UNITED STATES of America, Plaintiff-Appellee, v. Thomas FLASCHBERGER, Defendant-Appellant, 408 F.3d 941

EASTERBROOK, Circuit Judge. Two small vocational sehools-the Lac Courte Orielles Ojibwa Community' College and the College of Menominee Nation — formed the Wisconsin Consortium of Indian Controlled Community Colleges to apply for federal grants, including funds under the Carl D. Perkins Vocational Education Act, 20 U.S.C. §§ 2301-2415. Thomas Flaschberger prepared the Consortium’s annual applications and certified at the end of each fiscal year its compliance with conditions placed on the grants. From 1994 through 2001 the Consortium received a little more than $900,000.

An audit that year revealed, however, that the applications and certifications had been false: the Consortium overstated the number of eligible students by about 40% and failed to provide them with any of the services for which the grants were supposed to pay. Instead all but about $4,000 of the funds had been treated as general tribal revenues. An indictment charged Flaschberger with mail fraud, see 18 U.S.C. § 1341, because the applications, certifications, and checks had been sent by mail, and with diverting some of the money to himself, in violation of 18 U.S.C. § 666.

The jury acquitted him of the latter charge but convicted him of mail fraud, and the judge sentenced him to 30 months’ imprisonment plus restitution of the whole $900,000. Flaschberger’s principal argument on appeal is that, because he relied on the colleges’ financial aid directors to calculate the number of eligible students, the evidence fails to demonstrate beyond a reasonable doubt that he intended to defraud. There are two problems with this line of argument.

First, Flaschberger did not move for an acquittal at the close of the evidence or after the trial and therefore can prevail now only by demonstrating plain error. See Fed.R.Crim.P. 29, 33; United States v. Owens, 301 F.3d 521, 527-28 (7th Cir.2002). The omission appears to have been part of his strategy rather than an oversight; Flaschberger argued to the judge that he had gone to trial only because of the § 666 charge and legal questions, as opposed, to a claim of factual innocence on the mail-fraud charge, and that he therefore should receive a lower sentence to reward acceptance of responsibility.

He made the current claim of factual innocence only after the judge concluded that he had not genuinely accepted responsibility for his deeds. Second, Flaschberger disregards the principal evidence against him. What he assured the grant-making authority is not simply that a certain number of students were eligible, but that the funds would be applied to authorized uses. The jury was entitled to find that Flaschberger knew that these representations were false.

Every fiscal year Flaschberger made at least four certifications — two applications and two year-end representations that the funds had been applied properly. (The Perkins grants funded two categories of services, program involvement and student support. Each had its own documentation.) Flaschberger repeatedly told the grant-making officials that the money would be used to underwrite particular services, which the applications described; at year end Flaschberger assured the officials that the money-had been applied to these services.

Yet ample evidence shows that neither of the colleges ever offered any of these services. Flaschberger does not contend that he relied on someone else for information about what services the colleges provided and how the funds would be used. He was the program director, both colleges are small, his job included accounting for the outlays, and he either knew that the services were not being rendered or had his eyes so tightly shut that the “ostrich” inference supports a finding of intent to deceive. See United States v. Ramsey, 785 F.2d 184 (7th Cir.1986); United States v.

Craig, 178 F.3d 891, 897 (7th Cir.1999) (applying the ostrich inference to another federal-grant fraud prosecution). Indeed, as we have said, Flaschberger does not even argue that the evidence with respect to the funds’ misapplication is insufficient. Flaschberger also contends that the acquittal on the § 666 charge demonstrates innocence of mail fraud, but there is no inconsistency; and if the verdicts conflicted that still would not entitle Flaschberger to relief, because an inconsistent acquittal on one count may demonstrate mercy or confusion rather than innocence. See United States v.

Powell, 469 U.S. 57, 105 S.Ct. 471, 83 L.Ed.2d 461 (1984). There is no plain error, and we proceed to the sentence. United States v. Booker, — U.S.-, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005), was released while this appeal was pending, and Flaschberger seeks its benefit by contending that the district judge committed plain error in making findings of fact (on a preponderance standard) while the Sentencing Guidelines were mandatory. But whether the sentence is proper under the governing statutes and guidelines is an antecedent question.

Flaschberger’s restitution must be recalculated even though, because there is no statutory maximum for restitution, the sixth amendment and Booker do not apply to that subject. See United States v. George, 403 F.3d 470 (7th Cir.2005); United States v. Behrman, 235 F.3d 1049, 1054 (7th Cir.2000). The district court ordered him to repay the whole sum that the Consortium received between 1994 and 2001. Yet the only crime of which he stands convicted is a scheme that, according to the indictment, spanned just three fiscal years: 1998-99, 1999-2000, and 2000-01.

Unless a defendant agrees to pay more, which Flaschberger did not, restitution is limited to the crime of conviction. See 18 U.S.C. § 3663A(a); Hughey v. United States, 495 U.S. 411, 110 S.Ct. 1979, 109 L.Ed.2d 408 (1990); United States v. Peterson, 268 F.3d 533 (7th Cir. 2001). Losses from the years preceding the scheme alleged in the indictment therefore must be subtracted from the award.

Flaschberger may be entitled in contribution or indemnity from the colleges and tribes some of all of what he must pay in restitution, for they wrongfully pocketed the money, but the fact that Flaschberger’s fraud did not feather his own nest does not relieve him of responsibility: restitution under § 3663A is based on the harm conduct causes to the victim and not on the wrongdoer’s personal gains. See United States v. Shepard, 269 F.3d 884 (7th Cir.2001). In this respect restitution under § 3663A is more closely related to civil damages than to the remedy in equity, which is gain-based.

As for the term of imprisonment: the-district court’s conclusion that the total loss was $900,000 added four offense levels,- compared with Flaschberger’s view that the loss attributable to his conduct was about $190,000. (Under the table in U.S.S.G. § 2B1.1, a loss between $120,000 and $200,000 yields 10 offense levels, while a loss between $400,000 and $1 million produces 14 offense levels.) Flaschber-ger’s argument for the lower number supposes that his sole misdeed was overstating the number of eligible students; the jury (and judge) were entitled to find him culpable for the whole amount because almost all of it was diverted from authorized uses.

Flaschberger says that overhead and indirect expenses should be deducted, but phantom programs do not-have ■allowable overhead and indirect charges. Once again, however, the $900,000 figure represents grants dating back to 1994. These may be included as relevant conduct under U.S.S.G. § IB 1.3 only if a single scheme or plan comprises all seven fiscal years. The district judge stated, at page 12 of the sentencing transcript, that the “guideline calculations take into account all acts and omissions that were part of the same course of conduct or common scheme or plan”; this is a formally correct finding.

But the judge may have been addressing only Flaschberger’s argument that the losses should be reduced because the colleges had many students eligible for "services. The judge never said why he thought the losses during the four fiscal years preceding the scheme charged in the indictment were part of that scheme. Flaschberger bears some of the responsibility for this silence. His lawyer failed to argue forcefully that there is a temporal issue (number of years) as well as a quantity issue (number of eligible students and extent of services they received).

But this inattention is at worse a forfeiture, not a waiver. In deciding how to respond to the problem, we must take account of Booker, which shows that a constitutional error occurred when the judge made-findings, on a preponderance of the evidence, while implementing a system that he viewed as mandatory. After Booker Flaschberger is entitled, at a minimum, to a limited remand so that the district judge may determine whether the extra discretion that exists in Booker’s wake would affect the sentence. See United States v. Paladino, 401 F.3d 471 (7th Cir.2005).

Because the term of imprisonment may well rest on an incorrect assumption about the treatment of losses from 1994 through 1998 — and because the restitution order certainly is erroneous, for this very reason — we think it best to vacate the entire sentence and remand for resentencing. This is not an application of the plain-error doctrine under Booker and Paladino, but an insistence that all Guideline calculations be done correctly before any Boo&er-based adjustments. That can be accomplished only if we vacate the sentence.

The limited Paladino remand leaves the sentence in place and asks for the district judge’s views. Because Flaschberger is entitled to some reduction (if only in restitution), it is best to give the district judge a full measure of leeway in adjusting the sentencing package, for the financial and imprisonment aspects of the sentence may interact. On remand the judge will be free to exercise the discretion that exists now that Booker has severed 18 U.S.C. § 3553(b)(1) from the Sentencing Reform Act.

The conviction is affirmed, but the sentence is vacated, and the-case is remanded for further proceedings consistent with this opinion and with Booker.