PATRICK J. SCHILTZ, District Judge.
Defendants William Garfield Sandison and Curtis Alan Martinson pleaded guilty to conspiring to commit bank fraud in connection with the failure of a real-estate development known as the Ramsey Town Center ("RTC"). The parties agree that Sandison and Martinson committed bank fraud. But the parties disagree about the amount of loss caused by that bank fraud — a crucial factor in determining the length of the sentences recommended by the United States Sentencing Guidelines ("USSG" or "Sentencing Guidelines"). The government puts the amount of loss at $6,617,000, while the defendants put it at zero. The parties also dispute a number of other sentencing factors, but those disputes are largely overshadowed by the dispute over the amount of loss.
The Sentencing Guidelines themselves contemplate that, in an unusual case, a defendant can commit bank fraud without causing any loss. See USSG § 2B1.1 cmt. n. 3(E). For the reasons described below, the Court finds that this is just such a case. Although the defendants successfully executed a scheme to defraud a number of banks, the defendants did not cause those banks to lose any money. Therefore, the Court agrees with Sandison and Martinson that, for purposes of the Sentencing Guidelines, the amount of loss in this case is zero.
Sandison was the president, and Martinson was the vice president, of Community National Bank ("CNB"). CNB arranged financing for the RTC — a $1.3 billion development that was intended to create a town center of residential, commercial, retail, office, and civic buildings for the growing city of Ramsey, Minnesota. This case arises out of a series of loans that funded the RTC before the project collapsed in about 2006.
To finance the RTC, CNB arranged and administered a participation loan, in which multiple banks jointly funded a $35 million loan to the developer. As it turned out, the participation loan was grossly inadequate for the RTC's needs. Sandison and Martinson therefore arranged to loan more money to the developer. First, Sandison (as well as his son and another CNB officer) personally issued a $990,000 line of credit to the developer, and secured that line of credit by placing a mortgage on RTC property. Sandison did not tell any of the participating banks about the loan or the mortgage, even though he was required to do so under the participation-loan agreements. Later, Sandison and Martinson (and others) personally loaned an additional $6 million to the developer and secured that loan with yet another mortgage. Sandison and Martinson informed the participating banks of this loan and mortgage, but they did so in an incomplete and misleading manner.
Sandison and Martinson ensured that the loans that they and their friends personally extended to the RTC developer — $617,000 that was advanced against the $990,000 line of credit, and the $6 million loan — would be repaid before the $35 million participation loan. The participating banks did not know that Sandison and Martinson had essentially cut in line by obliging the RTC developer to repay their loans before the participation loan. When the RTC went into foreclosure, all of the participating banks suffered large losses on the $35 million participation loan, but Sandison and Martinson emerged largely unscathed. Because of the way that they had manipulated RTC's repayment obligations, Sandison and Martinson — unlike the participating banks — got their money back before the RTC project collapsed.
The parties are represented by excellent attorneys, and those attorneys have worked hard to reach a stipulation regarding the relevant details of the crimes committed by Sandison and Martinson. The Court is grateful for the parties' assistance. The parties' stipulation is as follows:
Stipulation of Fact [Docket No. 90].
Sandison and Martinson each pleaded guilty to one count of conspiring to commit bank fraud — Sandison on August 26, 2009, and Martinson on October 16, 2009. The Probation Office circulated a draft presentence-investigation report ("PSR") for Sandison on December 11, 2009, and a draft PSR for Martinson on December 22, 2009. Each defendant filed numerous objections to his draft PSR. The Probation Office responded to those objections and circulated revised PSRs for the defendants in March 2010. Each defendant then filed numerous objections to his revised PSR — Sandison on April 9, 2010 [Docket No. 69] and Martinson on April 13, 2010 [Docket No. 73]. Each defendant made both legal and factual objections.
After months of negotiations, the government and the defendants filed the stipulation quoted above on November 11, 2010. Docket No. 90. That stipulation resolved some, but not all, of the defendants' factual objections. The government further addressed the defendants' factual objections in a letter to the Court dated November 26, 2010. Docket No. 96.
On November 30, 2010, the Court conducted a lengthy sentencing hearing to address the many unresolved legal and factual objections to the revised PSRs. At the hearing, the defendants withdrew certain factual objections, and the government concurred with other factual objections made by the defendants. At the conclusion of the hearing, all of the defendants' legal objections and a couple of their factual objections remained to be resolved. At the Court's request, Martinson and the government filed post-hearing letters addressing one aspect of Martinson's objections. Docket Nos. 99 and 100.
According to Sandison's revised PSR, the Sentencing Guidelines recommend a sentence of 70 to 87 months, based on an offense level of 27 and a criminal-history category of I. According to Martinson's revised PSR, the Sentencing Guidelines recommend a sentence of 57 to 71 months, based on an offense level of 25 and a criminal-history category of I. By statute, though, the maximum term of imprisonment that can be imposed on either defendant is 60 months. 18 U.S.C. § 371. The Court now turns to the defendants' objections.
Sandison and Martinson raise four legal objections to the PSRs' calculation of the sentences recommended by the Sentencing Guidelines. First, both Sandison and Martinson object to the 18-level enhancement in ¶ 50
Paragraph 50 of the PSR adds 18 levels to Sandison's and Martinson's offense levels under § 2B1.1(b)(1)(J) of the Sentencing Guidelines. Under that provision, an 18-level enhancement is applied if the loss exceeds $2,500,000 but does not exceed $7,000,000. The amount of loss is the intended loss or the actual loss, whichever is greater. USSG § 2B1.1 cmt. n. 3(A). The government conceded at the sentencing hearing that the defendants did not intend to cause any loss to any of the participating banks. Evid. Hrg. Tr. at 20. That concession was well advised. Clearly, Sandison and Martinson acted in an attempt to help the RTC project succeed and eventually repay all of its loans, including the $35 million participation loan.
Because Sandison and Martinson did not intend to cause any loss to the participating banks, the applicability of the 18-level enhancement turns on whether the participating banks suffered an actual loss as a result of Sandison and Martinson's crime. "Actual loss" is "the reasonably foreseeable pecuniary harm that resulted from the offense." USSG § 2B1.1 cmt. n. 3(A)(i). The Sentencing Guidelines provide that a court need not calculate the actual loss with precision, but merely "make a reasonable estimate." USSG § 2B1.1 cmt. n. 3(C).
The government contends that the defendants committed bank fraud in three ways: First, the government alleges that the defendants committed bank fraud when, at the time that they extended the $990,000 line of credit, they hid the mortgage securing that line of credit from the participating banks. Second, the government alleges that the defendants committed bank fraud when, at the time of the closing on the $6 million loan, they continued to hide the existence of the mortgage securing the $990,000 line of credit by implying that the mortgage securing the $6 million loan was the third mortgage on RTC property — when, in fact, it was the fourth. Third, the government alleges that the defendants committed bank fraud when, faced with the possibility that the RTC project would fail, they jumped ahead of the participating banks to ensure that their personal loans to the developer — the $617,000 drawn against the line of credit and the $6 million loan — would be repaid before the $35 million participation loan.
The government has not even tried to prove that the first two instances of fraud caused a loss to the participating banks, and for good reason. No one can know whether any participating bank lost money because the defendants hid — and then continued to hide — the existence of the mortgage securing the $990,000 line of credit. The government suggests that, if the participating banks had known about the existence of that mortgage, they might have acted differently with respect to the $35 million loan, such as by replacing CNB as the lead bank. But this is sheer speculation. It is just as likely that the participating banks would have raised no objection whatsoever. After all, after Minnwest replaced CNB as the lead bank, the other participating banks readily agreed to a bridge loan that was to be repaid before the $35 million participation loan — and readily agreed to a new mortgage securing that bridge loan. Stip. at ¶ 15.
Because the government cannot prove that the first two instances of fraud caused any loss, the government's contention that the defendants caused the participating banks to lose $6,617,000 rests solely on the third instance of fraud — that is, on the steps taken by the defendants to ensure that their personal loans would be repaid ahead of the $35 million participation loan.
The Court disagrees. The defendants have pleaded guilty to conspiring to commit bank fraud. Bank fraud occurs when a defendant executes a "scheme . . . to defraud a financial institution." 18 U.S.C. § 1344. To assess the amount of loss caused by such a scheme, the Court must consider the impact of the scheme as a whole. The amount of loss caused by a scheme can be calculated only by comparing the victim's financial position after the scheme was executed with what the victim's financial position would have been had the scheme never been executed.
The problem with the government's calculation of the amount of loss caused by Sandison and Martinson is that the government divides their scheme into parts, ignores the part of the scheme that benefitted the participating banks, and considers only the part of the scheme that harmed the participating banks. Put differently, the government calculates the amount of loss by comparing the participating banks' financial position after the scheme was executed with what the participating banks' financial position would have been if only part of the scheme had been executed. The unfairness of the government's approach can be illustrated by an analogy.
Suppose that A and B hatch a scheme to defraud C. Under the scheme, A and B will sell authentic Rolex watches to C at steep discounts — and then, after they have gained C's trust, A and B will take advantage of that trust by selling fake Rolex watches to C. Pursuant to the scheme, A and B first sell $25,000 worth of watches to C for $5,000. C is impressed. A and B then sell $100,000 worth of watches to C for $20,000. C is again impressed. Taking advantage of C's trust, A and B then sell C a crate of worthless fake Rolex watches for $500,000 and abscond with the funds.
Using the approach that it has taken in this case, the government would calculate the amount of loss caused by A and B as $500,000. Obviously, though, that would be unfair to A and B. Their scheme had two parts: (1) gain the trust of C by selling him authentic Rolex watches at ridiculously low prices, and then (2) take advantage of that trust by selling C a large number of fake Rolex watches. In others words, the scheme involved both putting money into C's pockets and taking money out of C's pockets. Had the scheme never been executed, C would not have paid $500,000 for the fake Rolex watches, but C also would not have received $125,000 worth of authentic Rolex watches for $25,000. Thus, the scheme resulted in C paying A and B a total of $525,000, and A and B giving C merchandise worth $125,000. The amount that C lost because of the scheme is $400,000 — not (as the government would have it) $500,000.
Similarly, the scheme in this case had two parts: (1) Sandison and Martinson loaned the RTC developer $6,617,000; and (2) Sandison and Martinson arranged for the RTC developer to repay the $6,617,000 loan before repaying the $35 million participation loan. The two parts of the scheme cannot be separated, for it is clear that Sandison and Martinson would not have gone forward with the first part of the scheme (loaning $6,617,000 to the RTC developer) without having in place the second part of the scheme (the RTC developer's promise to repay them before the participating banks). All evidence in the record suggests
Obviously, then, had the defendants' scheme never been executed, Sandison and Martinson would not have taken $6,617,000 from the RTC developer, and that money would have been spent for the benefit of the participating banks. But had the defendants' scheme never been executed, Sandison and Martinson also would not have paid $6,617,000 to the RTC developer — money that (as explained below) was spent for the benefit of the participating banks. It cannot be emphasized enough that Sandison and Martinson did not first loan $6,617,000 to the RTC developer and then, some months or years later, decide to defraud the participating banks by inducing the developer to repay their loan ahead of the participation loan. If that had happened, this would be a different case; the defendants' scheme would have related solely to getting preexisting loans wrongfully repaid, and the amount of loss caused by the scheme would have been $6,617,000. Rather, Sandison and Martinson provided $6,617,000 to the RTC developer only because the RTC developer first agreed to repay that loan ahead of the participation loan. In short, Sandison and Martinson's scheme — like the scheme of A and B in the hypothetical described above — involved both giving money to the "victim" and taking money from the "victim."
To be sure, Sandison and Martinson provided $6,617,000 to the RTC developer, rather than to the victim banks — just as Sandison and Martinson later took payments from the RTC developer, rather than from the victim banks. But just as each dollar taken from the RTC developer came out of the pockets of the participating banks, each dollar given to the RTC developer went into the pockets of the participating banks. The RTC developer used the $617,000 drawn against the line of credit to make an interest payment to the participating banks, Stip. ¶ 8; Evid. Hrg. Tr. at 13-14, and the RTC developer used the $6 million borrowed from Sandison and Martinson to pay subcontractors, Stip. ¶ 9. If the RTC developer had not paid those subcontractors, the subcontractors would have placed $6 million in liens on RTC property. Thus, every single dollar that Sandison and Martinson loaned to the RTC ended up benefitting the participating banks — either directly (in the form of interest payments) or indirectly (in the form of reducing the amount of the liens on the collateral that the participating banks eventually sold after foreclosing on the participation loan).
The defendants' scheme is therefore unlike the offense in United States v. Mancuso, 42 F.3d 836 (4th Cir. 1994) and unlike the breach of fiduciary duty in Helm Financial Corporation v. MNVA Railroad, Inc., 212 F.3d 1076 (8th Cir. 2000) — two cases that the government cites in support of its amount-of-loss theory. The defendants in those cases only took money from their victims; they never provided money to their victims as part and parcel of their fraudulent schemes. In this case, by contrast, the fraudulent scheme of Sandison and Martinson involved both (1) putting $6,617,000 into the pockets of the participating banks, and (2) taking $6,617,000 out of the pockets of the participating banks. Thus, the amount of loss suffered by the participating banks as a result of Sandison and Martinson's scheme — the entire scheme, not just the part of the scheme on which the government focuses — was zero.
Paragraph 51 of the defendants' PSRs applies a two-level enhancement pursuant to § 2B1.1(b)(2)(A)(i) of the Sentencing Guidelines. Under that provision, a two-level enhancement is appropriate when the offense had 10 or more, but fewer than 50, victims. The PSR applied the enhancement based on a finding that there were at least 18 victim banks in this case.
The parties agree that, under the Eighth Circuit's decision in United States v. Miller, if the amount of loss is zero for purposes of § 2B1.1(b)(1), then "[i]t necessarily follows that there were no `victims' within the meaning of § 2B1.1(b)(2)(A)(i)." 588 F.3d 560, 567-68 (8th Cir. 2009); see also Govt Ltr. at 1-2 [Docket No. 100]; Martinson Ltr. at 1-2 [Docket No. 99]. The government further agrees that, although Sandison stipulated to the enhancement in his plea agreement, the Court nevertheless should not apply the enhancement to him if it finds that no loss occurred. Govt. Ltr. at 2.
Paragraph 54 of Sandison's PSR applies a two-level role enhancement to Sandison pursuant to § 3B1.1(c). Under § 3B1.1(c), a two-level enhancement is to be applied "[i]f the defendant was an organizer, leader, manager, or supervisor" of the offense. For purposes of this enhancement, the Court must construe the terms "manager" and "supervisor" broadly, and the enhancement may be applied even if the defendant directed the actions of only one other participant. United States v. Umanzor, 617 F.3d 1053, 1060 (8th Cir. 2010).
Sandison objects that he was not an organizer, leader, manager, or supervisor of the criminal activity, but merely a co-equal participant. The Court disagrees. Sandison (who was the president of CNB) and Martinson (who worked under Sandison as a vice president of CNB) have pleaded guilty to conspiring to commit bank fraud. In furtherance of this conspiracy, Martinson sent a materially misleading letter dated August 10, 2004, to the participating banks. Sandison Plea Agreement at 3; Change of Plea Tr. at 31. Sandison admits that he directed Martinson to send the misleading letter. Stip. at ¶ 10; Change of Plea Tr. at 31. Because Sandison admits that he directed Martinson (his subordinate) to commit an act in furtherance of the conspiracy, Sandison's offense level should be increased by two levels under § 3B1.1(c).
Paragraph 55 of the PSR applies a two-level abuse-of-trust enhancement as to both Sandison and Martinson. Under § 3B1.3 of the Sentencing Guidelines, a two-level enhancement is appropriate "[i]f the defendant abused a position of public or private trust. . . ." USSG § 3B1.3. A position of public or private trust is "characterized by professional or managerial discretion (i.e., substantial discretionary judgment that is ordinarily given considerable deference)." USSG § 3B1.3 cmt. n. 1. Individuals who hold positions of public or private trust "ordinarily are subject to significantly less supervision than employees whose responsibilities are primarily non-discretionary in nature." Id. Once the sentencing court determines that a person occupies a position of trust, the court must then determine whether the position of trust "`contributed in some significant way to facilitating the commission or concealment of the offense' before the enhancement may apply." United States v. Waldner, 580 F.3d 699, 706 (8th Cir. 2009) (quoting USSG § 3B1.3 cmt. n. 1). An abuse-of-trust enhancement is warranted, for example, in the case of a bank executive's fraudulent loan scheme, but is not warranted in the case of embezzlement by a bank teller. USSG § 3B1.3, cmt. n.1.
As an initial matter, the Court agrees with Sandison and Martinson that, at the time the participation loan closed, they did not hold positions of trust with respect to the participating banks (with the exception of CNB, where they were officers). The participation loan was an arms-length transaction, and the participating banks did not place any particular trust in the defendants. See, e.g., Saxena Aff. Ex. A [Docket No. 62-1] (in which the participating banks expressly agree that they were "not relying upon [CNB's] judgment, and that [CNB] . . . made no warranty of any kind, express or implied, in connection with the Loan or . . . regarding the legality, perfection, enforceability, or priority of any security interests, mortgages, guaranties, or similar documents issued in connection with the Loan.").
But once the participation loan closed and CNB began administering it, CNB (and its officers, including Sandison and Martinson) did hold a position of trust with respect to the other participating banks. The participation agreement granted CNB broad discretion to "administer the Loan and any related security and guaranties as though it were the sole owner and holder" of the loan and to "make all decisions concerning the servicing of the Loan." Govt. Ex. 2 at ¶ 12A ("Participation Agreement"). CNB was to provide the participating banks updates on the loan, including "any factual information bearing on the Borrower's continuing credit worthiness." Participation Agreement ¶ 10. The participation agreement further provided that CNB "will use the same degree of care in servicing and collecting the Loan as it would for its own accounts." Participation Agreement ¶ 12B. Under the participation agreement, CNB and the other participating banks "will promptly notify each other should either receive actual notice or knowledge of any . . . change in financial condition of any Obligor under the Loan, which will have a material adverse effect upon continuation of payments under the Loan or its repayment on default." Participation Agreement ¶ 18.
In short, the participation agreement contemplated that, once the participation loan closed, the participating banks would trust and rely upon CNB's officers — including Sandison and Martinson — to act honestly in servicing and collecting the loan and in keeping the participating banks apprised of material facts regarding the loan. Sandison and Martinson abused the trust placed in them by the participating banks when they failed to keep the participating banks apprised of the loan's status, did not tell the participating banks that the RTC was taking on additional encumbrances, and did not act honestly in servicing and collecting the loan. Because the defendants abused their position of trust with respect to the other participating banks, application of a two-level enhancement under § 3B1.3 is warranted.
Even if Sandison and Martinson did not hold a position of trust with respect to all of the participating banks, they unquestionably held a position of trust with respect to one of the participating banks: CNB. See United States v. Morris, 18 F.3d 562, 568 (8th Cir. 1994) (bank officer and director occupied a position of trust with respect to her employer bank). When the defendants defrauded the participating banks, they defrauded CNB, and when they defrauded CNB, they abused the trust that had been placed in them by CNB.
In sum, the Court finds that both Sandison and Martinson abused a position of private trust vis-a-vis the participating banks generally and vis-a-vis CNB specifically. The Court therefore overrules Sandison's and Martinsons' objection to application of the two-level enhancement under § 3B1.3.
The defendants object to the PSRs' assertion in ¶ 43 that they owe $6,617,000 in restitution. The Court sustains the objection. As the Court has explained at length, the defendants did not cause a loss to the participating banks, and thus they are not required to pay restitution to those banks. Accordingly, ¶ 43 of the PSRs should be amended to reflect that neither defendant owes restitution.
The Court ruled on a number of the defendants' factual objections at the November 30 hearing. The reasons for those rulings are reflected in the transcript of that hearing.
First, Sandison objects to the heading for ¶¶ 25-28 of his PSR — "Misapplication of Funds" — on the grounds that there was no loss and no misapplication of funds. The Court determines that a ruling on this objection is unnecessary because this heading "will not affect sentencing." Fed. R. Crim. P. 32(i)(3)(B). The Court is well aware of the facts of this case; it will not be influenced by the words used in the headings of the PSR.
Second, Sandison objects to ¶¶ 29-31 of the PSR. Sandison argues that "many" of the statements in these paragraphs "are simply incorrect" because he did not mislead the Office of the Comptroller of Currency ("OCC") about the $6 million loan, as the PSR reports. Sandison Pos. on Sent. at 10. The government disagrees with Sandison's objection. At the hearing, the parties agreed that the Court could resolve Sandison's objection in light of the parties' written factual stipulation and government exhibits 15, 16, and 31. After careful review of the relevant materials, the Court overrules Sandison's objection.
As an initial matter, although Sandison objects to ¶¶ 29, 30, and 31, his objection seems to apply to only ¶ 30 because only that paragraph suggests that Sandison misled the OCC with respect to the $6 million loan. Specifically, ¶ 30 states that in August 2004, Sandison made false and misleading statements to OCC examiners when he claimed that "`not a penny' would go to payment of the $6 million loan until all of the participating banks were paid in full first." PSR ¶ 30. According to Sandison, the evidence shows that his representations to OCC examiners regarding the $6 million loan were true — or, at worst, that the evidence is ambiguous.
Sandison is correct that some of the evidence is ambiguous. For example, the Asset Quality Review Sheet completed by an OCC examiner in August 2004 — essentially a compilation of the examiner's notes of his investigation — states that "B.[]Sandison stated no interest or [sic] has been collected to date and none will until the bank is paid out." Gov't Ex. 15 at 1. A word is plainly missing from the sentence after "interest or." Interest or what?
Sandison contends that the examiner most likely meant to write that "no interest or fees has been collected to date and none will until the bank is paid out." Such a statement from Sandison would have been consistent with the language that Martinson used in the August 10, 2004 letter to the participating banks — the letter that Sandison directed Martinson to send. Gov't Ex. 12 at 3 (stating that the lenders on the $6 million loan were "not being paid any interest or related fees at this time."). If the OCC examiner meant to write in the Asset Quality Review Sheet that Sandison represented to him that the RTC developer would not pay interest or fees on the $6 million loan, Sandison argues, then that must mean that Sandison was silent as to when the RTC developer would repay the principal. And if Sandison said only that no interest or fees would be paid ahead of the $35 million participation loan, then Sandison spoke the truth: In accordance with the agreement governing the $6 million loan, RTC was not obligated to pay interest or fees on the $6 million loan before paying off the $35 million loan. See Gov't Ex. 14a ¶¶ 9-10.
Sandison's argument is highly speculative. It seems as likely that the OCC examiner meant to write "no interest or principal has been collected to date and none will until the bank is paid out" as it is that he meant to write "no interest or fees has been collected to date and none will until the bank is paid out interest or fees." Indeed, the former may be more likely, given that it is grammatical; one would not normally say "no interest or fees has been collected." More importantly, though, Sandison's argument is contradicted by the testimony given by the OCC examiner before the grand jury. The OCC examiner testified that Sandison told him that the $35 million participation loan would be repaid in full before any payments of any kind would be made on the $6 million loan:
Gov't. Ex. 31 at 18-19.
This exchange supports the PSRs' factual assertion that Sandison made false and misleading statements to the OCC. According to the OCC examiner, Sandison told him that "not a penny would go to anyone else but [the participating banks]" until the participating banks were paid in full. That was a lie; Sandison knew that many millions of "pennies" would go to Sandison, Martinson, and the other lenders on the $6 million loan before the participation loan was repaid.
The OCC examiner also testified before the grand jury that Sandison told him the same thing that Martinson told the participating banks in the August 10, 2004 letter — namely, that:
Gov't Ex. 31 at 21-22. This later testimony is somewhat ambiguous, but it certainly does not contradict the OCC examiner's earlier testimony that Sandison told him that no payments would be made on the $6 million loan before the $35 million participation loan had been repaid.
Based on the evidence in the record, the Court finds that the government has shown, by the preponderance of the evidence, that Sandison falsely and misleadingly told the OCC that no payments of any kind would be made on the $6 million loan until the participation loan had been repaid in full. Therefore, the Court overrules Sandison's objection to ¶ 30 of his PSR.
Based on the foregoing, and on all of the files, records, and proceedings herein, IT IS HEREBY ORDERED THAT:
Moreover, even if Application Note 3(B) applied, the Court would reach the same conclusion, for the amount of gain to Sandison and Martinson was zero. Again, Sandison and Martinson paid $6,617,000 to the RTC developer and got $6,617,000 back from the RTC developer. They did not enjoy a gain. Indeed, because Sandison and Martinson borrowed the money that they loaned to the RTC from United Bankers Bank — and because the defendants had to repay United Bankers Bank with interest — Sandison and Martinson actually lost money as a result of their crime.