The appeal of Martin Geevers, who pleaded guilty to one count of bank fraud arising out of a check kiting scheme, requires us to determine once again when application of the Sentencing Guidelines may result in the imposition of a sentence on the basis of intended loss when the actual loss was significantly less. Geevers argues that because a passer of worthless checks could not possibly abscond with the full face amount of his worthless deposits, the District Court erred in calculating his intended loss under a "worst case" scenario. Though Geevers's argument possesses strong intuitive appeal, we will uphold the District Court's full face amount finding.

We base our conclusion on three separate considerations. First, we note that there is a distinction between intending a loss and expecting a loss. While we agree that Geevers may not have reasonably expected to extract the full face value of his fraudulent checks from the banks, it does not necessarily follow that he did not intend to extract every cent possible. Second, the commentary to S 2F1.1 makes clear that losses "need not be determined with precision." U.S.S.G. S 2F1.1 Application Note 9. A district court is therefore not barred from considering the full amount of the fraudulent checks to be the intended loss although that figure may overstate the actual intended figure. Finally, our precedent allows some limited burden shifting in the proving of intended loss under the guidelines. We have previously recognized that though the government bears the burden of proof in guidelines cases, the burden of production may shift to the defendant once the government presents prima facie evidence of a given loss figure. See United States v. Evans, 155 F.3d 245, 253 (3d Cir. 1998). Under this regime, intended loss does not equal the face value of the deposited checks as a matter of law. Rather, a defendant is free to proffer evidence about his or her true intentions in order to rebut the presumption that his or her fraudulent deposits may create. A district court does not, however, commit error when, in the absence of sufficient evidence to the contrary, it fixes the guidelines range based upon a presumption that the defendant intended to defraud the banks of the full face amount of the worthless checks.

Geevers also contends that if the District Court correctly calculated the intended loss figure from his check passing activities, he still should have received a three-level reduction in his guidelines calculation because he had not completed his attempt. This argument raises questions about the interpretation of U.S.S.G. S 2X1.1 and its relation to the guideline on intended loss, but we reject Geevers's contention. The guideline clearly precludes granting a downward departure in situations in which the attempted conduct was prevented solely through the intervention of the victim or law enforcement. Because such is the case here, we conclude that Geevers is ineligible for the downward departure.

Pursuant to a plea agreement, Geevers pleaded guilty to a violation of 18 U.S.C. S 1344 (bank fraud) for opening a bank account at Bankers Savings bank in Woodbridge, New Jersey, with a $75,000 check drawn on a closed account at Merrill Lynch. Merrill Lynch advised Bankers Savings that the check was not covered by sufficient funds, and Bankers Savings closed Geevers's account before he attempted to draw any funds on the account. Though this transaction produced no loss to the involved banks, it was just one of many attempts by Geevers to profit by fraudulently inflating bank balances. Between 1996 and 1997, according to the Presentence Investigation Report ("PSI"), Geevers repeatedly opened accounts in various banks by depositing checks from closed accounts or accounts with insufficient funds and then attempted to withdraw or transfer a portion of the deposited funds before the victim banks realized that the funds were not backed.

All told, including both offense conduct and relevant conduct recounted in the PSI, Geevers deposited or sought to cash checks with face values approximating $2,000,000 in total. Prior to his apprehension, he attempted to withdraw or transfer about $400,000. He actually managed to withdraw or transfer over $160,000. The PSI also included as relevant conduct the losses arising from a fraudulent real estate scheme he perpetrated several years earlier.

The parties agreed that the sentence would be calculated under U.S.S.G. S 2F1.1, which is the guideline covering offenses of fraud or deceit.1 For purposes of calculating the sentence, the government argued that the loss amount relevant for sentencing should be the total face value of Geevers's deposited checks, which is the potential loss. Geevers maintained that he did not intend to cause the full amount of the potential loss because he could not have successfully withdrawn those funds even if he had wanted to. In the alternative, he maintained that because he did not complete the acts necessary to effect that loss, he was at least entitled to a three-level reduction of his offense score under S 2X1.1, the guideline pertaining to attempts.

The District Court disagreed and adopted a loss figure of $2,188,575 and rejected the downward adjustment under S 2X1.1. Geevers's base offense level under U.S.S.G. S 2F1.1 of 6 was therefore increased by 12 levels for a loss in excess of $1.5 million and also by 2 levels for more than minimum planning. It was then reduced 3 levels for acceptance of responsibility under S 3E1.1. The resulting offense level of 17 carried an imprisonment range between 30 and 37 months, and the District Court imposed a sentence in the middle of the range--33 months.

We have jurisdiction to review Geevers's claim that the District Court incorrectly applied the sentencing guidelines under 18 U.S.C. S 3742(a)(2). See United States v. Shoupe, 988 F.2d 440, 444 (3d Cir. 1993). Our review of the District Court's interpretation and application of the guidelines is plenary, but where the District Court's application is based on factual conclusions, we will reverse only if its conclusion is clearly erroneous. See United States v. Hallman, 23 F.3d 821, 823 (3d Cir. 1994).

We first consider the question whether the District Court erred in considering the face amount of Geevers's fraudulent checks in determining the intended loss of his scheme under S 2F1.1.2 The section establishes a base offense level of 6 for crimes involving fraud or deceit, and sets forth a range of possible increases tied to the amount of loss that the crime involved. The commentary to the guidelines, which is binding, see Stinson v. United States, 508 U.S. 36, 38 (1993), explains that a sentencing court is to consider intended loss in determining the loss figure. "Consistent with the provisions of S 2X1.1 (Attempt, Solicitation, or Conspiracy), if an intended loss that the defendant was attempting to inflict can be determined, this figure will be used if it is greater than the actual loss." U.S.S.G. S 2F1.1 Application Note 8. Note 9 explains that "the loss need not be determined with precision. The court need only make a reasonable estimate of the loss, given the available information."

At his sentencing hearing, Geevers contended that the face amount of the deposited checks cannot be the figure employed in sentencing because no reasonable check kiter would think that he or she could get away with withdrawing the full face amount of the checks. Pointing to the difficulty of calculating a more precise figure of intended loss, the District Court rejected Geevers's argument.

I think I grasp your argument, but I find it difficult to ascertain how a Court would be able to determine what the intended loss was in a check kiting scheme, except to take the defendant at his actions, which is to count the full amount of any fraudulent checks from the time that the fraudulent check is deposited. I mean, there's no rule of thumb that the Court can come up with and say well, your typical check kiter is only going to net 5 percent or 15 percent of what he's deposited by way of phoney checks. And really that's what you're asking the Court to try to figure out.

App. 58.

Geevers renews this argument on appeal. His argument can be broken into two related contentions: 1. The government has not proved that he intended to take the face value of the false checks; and 2. He could not have possibly taken that much, and the intended amount should therefore not reflect an impossible amount. We will consider these contentions in turn, but first we pause to address the import of United States v. Torres, 209 F.3d 308 (3d Cir. 2000), filed after oral argument in this case.

In Torres, the defendant, posing as another individual, opened a money market account at a bank with the deposit of a worthless third-party check in the value of $240.65. On the same day, another party to the conspiracy, using the same false name, deposited a stolen U.S. Treasury check in the amount of $66,021.94 in the same account. The defendant was arrested after attempting to withdraw $24,900 from the account and was convicted of one count of bank fraud in violation of 18 U.S.C. S 1344. See id. at 310. The district court calculated his attempted loss as being $66,262.59, that is, the combination of the two deposits, despite the defendant's contention that the proper amount was the $24,900 he actually attempted to withdraw. We affirmed, ruling that "[i]t was eminently reasonable for the District Court to infer that Torres intended to withdraw the balance of the deposits before the stolen check surfaced as stolen and would have done so had he not been arrested." Id. at 312.

Though it affirmed the entire calculation of the loss amount, the Torres panel did not address whether there was any difference between the intent behind depositing a worthless check, the issue before us, and depositing a stolen one. For purposes of that appeal, any distinction would have been irrelevant to the holding, as it was the value of the stolen check (in excess of $66,000) that placed the defendant into the challenged sentence range, while the worthless check (worth only $240.65) was not of sufficient value to affect the guidelines. Indeed, the latter issue was not briefed in the case.3 Moreover, though the text of the opinion refers to the deposited check as "rubber," the facts as recited in the opinion do not clarify whether the check deposited was a pure fabrication or an attempt to draw down the account of a third party. Additionally, the Torres panel was not presented with the issue of how to address an elaborate false check scheme with the complexity of Geevers's.

Torres would thus not appear to be dispositive. At all events, the extent of the precedential impact of Torres is ultimately unimportant as we reach the same conclusion as did that panel. Because there are differences between a stolen and a kited check, we find it helpful to analyze and explain why the District Court's conclusions were appropriate under this Court's precedent.

Aside from implications from Torres, our precedent offers no clear direction on how to address a defendant in Geevers's position.4 Nor do the guidelines or the accompanying commentary specifically address this situation.5 Our sister circuits are also divided on the matter. Compare, e.g., United States v. Watkins, 994 F.2d 1192 (6th Cir. 1993) (remanding for findings whether defendant intended to withdraw or could have withdrawn face value of deposited checks), with United States v. Strozier, 981 F.2d 281 (7th Cir. 1992) (upholding equation of face value of checks with intended loss).

Though our pre-Torres precedent is unclear, it does offer some helpful guideposts. It is clear that a district court errs when it simply equates potential loss with intended loss without deeper analysis. In United States v. Kopp, 951 F.2d 521 (3d Cir. 1991), we considered the applicability of S 2F1.1 to fraudulent misrepresentations to obtain a bank loan, holding that loss should not be equated with potential loss at the time of the crime. See id. at 536. In reaching this conclusion, we stated that "[t]he fraud guideline thus has never endorsed sentencing based on the worst-case scenario potential loss . . ." Id. at 529 (emphasis in the original). Moreover, we declared that equating "possible loss" with "probable or intended loss" was a "linguistic stretch," which we rejected. Id. at 533. Because Kopp dealt with an equation of the amount obtained with actual or intended loss, we made no explicit holding on the meaning of intended loss, but the quoted passages suggest that potential loss should not be equated with the intended loss.

In United States v. Yeaman, 194 F.3d 442, 460 (3d Cir. 1999), we applied Kopp to explain that "[i]ntended loss refers to the defendant's subjective expectation, not to the risk of loss to which he may have exposed his victims." Yeaman makes clear, therefore, that the government's burden is to prove intended, not possible, loss if it seeks to increase the guideline levels faced by the defendant under S 2F1.1.

Though Kopp and Yeaman appear to aid Geevers's cause, the argument that intended loss is not per se equivalent to potential loss only takes him so far. While intended loss may not be automatically determinable based on what the potential loss is, intended loss may still equal potential loss. The District Court must determine Geevers's subjective intention, and it can draw inferences from the nature of the crime that he sought to perpetrate.

We therefore must resolve the question whether a reasonable inference may be drawn that a defendant in Geevers's position intends to cause the full loss of the face value of his false checks. We hold that such an inference may be made. Concomitantly, we conclude that the matter is not to be determined as a question of law, but as one of fact.

It seems likely that a defendant in Geevers's position does not expect to obtain the full amount of his fraudulent checks. Common sense suggests that a check kite will always be incomplete, and that a kiter will either abscond or be discovered before exhausting the kite. But expectation is not synonymous with intent when a criminal does not know what he may expect to obtain, but intends to take what he can. We believe that a sentencing court may plausibly conclude that a defendant like Geevers would likely have taken the full amount of the deposited checks if that were possible. Indeed, in one of Geevers's transactions, as the government points out, Geevers was actually able to cause a loss worth 85 percent of the amount he deposited.

To assume that Geevers did not want it all is to assume that had one of the banks somehow failed to detect his fraud and started sending Geevers monthly balance reports, Geevers would have refrained from taking any more of the money. Given Geevers's conduct, the District Court could reject this proposition as unlikely. Though he may not have expected to get it all, he could be presumed to have wanted to.6 As the District Court explained, "the actual loss sustained by the victim becomes a matter only of how quickly and how effectively the victim shuts the barn door before the rest of the horses have gotten out." App. 88. We conclude that it is not an error of law for a court to draw inferences from the face value of the checks in arriving at the factual conclusion that the defendant intended to let all the horses out if possible. If the preceding is correct, then Geevers is free to come forward with evidence to demonstrate that he actually intended something less, but the government has made its prima facie case.

Our ruling is consistent with our prior precedent that has addressed the government's burden of proof when proving subjective intent. In United States v. Evans, 155 F.3d 245, 253 (3d Cir. 1998), we stated that though the defendant does not have to "prove the negative" vis-a-vis his absence of intent, "the burden of production shifts to the defendant once the government has made out a prima facie case." See also United States v. Raven, 39 F.3d 428, 434-35 (3d Cir. 1994) (once the government makes a prima facie case in sentencing, the party challenging the government's case has the burden of coming forward with evidence at a sentencing hearing that shows the government's evidence is incomplete or incorrect). Of course, the burden of persuasion always remains with the government. See Evans, 155 F.3d at 253; Raven, 39 F.3d at 435. As discussed above, intent need not jibe perfectly with expectation when a defendant may not know precisely how much he or she will get away with before discovery, so the amount on the fraudulently deposited checks may be used by a district court in determining whether the government's burden has been met in establishing a given intended loss figure.

There remains the issue of why the face value of the deposited checks may be used to make the prima facie case that Geevers intended the full loss when even Geevers's most harmful transaction considered for sentencing purposes only caused eighty-five percent of the potential loss. The answer to this question begins with the admonition of Application Note 9 of S 2F1.1, which tells us that loss need not be proven precisely. "The court need only make a reasonable estimate of the loss, given the available information." Therefore, because the face value of the deposited checks is germane to analysis of intent, the government's presentation of the face value as evidence may make a prima facie case that the defendant intended to cause the full loss of those amounts. Applying this figure is not unfair to the defendant insofar as it represents a sum that was entirely under his control when he elected to act in contravention of the law. Once the government presented this information to the District Court, therefore, we conclude that the Court was free to accept the loss figure in the absence of persuasive evidence from Geevers that his intent was to steal a lesser amount.

To be clear, the face value of the deposited checks is not to be mechanically assumed to be the intended loss. We merely hold that a sentencing court may consider that as sufficient evidence that it was the intended loss. A defendant may, of course, produce evidence of his or her own in an attempt to convince the court that another figure was intended. It is easy to imagine circumstances where this may be so. For example, if a man needed $10,000 for surgery for his wife and sought to acquire the sum by engaging in a check kite, he might make a worthless deposit of $50,000 in order to inflate his "balance" to a high enough level that the bank would honor a $10,000 check. Such a defendant would likely be able to demonstrate that his subjective intent was only to take $10,000.

It is certainly possible, therefore, for a defendant to convince a sentencing court that he or she did not in fact intend any actual loss. Geevers failed to make such arguments here, beyond his claim that he could not have expected to get it all and a statement by his attorney at sentencing that "somewhere in his mind [Geevers] believed he was going to catch up." App. 104. He also did not present any evidence of what his actual intent was, which he could have requested to do under Fed. R. Crim. P. 32(c). As a result, the District Court was left with only Geevers's assertions of what a "sensible check kiter could . . . have reasonably expected." App. 88. Without having been provided by the defendant with an alternative reasonable basis on which to arrive at a more precise intended loss figure, the District Court did not commit clear error in making the guidelines calculation that it did.7

Allowing district courts to consider the face value of checks as probative of intended loss avoids one negative consequence of the reading urged by Geevers: the potential equation, for culpability purposes, of disparate forms of conduct. We think that a defendant who falsifies checks for large sums of money is more culpable than one who does so for lesser sums. Were we to adopt Geevers's position, however, this distinction would be eroded. Adopting his argument would equate the culpability of someone who falsifies checks for large sums of money with that of one who writes false checks for small amounts in situations where there is no differential in the amount successfully stolen. Given the creation of a risk to financial institutions that attends the presentment of such checks, we view such a result as a troubling one that should be avoided to the extent possible and consistent with the legal framework surrounding the guidelines.8

Our approach enjoys the added advantage of consistency with the duty of sentencing courts to consider the goals of affording adequate deterrence to criminal conduct, see 18 U.S.C. S 3553, because it provides incremental levels of deterrence against criminals who may be considering inflating the face values of the worthless checks that they plan to deposit. If sentencing courts are allowed to sentence based on the value of the falsely deposited check, check kiters will learn that the more they write, the more they risk, potentially reducing the face value of false deposits. As it is generally true that the amount that a kiter is able to steal will be a fraction of the amount deposited, deterring large fraudulent deposits will serve to reduce the amount of money ultimately lost to such frauds. As for the reverse danger, over deterrence, we do not think the potential of over-deterring the deposit of false checks is of particular concern.

Geevers also argues that he cannot be ascribed with intending to take money that would have been impossible to take. The question whether it was truly impossible for Geevers to abscond with the full face value of his fraudulent deposits does not appear to have been litigated at any length in the District Court. Assuming arguendo, however, that these assertions are correct, we do not view them as necessitating a different outcome. Rather, we join the majority of courts of appeals in holding that impossibility is not in and of itself a limit on the amount of intended loss for purposes of calculating sentences under the guidelines. The majority rule is that impossibility does not require a sentencing court to lower its calculation of intended loss. Compare, e.g., United States v. Klisser, 190 F.3d 34, 36 (2d Cir. 1999) (per curiam); United States v. Blitz, 151 F.3d 1002, 1010 (9th Cir. 1998); United States v. Studevent, 116 F.3d 1559, 1563 (D.C. Cir. 1997); United States v. Wai-Keung, 115 F.3d 874, 877 (11th Cir. 1997), with United States v. Galbraith, 20 F.3d 1054, 1059 (10th Cir. 1994); United States v. Watkins, 994 F.2d 1192, 1196 (6th Cir. 1993).

As a matter of guidelines interpretation, we conclude that the majority position is correct. Intent is intent. Impossibility bears on what is reasonable for the person to have intended to do, but the language in Application Note 8 of S 2F1.1 does not give any indication that adventuresome or deluded criminals are to have their actions interpreted differently than those of their more sober counterparts. Nor does the attempt guideline, S 2X1.1, make any provision for consideration of impossibility.

Moreover, Application Note 11 of S 2F1.1 specifically provides an escape valve for situations in which the intended loss may "overstate the seriousness of the offense." The note provides an example that specifically contemplates a situation of impossibility. "[Overstatement] may occur, for example, where a defendant attempted to negotiate an instrument that was so obviously fraudulent that no one would seriously consider honoring it." In light of the specific contemplation of impossibility in the context of departures, grafting an impossibility exception on to the setting of the offense level is inconsistent with the language and structure of the guidelines. Cf. United Steelworkers of Am. v. North Star Steel Co., Inc., 5 F.3d 39, 43 (3d Cir. 1993) ("[A] statute's provisions should be read to be consistent with one another, rather than the contrary.").9

We therefore conclude that the purported impossibility of Geevers's absconding with the full amount of his worthless deposits did not render erroneous the District Court's factual conclusions about Geevers's intended loss.

Finally, Geevers argues that if the District Court was correct to consider the face value of his deposited checks in calculating his intended loss, he should at least receive the benefit of U.S.S.G. S 2X1.1(b)(1), which provides for a three-level guideline reduction for attempts. Geevers maintains that he qualifies for this reduction because he had not drawn the full amount of money that he falsely deposited at the time of his arrest. The District Court based its denial of this argument on both the guideline's applicability and its terms:

For the record, I do not believe that it applies, but assuming for the sake of argument that 2X1.1 applies to any degree, given the factual circumstances here, then I do find as a matter of fact, that the 3-point downward adjustment is not available because it was the intervention of a third party, either apprehension by the bank or law enforcement, that prevented . . . the full flowering of the acts of the offense from occurring.

App. 64.

It is not clear from the record whether Geevers should be viewed as having pleaded guilty solely to a completed crime or an attempt. Section 2X1.1(b)(1) provides for the reduction for an attempt "unless the defendant completed all the acts the defendant believed necessary for successful completion of the substantive offense or the circumstances demonstrate that the defendant was about to complete all such acts but for apprehension or interruption by some similar event beyond defendant's control." Application Note 2 explains that " `[s]ubstantive offense,' as used in this guideline, means the offense that the defendant was convicted of soliciting, attempting, or conspiring to commit." The government contends that Geevers was not convicted of an attempt, but rather the substantive crime of bank fraud set forth in 18 U.S.C. S 1344, but this does not dispose of the issue. The statute incorporates both the completed act and the attempt. See 18 U.S.C. S 1344 ("Whoever knowingly executes, or attempts to execute, a scheme or artifice . . . .") (emphasis added). Indeed, the District Court did not clearly state whether it thought that the offense pleaded to resolved the issue. See App. 57 ("18 U.S.C. Section 1344 contains both the completed substantive offense and the attempt. So the offense statute does not really tell us one way or the other whether we go to 2X."). The Court further acknowledged that Geevers's crime "theoretically, [ ] could be viewed as a series of attempts to some degree." App. 63.

Moreover, Application Note 10 of S 2F1.1 directs the sentencing court to S 2X1.1.

In the case of a partially completed offense (e.g., an offense involving a completed fraud that is part of a larger, attempted fraud), the offense level is to be determined in accordance with the provisions of S 2X1.1 . . . whether the conviction level is for the substantive offense, the inchoate offense . . . or both; see Application Note 4 in the Commentary to S 2X.1.

We therefore conclude that S 2X1.1 may apply to Geevers. We turn then to consideration of whether a defendant in Geevers's shoes fits under the language of the guideline.

Section 2X1.1(b)(1) states that there is no reduction if "the circumstances demonstrate that the defendant was about to complete all such acts but for apprehension or interruption by some similar event beyond defendant's control." Geevers argues that with respect to the offense to which he pleaded guilty, the depositing of the worthless $75,000 check at Bankers Savings, he had never attempted to draw funds and had even returned the checks that had been issued him. This argument for applying S 2X1.1(b)(1) fails based on the facts as represented by Geevers. By Geevers's own admission, Bankers Savings closed his account after Merrill Lynch, which had hosted the closed account on which Geevers attempted to draw, notified Bankers Savings that the check was not backed by sufficient funds. This intervention by a third party prevented Geevers from even attempting to draw on his worthless check.

There was no legal error in the District Court's consideration of this fact, nor clear error in the factual finding that followed: that Geevers would have completed his intended fraud but for the intervention of a third party. See United States v. Strozier, 981 F.2d 281, 286 (7th Cir. 1992) (applying similar analysis). Even more damaging to Geevers's argument is the District Court's explicitfinding during the colloquy that Banker's Savings closed Geevers's account upon the warning of Merrill Lynch that Geevers's deposit was not backed by adequate funds, thereby placing Geevers's conduct squarely within the limiting language of S 2X1.1(b) for attempts interrupted by events beyond the defendant's control.

Nor does the language of Application Note 4 aid Geevers. It provides:

In certain cases, the participants may have completed (or have been about to complete but for apprehension or interruption) all of the acts necessary for the successful completion of part, but not all, of the intended offense. In such cases, the offense level for the count . . . [is the greater of the offense level for the completed offense minus three or the offense level for the completed part of the offense]. For example, where the intended offense was the theft of $800,000 but the participants completed (or were about to complete) only the acts necessary to steal $30,000, the offense level is the offense level for the theft of $800,000 minus 3 levels, or the offense level for the theft of $30,000, whichever is greater.

Because the District Court concluded that Geevers would have taken the rest of the money but for the interruption by law enforcement and the intervention of the victim banks, this commentary is inapplicable.

We acknowledge that an extreme reading of this application note could be taken to mean that failure to complete the full range of acts necessary to complete the fraud automatically demands the reduction, notwithstanding the interference of third parties. Such a reading would, however, clash directly with both the text of the guideline, see S 2X1.1(b)(1), and another part of the official commentary. The "background" note to S 2X1.1 further explains (with emphasis added):

In most prosecutions for conspiracies or attempts, the substantive offense was substantially completed or was interrupted or prevented on the verge of completion by the intercession of law enforcement authorities or the victim. In such cases, no reduction of the offense level is warranted. Sometimes, however, the arrest occurs well before the defendant or any co-conspirator has completed the acts necessary for the substantive offense. Under such circumstances, a reduction of 3 levels is provided . . . .

This note, in conjunction with the guideline's text, makes clear that interruption, except in the early planning stages, precludes the three-level reduction. On the facts before us, it was not clear error for the District Court to have concluded that the intervention by the banks was not"well before" the completion of the necessary acts for the substantive offense.10 See United States v. Torres, 209 F.3d 308, 312 (3d Cir. 2000).

For the foregoing reasons, the judgment of the District Court will be affirmed.