Appeal from the United States District Court for the Northern District of Illinois, Eastern Division. No. 07 CR 298-Milton I. Shadur, Judge.
The opinion of the court was delivered by: Posner, Circuit Judge
Before POSNER, MANION, and HAMILTON, Circuit Judges.
Edward Vrdolyak pleaded guilty to conspiracy to commit mail and wire fraud and agreed in the plea agreement that the loss intended by his fraud was between $1 million and $2.5 million. He was sentenced to five years of probation, with a community-service obligation but no confinement, and to pay a $50,000 fine (a modest amount, because the defendant has a high income, and a net worth in excess of $1 million if his large loans to members of his family are included). The government appeals, contending that the judge miscalculated the sentencing-guidelines range applicable to the defendant's crime and committed other errors. Although a judge is no longer required to give a guidelines sentence, he is required to make a correct determination of the guidelines sentencing range as the first step in deciding what sentence to impose. Gall v. United States, 552 U.S. 38, 50 (2007); United States v. Gibbs, 578 F.3d 694, 695 (7th Cir. 2009).
The Chicago Medical School (as it was then known) wanted to sell a property in Chicago that it owned consisting of a lot with a building on it. Stuart Levine was a trustee of the medical school and the chairman of the board's real estate committee, and he agreed with the defendant to use his position as a trustee to steer the sale of the property to a buyer of the defendant's choice. The defendant lined up Smithfield Properties to be the favored buyer in exchange for a $1.5 million fee that Smithfield agreed to pay him, and he in turn agreed to give Levine half the fee. The medical school was not told about this corrupt arrangement. Levine like the defendant has pleaded guilty to his part in the fraud and has agreed not to contest a prison sentence of up to 67 months that the sentencing judge might impose.
Smithfield's initial offer for the building-$9.5 million-was lower than two other potential buyers-the Farley Group and Loyola University-were willing to pay. The defendant advised Smithfield to up its offer, and it did, to $15 million. Farley and Loyola remained inter- ested in buying the property. To head them off, Levine arranged for an "emergency" meeting of the medical school's board of trustees to consider offers for the property. At the meeting, although Farley had offered $15 million and Loyola $15.5 million for the property, the board, persuaded by Levine, decided to accept Smithfield's offer and negotiate no further with Farley or Loyola. The board discounted Loyola's bid because Loyola had not actually inspected the property before bidding-Levine had seen to that. The board rejected Farley's bid because Levine strongly urged approval of Smithfield's bid, noting that Farley's was lower because it included a 4 percent brokerage fee that would be deducted from the amount paid to the school. This was misleading, because Smithfield's bid was contingent on obtaining zoning approvals and Farley's was not. And a week later Farley upped its bid to $16 million, which in pure dollar terms was higher than Smithfield's even after deduction of the brokerage fee. Farley was told that it was too late.
The government was prepared to offer an affidavit from Loyola's broker that Loyola would have increased its offer had it been given an opportunity to do so. And a representative from Farley was prepared to testify that if necessary Farley would have increased its offer to some-where between $18 and $20 million. By convening the emergency meeting Levine had made sure that Smithfield's bid would be accepted and that he and the defendant would split the finder's fee. Although we use "bid" and "bidder" as synonyms for "offer" and "offeror," no formal auction was ever contemplated and so there was no reason to consider Farley's higher bid untimely.
The district judge concluded that the defendant's fraud had inflicted neither actual nor intended loss on the medical school. His finding that it had inflicted no actual loss was based on the fact that Smithfield's bid was the highest one considered at the "emergency" meeting. The judge gave no weight to Farley's week-later offer of $16 million and refused to consider the evidence that Farley would have bid $18 million to $20 million if given the chance and that Loyola was also prepared to offer more than $15.5 million. These rulings were erroneous. No emergency required the medical school's board of trustees to act with haste to award the sale contract. The "emergency" was a ruse to preclude competition with Smithfield.
The judge's refusal to consider the evidence of what Loyola or Farley would have done if given the chance to sweeten their bids was based on his belief that uncommunicated intentions are unworthy of consideration by a finder of fact. That is not correct. No rule of evidence or principle of common sense makes a person's testimony about his own intentions-testimony uniquely based on his personal knowledge-inadmissible in a sentencing proceeding any more than in any other proceeding in which intention is material. United States v. Young, 247 F.3d 1247, 1252--53 (D.C. Cir. 2001). Who better than a potential buyer knows what he would bid for a property?
The judge himself speculated at the sentencing hearing about the defendant's uncommunicated intentions in conspiring with Levine to defraud the medical school- that he had acted out of friendship for Levine. A defendant's testimony about his uncommunicated intentions is no more credible than the testimony of an honest third party about his uncommunicated intentions. To believe the former and refuse even to listen to the latter is error.
The weight to be given a piece of evidence is one thing, and is ordinarily within the discretion of the trier of fact to determine. Admissibility is another matter. A judge is not permitted to have his own rules of admissibility-to say for example that "[i]n my court no exceptions to the hearsay rule will be recognized." As we shall be emphasizing throughout this opinion, our concern is not with the leniency of the defendant's sentence as such but with procedural errors committed by the judge en route to the determination of the sentence.
The judge's refusal to listen to the evidence of the potential buyers was an egregious error because the evidence was corroborated. The medical school's property had recently been appraised for $15 million on the assumption that its best use was as a luxury residential development, a use that would require tearing down the building on the property. If the building was not torn down (an expensive undertaking), the land alone, according to the appraisal, was worth $16.5 million. Loyola didn't want to tear the building down; it wanted to use it for student housing. It had every reason therefore to offer more than Smithfield. Farley had no intention of demolishing the building either, and its intention to top Smithfield's bid was corroborated by the $16 million offer that it made for the property.
The judge was impressed by the fact that the defendant had told Smithfield that $9.5 million was too low an offer. By doing so, the judge reasoned, he had conferred a benefit on the school. But that was not the defendant's intention. His intention was to make sure that Smithfield was the winning bidder, since the finder's fee was contingent on Smithfield's getting the property. Whether in an honest bidding process the school would have obtained more than $15 million from Farley or Loyola or perhaps from some other potential buyer can't be determined with certainty because Levine prevented Farley and Loyola from keeping the bidding going and prevented everyone else who had expressed interest from even making offers.
The judge thought the defendant's interests perfectly aligned with the school's-thought that the more Smithfield bid, the more the school would receive, as well as the defendant. That is not true. The defendant did want Smithfield to be the high bidder, but he also wanted the bidding process to be rigged, to make sure Smithfield was the high bidder so that he would get his fee. The result of the rigging was to prevent the medical school from considering higher bids from Farley and Loyola and perhaps others.
In determining pecuniary loss for purposes of calculating a sentencing-guidelines range, the judge is required to determine the loss that the defendant "reasonably should have known, was a potential result of the offense." U.S.S.G. § 2B1.1, Application Note 3(A)(iv). That potential loss in this case was the amount above $15 million that another bidder might have decided to pay for the property had the bidding been fair and open. As an experienced lawyer and businessman, the defendant must have known that a fair and open bidding process might well yield a higher price than Smithfield offered. In fact he knew that both Loyola and Farley wanted to pay more than Smithfield, which made sense because, as we said, both bidders wanted to use the building on the property rather than tear it down.
The judge's finding that the defendant had caused no loss blocked the alternative measure of loss in cases in which there is a loss but the precise amount of the loss cannot be determined: in such a case the criminal's gain is treated as the measure of loss. U.S.S.G. § 2B1.1, Application Note 3(B); United States v. Serpico, 320 F.3d 691, 698 (7th Cir. 2003); United States v. Bhutani, 266 F.3d 661, 668 (7th Cir. 2001); United States v. Chatterji, 46 F.3d 1336, 1340 (4th Cir. 1995). That makes good sense in this case. Smithfield was willing to pay $1.5 million to the defendant to obtain the property, and it must have thought that if it didn't pay that amount it would have to up its bid by at least that much to win an unrigged bidding contest. Only on that assumption did the kick-back make sense from Smithfield's standpoint. From the defendant's standpoint, the more Smithfield paid, the better; but from Smithfield's standpoint, the goal of paying a finder's fee was to enable Smithfield to obtain the property for a smaller total outlay (price plus finder's fee) than it would have had to pay otherwise.
There was at the very least a probable loss, and that is "loss" within the meaning of the guideline. United States v. Johnson, 16 F.3d 166, 170 (7th Cir. 1994); United States v. Schneider, 930 F.2d 555, 558 (7th Cir. 1991); United States v. Stanley, 12 F.3d 17, 21 (2d Cir. 1993). It is true that cases involving probable loss usually are ones in which the illegal scheme is interrupted, so that its consequences cannot be determined with certainty. Here it was not interrupted. But the consequences still cannot be determined with certainty, and it would be even more anomalous to give the defendant a sentencing break when there is no interruption by some outside force but instead the very nature of the scheme precludes a certain determination of loss.
The gain (and thus alternative measure of the loss) was the $1.5 million finder's fee. It is true that when originally negotiated, the fee was contingent on certain factors. But by the time of the defendant's sentencing, the contingencies had been dispelled and the defendant would have been entitled, had the scheme not been detected, to the full $1.5 million. That the fee was to be split with a coconspirator is of no significance. U.S.S.G. § 1B1.3(a); United States v. Thomas, 199 F.3d 950, 952-54 (7th Cir. 1999); United States v. Boatner, 99 F.3d 831, 834-37 (7th Cir. 1996). Dividing the gain by the number of conspirators would mean that the larger the conspiracy, the milder the punishment of each one. Anyway the defendant stood to gain $750,000 from his crime-not a negligible haul.
The zero loss found by the district judge created a guidelines sentencing range of zero to six months in prison; the correct loss figure of $1.5 million (which incidentally was within the range that the defendant agreed in the plea agreement was the intended loss attributable to his crime) ups the sentencing range to 33 to 41 months.
Ordinarily we would stop here and remand for resentencing. But the judge went on to rule that if he was wrong and there was a loss of $500,000, which would create a guidelines range of 27 to 33 months in prison, he would give the defendant a below-guidelines sentence of no prison-in fact the identical sentence that he imposed on the assumption of zero loss.
But $500,000 was also error. And while a judge can give a below-guidelines sentence, the sentence cannot stand if it is based on a legal, factual, or analytic (connecting law and fact) error that is not harmless. The court of appeals must "ensure that the district court committed no significant procedural error, such as failing to calculate (or improperly calculating) the Guidelines range, treating the Guidelines as mandatory, failing to consider the § 3553(a) factors, selecting a sentence based on clearly erroneous facts, or failing to adequately explain the chosen sentence-including an explanation for any deviation from the Guidelines range." Gall v. United States, supra, 552 U.S. at 51. "The allowable band of variance [in sentencing] is greater after Booker than before, but intellectual discipline remains vital." United States v. Kirkpatrick, 589 F.3d 414, 416 (7th Cir. 2009); see also United States v. Peña-Hermosillo, 522 F.3d 1108, 1112 (10th Cir. 2008).
The judge committed three errors in his alternative ruling. First, the $500,000 figure was erroneous for the reasons we've given already. The correct figure was $1.5 million and the guidelines range was therefore higher than the judge thought. Second, repeating an error in his computation of loss, the judge thought that the defendant deserved leniency because he had intended no harm to the medical school, but on the contrary had intended a benefit-that the school should receive the highest bid from Smithfield. Notice the equivocation implicit in "highest bid from Smithfield." The highest bid from Smithfield is the bid that ...