ILANA DIAMOND ROVNER, Circuit Judge.
This appeal raises the question of how “loss” should be calculated under Sentencing Guidelines section 2F1.1 in cases involving a “Ponzi” or pyramid scheme, when defendants have partially repaid fraudulently-obtained funds before detection of the scheme.
I.
Between January 1982 and April 1988, Richard Holiusa and his co-conspirators solicited investments in various companies, representing that they would reinvest the money in silver futures and high-yield government securities. They promised investors that their principal would not be at risk and that they would realize high rates of return. In fact, the funds were used to cover the operating expenses of the various companies and were never reinvested. The conspirators sent investors fraudulent weekly and monthly statements detailing both the principal investment and the interest that had purportedly accrued. They perpetuated the scheme by paying off earlier investors with the money of new investors. Thus, although [1045]*1045a total of $11,625,739 was invested in the fraudulent operation, slightly more than $8,000,000 was returned to investors before the scheme was discovered.
After being charged in a ten-count indictment, Holiusa pled guilty to mail fraud, conspiracy to commit mail fraud, and failure to report a currency transaction. He received a pre-Guidelines sentence of five years on the mail fraud count, and was sentenced under Sentencing Guidelines section 2F1.1 on the remaining two counts. That guideline, which applies to offenses involving fraud, provides for a base offense level of 6, which is to be increased depending on the amount of “loss.” Finding the loss in this case to be the full $11,625,739, the district court increased Holi-usa’s sentence by eleven levels pursuant to section 2Fl.l(b)(l)(L).1 After making various other adjustments, the court arrived at a total offense level of 23. In conjunction with criminal history category I, that offense level produced a sentencing range of 46-57 months. The district court sentenced Holiu-sa at the very top of the range to 57 months, to be followed by the 5 year pre-Guidelines sentence and three years of supervised release.
On appeal, Holiusa contests the district court’s calculation to the extent that it was based on a loss amount of $11,625,739. He argues that because over $8,000,000 was returned to investors, the actual loss was approximately $3,500,000.2 The government contends that the full amount should be considered, even though much of it was returned, because the money was not reinvested as investors had been promised. The district court agreed with that rationale:
In this case the gravity of the completed crime was more substantial than the ultimate loss suffered by the victims. This Court finds in this ease that the defendant never intended to invest the monies taken from the victims; that the intent of this defendant was to defraud all of the victims of their money.
Pursuant to the exhibits and the testimony, this court now finds that the amount of intended or probable loss would be ... $11,625,739.
(Nov. 23, 1992 Tr. at 164-65).
Although the district court’s loss calculation is a factual finding that we review for clear error, the meaning of “loss” for purposes of section 2F 1.1 is a question of law that is subject to de novo review. United States v. Chevalier, 1 F.3d 581, 585 (7th Cir.1993); United States v. Strozier, 981 F.2d 281, 283 (7th Cir.1992).
II.
Section 2F1.1 takes into account more than actual losses. Application Note 7 to the section explains:3
In keeping with the Commission’s policy on attempts, if a probable or intended loss that the defendant was attempting to inflict can be determined, that figure would be used if it was larger than the actual loss.
See also United States v. Schneider, 930 F.2d 555, 556 (7th Cir.1991). In addition to actual loss, then, we must consider the loss that was “probable or intended.” Although “intended” is straightforward enough, “probable,” which was deleted from the note in 1991 (see n. 3), is unclear and might be understood to greatly expand the loss inquiry. The term’s meaning is limited, however, by “attempt,” with which it is twice linked (“In keeping with the Commission’s policy on attempts, if a probable ... loss that the defendant was [1046]*1046attempting to inflict ... )• Thus, as the Third Circuit has explained, “[t]he fraud guideline ... has never endorsed sentencing based on the worst-case scenario potential loss.” United States v. Kopp, 951 F.2d 521, 529 (3d Cir.1991) (emphasis in original). That reading is supported by the fact that when the Commission deleted “probable” in 1991, it did not intend to substantively change the note, but only to “eonform[ ] the wording of Application Note 7 of the Commentary to § 2F1.1 to Application Note 2 of the Commentary of § 2B1.1 to make clear that the treatment of attempts in cases of fraud and theft is identical.” U.S.S.G. Appendix C, Amendment 393. See also Kopp, 951 F.2d at 529; United States v. Bailey, 975 F.2d 1028, 1031 (4th Cir.1992).
In addition, the note directs us to consider the “intended or probable loss that the defendant was attempting to inflict” only if it is greater than the actual loss. Thus, if the defendant intends to take a greater amount than he succeeds in taking before detection of the scheme, he is sentenced for the larger amount. In Strozier, 981 F.2d at 283-85, for example, the defendant had deposited $405,000 worth of bad checks into his bank account, but had withdrawn only $36,-000 before his arrest. We found that a sentence based on the full amount was appropriate because the evidence clearly indicated that Strozier would have withdrawn that amount if his scheme had not been detected.
At the same time, unrealized plans to repay do not reduce the loss amount. If the defendant intended to return the money but did not, then the actual loss is greater than the intended loss and the intended loss becomes irrelevant. As we explained in United States v. Mount, 966 F.2d 262, 266 (7th Cir.1992):
An embezzler who abstracts $10,000 to invest in the stock market causes a “loss” of $10,000 even if he plans to repay before the next audit (to avoid detection) and even if he invests only in blue chip stocks. ******
The embezzler causes loss in the full amount taken, despite plans to repay, because the employer is at risk in the interim and lacks a ready source of recompense.
In contrast to those situations, however, the full amount involved is not considered if the defendant both intended to and did return part of that amount before detection of his scheme.
Free access — add to your briefcase to read the full text and ask questions with AI
ILANA DIAMOND ROVNER, Circuit Judge.
This appeal raises the question of how “loss” should be calculated under Sentencing Guidelines section 2F1.1 in cases involving a “Ponzi” or pyramid scheme, when defendants have partially repaid fraudulently-obtained funds before detection of the scheme.
I.
Between January 1982 and April 1988, Richard Holiusa and his co-conspirators solicited investments in various companies, representing that they would reinvest the money in silver futures and high-yield government securities. They promised investors that their principal would not be at risk and that they would realize high rates of return. In fact, the funds were used to cover the operating expenses of the various companies and were never reinvested. The conspirators sent investors fraudulent weekly and monthly statements detailing both the principal investment and the interest that had purportedly accrued. They perpetuated the scheme by paying off earlier investors with the money of new investors. Thus, although [1045]*1045a total of $11,625,739 was invested in the fraudulent operation, slightly more than $8,000,000 was returned to investors before the scheme was discovered.
After being charged in a ten-count indictment, Holiusa pled guilty to mail fraud, conspiracy to commit mail fraud, and failure to report a currency transaction. He received a pre-Guidelines sentence of five years on the mail fraud count, and was sentenced under Sentencing Guidelines section 2F1.1 on the remaining two counts. That guideline, which applies to offenses involving fraud, provides for a base offense level of 6, which is to be increased depending on the amount of “loss.” Finding the loss in this case to be the full $11,625,739, the district court increased Holi-usa’s sentence by eleven levels pursuant to section 2Fl.l(b)(l)(L).1 After making various other adjustments, the court arrived at a total offense level of 23. In conjunction with criminal history category I, that offense level produced a sentencing range of 46-57 months. The district court sentenced Holiu-sa at the very top of the range to 57 months, to be followed by the 5 year pre-Guidelines sentence and three years of supervised release.
On appeal, Holiusa contests the district court’s calculation to the extent that it was based on a loss amount of $11,625,739. He argues that because over $8,000,000 was returned to investors, the actual loss was approximately $3,500,000.2 The government contends that the full amount should be considered, even though much of it was returned, because the money was not reinvested as investors had been promised. The district court agreed with that rationale:
In this case the gravity of the completed crime was more substantial than the ultimate loss suffered by the victims. This Court finds in this ease that the defendant never intended to invest the monies taken from the victims; that the intent of this defendant was to defraud all of the victims of their money.
Pursuant to the exhibits and the testimony, this court now finds that the amount of intended or probable loss would be ... $11,625,739.
(Nov. 23, 1992 Tr. at 164-65).
Although the district court’s loss calculation is a factual finding that we review for clear error, the meaning of “loss” for purposes of section 2F 1.1 is a question of law that is subject to de novo review. United States v. Chevalier, 1 F.3d 581, 585 (7th Cir.1993); United States v. Strozier, 981 F.2d 281, 283 (7th Cir.1992).
II.
Section 2F1.1 takes into account more than actual losses. Application Note 7 to the section explains:3
In keeping with the Commission’s policy on attempts, if a probable or intended loss that the defendant was attempting to inflict can be determined, that figure would be used if it was larger than the actual loss.
See also United States v. Schneider, 930 F.2d 555, 556 (7th Cir.1991). In addition to actual loss, then, we must consider the loss that was “probable or intended.” Although “intended” is straightforward enough, “probable,” which was deleted from the note in 1991 (see n. 3), is unclear and might be understood to greatly expand the loss inquiry. The term’s meaning is limited, however, by “attempt,” with which it is twice linked (“In keeping with the Commission’s policy on attempts, if a probable ... loss that the defendant was [1046]*1046attempting to inflict ... )• Thus, as the Third Circuit has explained, “[t]he fraud guideline ... has never endorsed sentencing based on the worst-case scenario potential loss.” United States v. Kopp, 951 F.2d 521, 529 (3d Cir.1991) (emphasis in original). That reading is supported by the fact that when the Commission deleted “probable” in 1991, it did not intend to substantively change the note, but only to “eonform[ ] the wording of Application Note 7 of the Commentary to § 2F1.1 to Application Note 2 of the Commentary of § 2B1.1 to make clear that the treatment of attempts in cases of fraud and theft is identical.” U.S.S.G. Appendix C, Amendment 393. See also Kopp, 951 F.2d at 529; United States v. Bailey, 975 F.2d 1028, 1031 (4th Cir.1992).
In addition, the note directs us to consider the “intended or probable loss that the defendant was attempting to inflict” only if it is greater than the actual loss. Thus, if the defendant intends to take a greater amount than he succeeds in taking before detection of the scheme, he is sentenced for the larger amount. In Strozier, 981 F.2d at 283-85, for example, the defendant had deposited $405,000 worth of bad checks into his bank account, but had withdrawn only $36,-000 before his arrest. We found that a sentence based on the full amount was appropriate because the evidence clearly indicated that Strozier would have withdrawn that amount if his scheme had not been detected.
At the same time, unrealized plans to repay do not reduce the loss amount. If the defendant intended to return the money but did not, then the actual loss is greater than the intended loss and the intended loss becomes irrelevant. As we explained in United States v. Mount, 966 F.2d 262, 266 (7th Cir.1992):
An embezzler who abstracts $10,000 to invest in the stock market causes a “loss” of $10,000 even if he plans to repay before the next audit (to avoid detection) and even if he invests only in blue chip stocks. ******
The embezzler causes loss in the full amount taken, despite plans to repay, because the employer is at risk in the interim and lacks a ready source of recompense.
In contrast to those situations, however, the full amount involved is not considered if the defendant both intended to and did return part of that amount before detection of his scheme. As Mount further elucidated:
Both the Sentencing Commission’s notes defining “loss”, see § 2B1.1 (commentary) and § 2F1.1, application note 7, and this court’s cases, e.g., United States v. Schneider, 930 F.2d 555 (7th Cir.1991), call for the court to determine the net detriment to the victim rather than the gross amount of money that changes hands. So a fraud that consists in promising 20 ounces of gold but delivering only 10 produces as loss the value of 10 ounces of gold, not 20. Borrowing $20,000 by fraud and pledging $10,000 in stock as security produces a “loss” of $10,000: “the loss is the amount of the loan not repaid at the time the offense is discovered, reduced by the amount the lending institution has recovered, or can expect to recover, from any assets pledged to secure the loan.” Guideline 2F1.1, application note 7(b).
******
The difference between gross and net loss matters if the offender has paid in part before detection, or the victim has access to a ready source of compensation such as the assets securing the fraudulently-obtained loan.
966 F.2d at 265-66.4
Such is the case here. The full amount invested was not the probable or intended loss because Holiusa did not at any point intend to keep the entire sum. Indeed, return of the money — that is payment of earlier investors with the funds of later inves[1047]*1047tors — was an integral aspect of Holiusa’s scheme, essential to its continuation. And, in line with his intentions, Holiusa returned over $8 million to investors before the scheme was detected. Because he did not intend to and did not keep the full $11.6 million, that amount does not reflect the actual or intended loss, and is not an appropriate basis for sentencing.
The government argues that the full amount should be considered even if all or part of it was returned before discovery of the scheme because the invested funds were “at risk” in the interim. The government cites fraudulent loan cases that have considered as loss the full value of the loan even though the bank was able to recover some of that amount. See United States v. Brack, 942 F.2d 141, 143 (2d Cir.1991) (suggesting in dicta that defendant would be hable for full amount of loan even if he had repaid it before fraud was discovered); United States v. Johnson, 908 F.2d 396, 398 (8th Cir.1990).5 The government also points to our own language in Schneider as supporting its position:
[T]he fact that the victims might have been able to recover some of the money taken from them by enforcing their security interests no more reduced the victims’ loss in the contemplation of the law than if a pickpocket got cold feet and returned his victim’s wallet before the victim discovered it had been missing. The crime is complete when the thief obtains the victim’s property_ What happens later is irrelevant.
930 F.2d at 559.
Notwithstanding that language, however, Schneider’s ultimate holding — that defendants should not be sentenced based on the full value of a fraudulently-obtained contract when they intended to perform the contract — supports the net loss approach.6 Moreover, the Sentencing Commission has since rejected the full value approach. In its November 1991 amendment, it added Application Note 7(b), which provides:
[I]f a defendant fraudulently obtains a loan by misrepresenting the value of his assets, the loss is the amount of the loan not repaid at the time the offense is discovered, reduced by the amount the lending institution has recovered, or can expect to recover, from any assets pledged to secure the loan.
Although the post-sentencing amendment is not binding here, the fact that the Sentencing Commission did not intend the 1991 amendments to substantively change the guideline suggests that this approach is also proper in pre-1991 cases. See United States v. Menichino, 989 F.2d 438, 441-42 (11th Cir.1993) (per curiam); United States v. Smith, 951 F.2d 1164, 1168 (10th Cir.1991); Kopp, 951 F.2d at 534-35. The above-quoted portion of Mount has also since adopted the net loss approach, without suggesting that its analysis was limited to any particular Guidelines version. In his concurring opinion in United States v. Miller, 962 F.2d 739, 747-49 (7th Cir.1992), a case that addressed but did not decide this issue, Judge Flaum also indicated his support for considering only net losses. See also Chevalier, 1 F.3d at 586-87. The same approach has been adopted by the Third Circuit in its extremely well-reasoned opinion in Kopp, 951 F.2d at 527-36, by the Tenth Circuit in Smith, 951 at 1167-68, and [1048]*1048by the Eleventh Circuit in Menichino, 989 F.2d at 441-42.7
III.
Holiusa should not have been sentenced based on amounts that he both intended to and indeed did return to investors. We vacate his sentence and remand for resentenc-ing.