, including lying, covering up, using funds for personal purposes, destroying the victims' lives, cheating victims out of significant sums of money that they needed, and taking advantage of personal relationships, that the guideline range doesn't adequately account for the harm that his conduct has caused. There is a need—despite the fact that there has been almost 28 years between his last conviction, there still is a need I believe in this case to protect the public from further crimes.
In addition, we note that many of the “life destroying impacts” were indivisible. For example, Jennifer R. attributed the dissolution of her 19–year marriage to Christensen's conduct. The dissent argues that it was inappropriate for the district court to consider this divorce because the PSR indicates that Jennifer R. lost only $23,017 due to Christensen's fraud. Dis. Op. at 27–29. In so doing, the dissent assumes Jennifer R.'s divorce was a matter of simple arithmetic: it resulted from the loss of a certain sum of money, and that sum must have been greater than $23,017; therefore, because Jennifer R. lost only $23,017, the district court was required to ignore her claim that Christensen's conduct had caused her divorce. This argument ignores the indivisible nature of the harm. The fact of the matter is that we do not know how much financial stress Jennifer R.'s marriage could have withstood. What we do know is that Jennifer R. and her husband entrusted their life savings to Christensen, that he fraudulently diverted a portion of it, that their marriage never recovered from the stress, and that neither the causes nor the impacts of a divorce break down neatly into dollars and cents. This is precisely the type of situation in which the Guidelines do not adequately account for the seriousness of the offense. As a result, it was proper for the district court to consider Jennifer R.'s divorce, together with the other “life destroying impacts” described by Christensen's victims.
If the district court had used non-offense “losses” to erroneously calculate the Guidelines range, this would be a much different case. But, where as here, the district court used the correct loss calculation to arrive at the correct Guidelines range, and appears to have only relied on extrinsic evidence as support for the Guidelines inadequate accounting for the harm caused by Christensen, we cannot say that the district court clearly erred in any manner. Therefore, we conclude that the district court committed no error in this regard.
Even assuming that the district court did err, Christensen has fallen short of establishing that his substantial rights were affected by any such error. Indeed, while the record provides ample support for the district court's 19–month upward variance, the record provides no basis for concluding that there is a reasonable probability Christensen would have received a more lenient sentence absent the asserted error.
At the conclusion of the sentencing hearing, the district court expressly stated that it was “imposing a sentence that was less than what I think the Court could have imposed given the egregiousness of the conduct in this case,” conduct which—in effect—amounted to stealing over a long period of time from people whose trust Christensen maintained through lies. The district court also stated that “[t]his could have been a sentence that was far greater than 60 months.” The district court said nothing to suggest that it would have imposed a lesser sentence if the overstated impacts of Christensen's egregious criminal conduct had not been considered. At best, it is highly uncertain whether Christensen would have received a lesser sentence if the district court had not considered those impacts, and such a high degree of uncertainty precludes a finding that any alleged error affected his substantial rights. See Jones v. United States, 527 U.S. 373, 394–95 (1999) (“Where the effect of an alleged error is so uncertain, a defendant cannot meet his burden of showing that the error actually affected his substantial rights.”); see also United States v. Gonzalez -Aguilar, 718 F.3d 1185, 1189 (9th Cir.2013) (explaining that a defendant cannot meet the third prong of the plain error test by demonstrating a mere “possibility” that he could have obtained a lesser sentence absent the alleged error); United States v. Rodriguez, 627 F.3d 1372, 1382 (11th Cir.2010) (explaining that “where the effect of an error on the result in the district court is uncertain or indeterminate—where we would have to speculate—the appellant has not met his burden of showing a reasonable probability of a different result” (internal quotation marks omitted)); United States v. Lorenzo, 995 F.2d 1448, 1457 n .4 (9th Cir.1993) (noting that “if plain error analysis applies, it appears that the appellants ․ pay the price for the inadequacy of the record”). Thus, even assuming error, it did not affect Christensen's substantial rights.
Likewise, Christensen's due process argument is unavailing. The Due Process Clause requires that a defendant not be sentenced on the basis of “misinformation of constitutional magnitude.” United States v. Tucker, 404 U.S. 443, 447 (1972). To establish that his due process rights were violated, Christensen must show that materially false or unreliable information was demonstrably made the basis for the sentence imposed by the district court. United States v. McGowan, 668 F.3d 601, 606 (9th Cir.2012). Here, because Christensen failed to raise his “erroneous facts” issue before the district court, Christensen must show not only that the district court committed error under the Due Process Clause but also that the error affected his substantial rights. As we have shown, Christensen cannot satisfy this burden.
Having found no merit to any of the issues raised by Christensen, we AFFIRM.
We review a district court's sentencing decisions for abuse of discretion, thus affording district courts a substantial degree of deference. But, as we recently observed en banc, “[t]he abuse of discretion standard is deferential, but it does not mean anything goes.” United States v. Ressam, 679 F.3d 1069, 1087 (9th Cir.2012) (en banc). And, as the Supreme Court has observed in a related context, we should “not indulge [in] post hoc rationalization” of the sentencing court's “decisionmaking that contradicts the available evidence․” Harrington v. Richter, 131 S.Ct. 770, 790 (2011). In short, “[t]he abuse of discretion standard of review is not a rubber stamp of all sentencing decisions made by a district court, United States v. Ruff, 535 F.3d 999, 1005 (9th Cir.2008) (Gould, J., dissenting), and we should not turn a blind eye when a district court distorts the sentencing process. Because that is precisely what the majority does, I respectfully dissent.
In affirming the sentence in the present case, the majority obfuscates a fact that is apparent from any fair reading of the record: the district court based its above-guidelines sentence on investor losses not caused by Christensen's criminal conduct.
This was “significant procedural error.” United States v. Carty, 520 F.3d 984, 993 (9th Cir.2008) (en banc). The district court either mistakenly believed that the losses in question did result from Christensen's criminal conduct, in which case the sentence was predicated on “clearly erroneous facts,” id. (internal quotation marks omitted), or it knowingly increased the sentence based on the impact of non-criminal conduct, a proposition that not even the government contends would be justified. Because victim impact formed the central basis for the upward variance, there is a “reasonable probability” that the district court's error affected the outcome of the proceedings. United States v. Whitney, 673 F.3d 965, 972 (9th Cir.2012) (internal quotation marks omitted). For these reasons, I believe Christensen has demonstrated plain error entitling him to relief.
As an initial matter, it is necessary to clarify the nature of the victims and their losses in this case. The presentence investigation report (“PSR”) identified fourteen victims of Christensen's criminal conduct. The PSR explained the basis on which this determination was made:
The victims of this offense are the numerous individuals who invested money into the various projects the defendant managed and who had their investment funds diverted for other uses which they did not authorize. It is noted victim statements were received from other individuals who lost investment funds due to the failure of all of Christensen's projects. As their funds were properly invested into the projects they designated and were not diverted, they are not victims for the purposes of this criminal matter.
Consistent with this approach, for each of the fourteen victims, the PSR separately listed the amount of money invested with Christensen, and the portion of that investment fraudulently diverted and lost as a result of the criminal conduct. Across the fourteen victims, the total amount of money lost due to Christensen's fraud was $985,994, representing less than half of the more than $2.1 million invested by these individuals.
The district court accepted the PSR's calculation of $985,994 as “the agreed upon sum in terms of loss in this case.” Indeed, this was the figure employed in calculating Christensen's advisory guidelines sentencing range and was the amount of restitution that Christensen was eventually ordered to pay. As described below, however, the district court disregarded the PSR's methodology in assessing the impact of Christensen's conduct on individual victims, conflating losses suffered due to the fraud and losses suffered from legitimate (i.e., non-criminal), but failed investments.
Although one would never know it from the majority opinion, the district court based its upward variance on the impact of Christensen's conduct on individual investors, specifically five victims identified in the PSR and one investor who was not a victim at all. This is abundantly clear from the record.
At the outset of explaining the basis for its sentence, the district court stated that it had “researched a number of cases” in which sentencing courts had varied upward “under similar circumstances.” The district court then described in detail three out-of-circuit cases that it considered particularly analogous, all involving defendants convicted of financial fraud who received above-guidelines sentences on the basis of victim impact. The district court characterized the final case it discussed, United States v. Schlueter, 634 F.3d 965 (7th Cir.2011), as the “most instructive.”1 According to the district court, the upward variance in Schlueter was imposed (and affirmed) because “the guideline range failed to adequately account for the harm that [the defendant's] fraud had caused his victims or for the egregiousness of his conduct.” The victims in Schlueter were friends of the defendant, many of retirement age, and, in the district court's words, the defendant “[took] advantage of personal relationships to cheat them out of significant sums that they needed at critical stages of their lives,” including “one victim [who] was unable to purchase a home because of the fraud.”
After recounting the details of Schlueter, the district court explained that it “went back to the victim impact statements and the presentence report to compare, in effect, Mr. Christensen to the defendant in Schlueter.” It was at this point that the district court described the alleged impact of Christensen's conduct on six specific investors. The district court prefaced this discussion by stating that it was “looking at the victims' losses [and] statements.”
For at least three of the six investors, however, the district court considered the impact of losses well beyond those caused by Christensen's criminal conduct. For instance, the district court described Robert G. as a victim who “has been unable to retire ․ as planned; he's had to continue to work.” Yet, according to the PSR, Robert G. only lost $5,496 due to the fraud. It is unlikely that this narrow loss was the cause of Robert G.'s inability to retire, especially given that Christensen repaid all but $192 of that loss more than two years prior to the sentencing hearing.
The district court also discussed an investor named Mehr, who apparently lost $128,000 through investing in one of Christensen's projects. The district court noted Mehr's statement that “if he had known and been told the truth, he obviously would not have invested in the project.” But Mehr is not listed in the PSR as a victim. Rather, as the government concedes, he is one of the investors not included in the PSR because his funds were properly invested and not diverted.
Finally, and most importantly, is Jennifer R., whom the district court described as “the victim that concerns me the most.” Jennifer R. submitted a letter to the probation officer indicating that she lost $330,000 of her life savings from investing with Christensen, which led to the disintegration of her marriage. The PSR, however, lists Jennifer R. as having lost only $23,017 due to Christensen's fraud.2 Nevertheless, in weighing the impact of Christensen's conduct, the district court described Jennifer R.'s experience as follows:
She invested $330,000 of her life savings, her and her husband's, which they had saved for 17 years. This was going to be their retirement. They were invited to Mr. Christensen's home. He befriended them, he talked them into investing and, as a result, they lost the entire amount of their life savings. And her marriage, as she writes, could not handle the loss. So not only did she lose her entire life savings, her 19–year marriage has ended as a result of the defendant's criminal behavior.
For all three of these investors (and likely two more of the remaining three),3 it is evident that the district court based its analysis on losses unrelated to Christensen's criminal conduct. These losses may have been caused by Christensen's poor, but non-criminal, management, the victims' poor judgment in investing with him, the deterioration of the economy, or some combination thereof. None of these are reasons why Christensen was convicted of a crime.
The district court's consideration of these losses can be explained in only one of two ways: (1) it mistakenly believed that the full investment losses were attributable to the fraud; or (2) it realized that just a portion of the losses resulted from the fraud, but it nonetheless knowingly based its upward variance on the impact of non-criminal conduct. For reasons described below, the record supports the former interpretation, but under either scenario, the district court committed significant procedural error.
If the district court operated under the erroneous belief that the full losses of Robert G., Mehr, and Jennifer R. were attributable to Christensen's crimes, there can be little dispute that the sentence was predicated on “clearly erroneous facts.” Carty, 520 F.3d at 993 (quoting Gall v. United States, 552 U.S. 38, 51 (2007)). Indeed, the government has never contested the accuracy of the PSR's loss calculations for these investors. The district court's statements at sentencing make plain that it did proceed under this incorrect assumption.
As described, the district court immediately led into its discussion of the six investors by stating that it was “looking at the victims' losses.” Then, with regard to Jennifer R., the most important victim in the district court's calculus, the district court stated that she lost her entire life savings of $330,000, as well as her 19–year marriage, “as a result of the defendant's criminal behavior.” (Emphasis added.) The district court could not have made a more direct statement indicating that it was treating and relying on full investment losses (including non-criminal losses) as attributable to Christensen's criminal activities.
Nearly as compelling an indication is the district court's reference to Mehr, who was not a victim at all. The district court's discussion of the six investors, including Mehr, followed its long description of the purportedly analogous cases in which sentencing courts had varied upward on the basis of victim impact. In fact, the district court used the word “victim” ten times in the three pages of the sentencing transcript preceding its discussion of these investors. It is inconceivable that the district court deemed it relevant to discuss Mehr's losses in this context while fully cognizant that Mehr was not a victim.
The majority offers no refutation of this evidence in holding that the district court did not premise its sentence on clearly erroneous facts. Instead, the majority attempts to elide the factual basis underlying the upward variance by pointing to two passing references that the district court made, later in the sentencing hearing, to the PSR's aggregate loss calculation of $985,994. Maj. Op. at 16. From these references, the majority reasons that the district court surely predicated its variance only on losses caused by the fraud. Id. at 16–17. This analysis ignores the reality that the district court's variance was based not on aggregate losses, but on the impact on individual victims. The district court emphasized as much over and over at sentencing. In evaluating those individual victims, the district court relied on clearly erroneous facts, and it thereby committed significant procedural error.
If, on the other hand, the district court appreciated that the referenced investor losses were not attributable to Christensen's crimes, this would necessarily mean that the district court intentionally rested its sentence, in part, on the impact of non-criminal conduct. Otherwise, the district court's extended discussion of the full investor losses would have been without purpose. At oral argument, however, the government expressly disclaimed reliance on any notion that the district court could permissibly base its upward variance on non-criminal conduct. Hence, if the majority is correct that the district court did not rely on erroneous facts in misperceiving the nature of the losses, the only remaining possibility depends on an argument that the government wisely has elected not to pursue, and that we should not advance in its stead. See United States v. Anekwu, 695 F.3d 967, 985 (9th Cir.2012) (arguments not raised by parties are waived).
Yet, because the majority has not clearly disavowed the possibility that the district court may have varied upward based on the impact of non-criminal conduct, reluctantly, I find it necessary to address that question. I believe it must be answered in the negative. In the analogous context of “relevant conduct” under U.S.S.G. § 1B1.3, used in calculating a defendant's guidelines offense level, the Courts of Appeals are unanimous in holding that the provision is limited to conduct that is criminal. See United States v. Catchings, 708 F.3d 710, 712, 720 (6th Cir.2013); United States v. Griffith, 584 F.3d 1004, 1013 (10th Cir.2009); United States v. Schaefer, 291 F.3d 932, 940–41 (7th Cir.2002); United States v. Dove, 247 F.3d 152, 155 (4th Cir.2001); United States v. Jain, 93 F.3d 436, 443 (8th Cir.1996); United States v. Peterson, 101 F.3d 375, 385 (5th Cir.1996); United States v. Dickler, 64 F.3d 818, 830–31 (3d Cir.1995). The reasoning behind this limitation applies equally here, i.e., to the basis for a variance. To allow for an upward variance based on non-criminal conduct “would allow individuals to be punished by having their [sentence] increased for activity which is not prohibited by law but merely morally distasteful or viewed as simply wrong by the sentencing court.” Peterson, 101 F.3d at 385.
The sentencing factors of 18 U.S.C. § 3553(a)(2) confirm this conclusion. With the exception of the rehabilitative goals of clause (D), all of the factors focus on the unlawful nature of the defendant's past or future behavior. The factors speak of the seriousness of “the offense,” the need to provide just punishment for “the offense,” the need to deter “criminal conduct,” and the need to protect the public from “further crimes.” See 18 U.S.C. § 3553(a)(2). The impact of a defendant's completed, non-criminal activities—in this case, losses caused by poor investment decisions—is not relevant to any of these inquiries. And given that sentencing courts must impose a sentence that is “sufficient, but not greater than necessary to comply with the purposes” of the § 3553(a)(2) factors, there is no place for increasing a defendant's sentence on the basis of this consideration. Id. § 3553(a) (emphasis added).
The majority's response on this point is perplexing. The majority seems to concede that it would have been improper for the district court to vary upward based on the magnitude of the losses caused by Christensen's non-criminal conduct. In other words, the majority seems to concede that the district court would have erred in stating, “I am varying upward because Christensen caused over $2 .1 million in investor losses, including over $1.1 million in losses not related to the fraud.” But the majority nonetheless contends that it was appropriate to vary upward based on the “life-destroying impacts” of those same losses. Maj. Op 18–19. This is a classic distinction without a difference.4
In its limited explanation, the majority vaguely references 18 U.S.C. § 3661, which allows sentencing courts to receive any information concerning a defendant's “background, character, and conduct.” Maj. Op. 19. However, this provision, which “remov[es] typical ‘jury trial’ evidentiary limitations” on the information that may be presented and considered at sentencing, United States v. Booker, 543 U.S. 220, 249 (2005), says nothing of the grounds upon which a district court may grant an upward variance. To the extent the majority implies that all information countenanced under § 3661 may serve as the basis for a variance, such is patently absurd. For then, a sentencing court could vary upward based on a defendant's eating or dressing habits, the tradition or school of yoga he favors, or the regularity with which he recycles, all of which “provide greater insight into [the defendant's] ‘background, character, and conduct.’ “5 Maj. Op. 18–19.
For these reasons, I would hold that it is significant procedural error to rest an upward variance on the impact of a defendant's lawful activities and conduct.
Christensen must also establish prejudice, and to do so, he must demonstrate that there is a “reasonable probability” that the district court's error affected the outcome of the sentencing. Whitney, 673 F.3d at 972 (quoting United States v. Marcus, 130 S.Ct. 2159, 2164 (2010)). Ordinarily, this would be a difficult task given the inherent challenges in identifying “the effect of any discrete set of errors on a district court's multifaceted sentencing determination.” Id. Here, however, Christensen is presented with no such difficulties given that the district court made clear that the issue on which it committed error provided the central basis for its upward variance.
In seeking to minimize the import of this error, the majority paints a grossly distorted picture of the district court's decisionmaking process. Rather than informing the reader of the district court's unambiguous emphasis on victim impact, the majority relays every snippet of the sentencing hearing it can locate that would suggest other considerations played a role in the sentencing decision. See Maj. Op. 7–9, 16.
The majority alternatively holds that, even if the district court did err in overstating the impact of Christensen's criminal conduct, Christensen cannot demonstrate prejudice because it is “uncertain” whether the outcome would have been different if not for the error. Maj. Op. 21–22. The majority's analysis in this regard places an incorrect, and virtually impossible to meet, burden on a defendant under plain error review. To demonstrate prejudice, Christensen need not show with absolute certainty that he would have received a lesser sentence in the absence of the error, he need only show a “reasonable probability” of such. See Whitney, 673 F.3d at 972; see also United States v. Gonzalez–Aguilar, 718 F.3d 1185, 1189 (9th Cir.2013). Given the overriding role that victim impact played at sentencing, I have little difficulty finding a reasonable probability that the district court would not have varied to the degree that it did—nineteen months above the high end of the guidelines range and almost double the low end of the guidelines range—had it not erred.
Indeed, to not find a reasonable probability of prejudice when a district court's error undermines the primary justification for its sentence makes a mockery of the requirement that the sentencing court adequately explain its decision, a requirement that is of particular importance in the context of a variance. See Gall, 552 U.S. at 46. It is true that, in the present case, the district court evidenced a desire to reach its eventual outcome regardless of the reasons. Prior to the second sentencing hearing, the district court indicated to the parties that it was considering varying on the basis of Christensen's personal use of the funds diverted. It was only after Christensen submitted briefing as to why it would be inappropriate to vary on this basis that the district court suddenly shifted gears and changed its justification at the second hearing, without advance notice, to victim impact. However, it would be perverse to find a lack of prejudice because we suspect that a district court was bent on reaching a particular outcome. The integrity of the sentencing process requires that we take a district court at its word as to the reasons for the sentence imposed. See id. at 50 (“[A district court] must adequately explain the chosen sentence to allow for meaningful appellate review and to promote the perception of fair sentencing.”). Here, those reasons were fundamentally flawed.
The district court committed “significant procedural error” in imposing an upward variance on the basis of investor losses that were not attributable to Christensen's criminal conduct. This was prejudicial error, and Christensen is entitled to resentencing. Accordingly, I would vacate the sentence and remand for resentencing. I respectfully dissent.
1. Christensen did object to the government's reading of a “victim impact statement” from W.H. Mehr. Christensen noted for the record that Mehr was not, in fact, listed as a “victim.” The district court responded to the objection by stating: “So noted.” Mehr's investment–$128,571.42–was not included in the $985,994 loss attributed to Christensen's criminal conduct.
2. Despite this plain and capacious language, the dissent insists that information considered under 18 U.S.C. § 3661 must be criminal in nature. See Dis. Op. at 31–34. We disagree. The Supreme Court “[has] recognized that ‘the broad language of § 3661’ does not provide ‘any basis for the courts to invent a blanket prohibition against considering certain types of evidence at sentencing.’ “ Pepper v. United States, 131 S.Ct. 1229, 1241 (2011) (quoting United States v. Watts, 519 U.S. 148, 152 (1997)). Therefore, subject to constitutional constraints, see id . at 1240 n.8, we take Congress at its word: a sentencing court may rely on any evidence relating to a defendant's background, character, and conduct when considering the sentencing factors found in 18 U.S.C. § 3553(a). See Pepper, 131 S.Ct. at 1236 (holding that a sentencing court may consider post-sentencing rehabilitation).Our decision in United States v. Mercado, 474 F.3d 654 (9th Cir.2007), cited by the dissent, does not alter this conclusion. See Dis. Op. at 34 n.5. In Mercado, the district court considered the criminal activity charged in several acquitted counts when it sentenced the defendant on the counts for which he was convicted. 474 F.3d at 655–56. We affirmed, holding that “the district court could constitutionally consider the acquitted conduct .” Id. at 657. Thus, although Mercado establishes that a district court may consider acquitted conduct, it does not suggest that information considered under § 3661 must be criminal in nature.
3. The dissent worries that our holding might allow sentencing courts to vary upward based on “the tradition or school of yoga [a defendant] favors.” Dis. Op. at 34. Of course, this fanciful scenario is not before us. The question presented here is much easier: Does 18 U.S.C. § 3661 authorize a sentencing court to consider life-destroying impacts that flow directly from a substantially criminal scheme? As explained above, the answer is “yes.”
1. The other two cases were United States v. Siman, 419 F. App'x 979 (11th Cir.2011), and United States v. Scherrer, 444 F.3d 91 (1st Cir.2006). The district court characterized Siman as a case where the sentencing court varied upward “relying on the impact experienced by the victims and who the victims were.” Similarly, the district court described Scherrer as a case where the sentencing court “focus[ed] on the victims” in imposing a variance.
2. The PSR lists Jennifer R. as having invested a total $90,000 with Christensen. It is not entirely clear why there is a discrepancy between this amount and the $330,000 described in her letter to the probation officer, although it may be that the additional funds were invested in projects from which Christensen did not fraudulently divert any funds (and therefore were not included the PSR's calculations). In any event, the relevant comparison is the $23,017 loss established by the PSR and the $330,000 figure on which the district court expressly relied.
3. For two of the other three victims discussed, William K. and James D., there was also a sizeable gap between the amount invested and the amount lost due to the fraud (an over $90,000 difference for William K. and a $51,000 difference for James D.). The district court did not mention specific dollar amounts lost by these two investors, but its discussion again suggested that it was considering total investment loss, not loss caused by the fraud. For only one of the six investors, Twila K., was the amount invested the same as the amount lost as a result of the criminal conduct.
4. The majority charges thatthe dissent assumes Jennifer R.'s divorce was a matter of simple arithmetic: it resulted from the loss of a certain sum of money, and that sum must have been greater than $23,017; therefore, because Jennifer R. lost only $23,017, the district court was required to ignore her claim that Christensen's conduct had caused her divorce.Maj. Op. 20. But I make no such assumption; instead, I take the district court at its word. It stated: “So not only did she [Jennifer R.] lose her entire life savings, her 19–year marriage has ended as a result of the defendant's criminal behavior.” (Emphasis added.) Even more disturbing than the majority's refusal to confront the facts is its assertion that because “many of the ‘life destroying impacts' were indivisible,” id., we need not and should not even attempt to parse out what harm was caused by a defendant's criminal conduct.
5. Our decision in United States v. Mercado, 474 F.3d 654 (9th Cir.2007), provides an example of when information received under § 3661 can be pertinent to a defendant's criminal activities. There, we cited § 3661 in holding that sentencing courts may consider criminal conduct charged in the indictment, but of which the defendant was acquitted, in calculating the defendant's guidelines range. See id. at 656–57. This scenario centers on the conduct at issue being criminal, requiring that the sentencing court find by a preponderance of the evidence that the defendant is culpable for the charged, but acquitted conduct. See id. at 655–56.
STAFFORD, Senior District Judge: