Convicted Politician Did Not Lay a Proper Foundation For Proferred Indirect Testimony of Lack of Intent (7/4/14).

In United States v. Beavers, ___ F.3d ___, 2014 U.S. App. LEXIS 12469 (7th Cir. 2014), here, the defendant was convicted of multiple counts of tax fraud.  The facts are within a range of what might be expected given his profession -- money flowed to him and he did not report or pay tax on it.

The interesting part of the case relates to the key element for tax fraud that the defendant act willfully -- with intent to violate a known legal duty.  How does a defendant unwilling to testify as to his intent -- thus invoking his Fifth Amendment privilege -- introduce indirect evidence of his lack of intent to blunt the Government's indirect proof of his intent?  (Similar issues going to the willfulness element are, for example, reliance on tax practitioner (how can the defendant put his reliance in play without testifying).  Can the defendant effectively mount the "defense" without taking the stand? This issue often comes up in the context of a defense request for a jury instruction on good faith -- an instruction meant to emphasize something inherent in the willfulness element, that if the defendant acted in good faith he did not act willfully.  I offer the following from my Federal Tax Crimes Book (footnotes omitted):

The so-called good faith defense – sometimes the Cheek good faith defense – is technically not a defense.  The Government must prove willfulness.  Willfulness does not exist if the defendant acted in good faith with a belief that the law did not impose the legal obligation he is alleged to have violated.  Does that mean that the Government, in order to prove willfulness, must disprove good faith as to the legal requirement?  Logically, it would and, certainly most of the time, the Government’s proof of willfulness will in fact be sufficient to permit the jury to infer beyond a reasonable doubt that the defendant lacked good faith. 

Defendants who want to emphasize their good faith with the hope that the evidence bearing on good faith will at least dissuade the jury from finding willfulness beyond a reasonable doubt.  Thus, they will want the judge, by separate instruction, to instruct that good faith negates willfulness.  That good faith nuance, while implicit in the general willfulness instruction, is not as explicit as defendants desire.  Generally, courts will give the specific good faith instruction only if the evidence somehow affirmatively puts good faith in play – makes it a real issue for the jury.  How does the defendant do that?  The most direct way is for the defendant to testify as to his or her good faith.  But, in order to do that, the defendant will be subject to cross-examination; frequently, the defense team will conclude that the potential benefits of the defendant testifying (including the good faith opportunity) are less than the risks of the defendant testifying.  So the defendant will not testify.  Notwithstanding some noises that the defendant is required to testify to put good faith in play, the courts soundly reject that notion.  Other circumstantial evidence, including perhaps lay opinion evidence as to the defendant’s mental state, may be sufficient to put that issue in play and, if it does, the trial judge should give the instruction.

So, basically, the Government must prove intent beyond a reasonable doubt via circumstantial evidence in the absence of the defendant's testimony.  In appropriate cases, the defendant should be able to do so also, provided that he gets to the jury by laying the proper foundation.

With that, let's turn to Beavers.

Evidence of Beaver's Remedial Actions - Filing Amended Returns

After the criminal investigation started, Beavers filed amended returns.  He thus took the risk that the amended returns would prove one element of the crime -- tax due -- while setting up his good faith defense on another element -- willfulness.  That is a dicey move.  Still that is what he did, hoping, I am sure, to disincentivize the IRS from continuing to investigate and then DOJ from prosecuting.  He also paid a loan he might not have otherwise paid and wanted to use the payment as proof that the original transaction was a loan.  He wanted to introduce proof of those remedial actions to negate willfulness.  The district court did not allow the evidence.  The reason for disallowing it and the Court of Appeals' resolution of the appeal on the issue are:

The district court ruled that the evidence would be admissible if Beavers could establish that each remedial action was relevant to his state of mind at the time he filed the original tax returns. Beavers ultimately elected not to testify, and he did not otherwise succeed in establishing the required evidentiary foundation. Thus, the evidence of his remedial actions was not presented at trial. Beavers argues that (1) the evidence of his remedial actions was relevant under Rule 401; (2) the district court's conditional admission impermissibly burdened Beavers' Fifth Amendment right against self-incrimination; and (3) the court's rulings deprived him of his constitutional right to present a meaningful defense. 

First, the district court conditioned the admission of evidence of Beavers' corrective actions upon a showing that these actions had a connection to Beavers' state of mind at the time he filed his incorrect returns. The logic of the ruling was that, in the absence of a foundation establishing this link, the amended tax returns (and evidence of other remedial actions) did not make it more likely that Beavers believed his original returns were accurate when he filed them. 

District judges making relevancy determinations in this type of situation should proceed on a case-by-case basis. Cf. United States v. Tishberg, 854 F.2d 1070, 1073 (7th Cir. 1988) (reasoning, in the sufficiency context, that the defendant's amended return "may demonstrate a good faith effort to correct his previous mistakes," but also noting that the return "does not negate the import of his previous action"). Nevertheless, courts have repeatedly affirmed the exclusion of evidence of remedial action taken after the taxpayer knows he is under investigation. See United States v. McClain, 934 F.2d 822, 834-35 (7th Cir. 1991) (finding "no reason" to disturb the district court's ruling that defendant's 1985 tax return was not probative of his state of mind at the time he filed his 1984 return, given his indictment in the intervening year); United States v. Radtke, 415 F.3d 826, 840-41 (8th Cir. 2005); United States v. Ross, 626 F.2d 77, 81 (9th Cir. 1980); Post v. United States, 407 F.2d 319, 325 (D.C. Cir. 1968); United States v. Stoehr, 196 F.2d 276, 282 (3d Cir. 1952); see also United States v. Philpot, 733 F.3d 734, 748 (7th Cir. 2013) (explaining, in political corruption case, that defendant's remedial "actions after his bonuses were reported in the press shed little or no light on his state of mind two years earlier" when he received those bonuses). A common thread in many of these cases is that subsequent remedial actions may not be probative of the defendant's prior state of mind because such actions are equally consistent with (1) promptly correcting a genuine mistake and (2) trying to cover up a purposeful lie in the hope of avoiding prosecution. Cf. Fed. R. Evid. 407 (barring evidence of subsequent remedial action to prove an admission of fault). 

In any event, the district court did not exclude the evidence entirely—it simply conditioned the evidence's admission on some kind of showing of its relevance to Beavers' state of mind at the time he filed his original returns several years earlier. Beavers did not make this showing. In fact, his theory of defense is at odds with his argument that the remedial evidence was relevant. Beavers' argument in defense (which he maintains on appeal) was that all of the transfers from his campaign committees were loans, not income. Thus, evidence that Beavers declared some of the campaign fund transfers as income on his amended tax returns after he was under investigation has little bearing on whether he considered the transfers to be loans at the time he took the funds. In sum, the district court's sensible approach to the remedial evidence was within its discretion. 

Beavers next argues that the district court's handling of this issue impermissibly burdened his Fifth Amendment right against self-incrimination, see generally Griffin v. California, 380 U.S. 609 (1965), because it forced him either to testify (in order to lay the appropriate foundation) or to forgo the opportunity to get evidence before the jury. Beavers' argument misunderstands the nature of this right, however. Criminal defendants often face difficult choices in weighing the costs and benefits of testifying. And the rules of evidence sometimes prevent defendants from getting their story—or evidence of their state of mind—before the jury in the particular manner they would prefer. Cf. U.S. ex rel. Harris v. Illinois, 457 F.2d 191, 197-98 (7th Cir. 1972) ("The State presently contends, and we agree, that a defendant in a criminal trial may not introduce evidence in his defense without laying a proper foundation therefor."). For instance, the general rule against hearsay, see Fed. R. Evid. 802, would typically prevent a defendant from calling a friend to the stand to relay an exculpatory statement the defendant said to her, in lieu of the defendant testifying himself. Thus, the rule against hearsay may burden the defendant's right to testify in the same way Beavers argues that his right was burdened here, since it puts him to the choice of taking the stand (and exposing himself to potentially damaging cross-examination) or losing the chance to get evidence before the jury. That sort of burden is not impermissible, however; it is simply part of a larger system in which "[t]he accused does not have an unfettered right to offer testimony that is ... inadmissible under standard rules of evidence." Taylor v. Illinois, 484 U.S. 400, 410 (1988). We see no meaningful difference between the hearsay situation and the burden Beavers faced here—both burdens are ordinary, well-established, and permissible. Indeed, we have previously approved the exclusion of purported state-of-mind evidence offered by a defendant where the defendant could have, but did not, supply the requisite foundation of relevance through his own testimony. See United States v. Scott, 660 F.2d 1145, 1165-67 (7th Cir. 1981). Accordingly, we reject Beavers' argument on this score.

Limitations  on Expert So As Not to Produce Indirect Evidence of Lack of Intent

A related issue occurred when the district court limited the testimony of the defendant's expert witness.  Here is the discussion:

We begin with state-of-mind limitations. First, the district court ruled that Gershinzon could not base his expert opinions on what Beavers had told him about his (Beavers') state of mind at the time of the charged offenses. This limitation was proper. Otherwise, Beavers could have gotten highly selective and favorable statements of his before the jury without having to face cross-examination. Later, the district court also ruled that Gershinzon could not testify as to whether he (Gershinzon) believed the 100 checks were loans. The court reasoned that such testimony would violate Rule 704(b) of the Federal Rules of Evidence, which prohibits expert witnesses in a criminal case from opining as to whether the defendant had the mental state necessary for the charged crime. As the judge explained, "[t]he relevant state of mind is whether defendant intended to treat withdrawals from his campaign coffers as income or as loans. The expert cannot state an opinion on this issue." For Gershinzon to have addressed this issue would have been the equivalent of opining on whether Beavers had the "willfulness" necessary for a tax offense. See United States v. Windfelder, 790 F.2d 576, 582 (7th Cir. 1986). We accordingly find no abuse of discretion in this limitation. 

Next, the relevance limitations prohibited Gershinzon from testifying about issues that were not relevant absent testimony or other evidence connecting them to Beavers' state of mind at the time of the charged offenses. For example, the court barred Gershinzon from offering his opinion that Cook County erred by omitting the $1,200 monthly stipend payments to Beavers. As explained above, this testimony was irrelevant absent evidence that Beavers knew of and relied on the county's alleged obligation. Similarly, the defense sought to ask Gershinzon about a form that Beavers received from the pension fund in early 2008, as part of the defense's attempt to show that Beavers' $68,763 pension payment was not income. The court prohibited Gershinzon from testifying about this form because Beavers received it in 2008, so it was irrelevant to Beavers' state of mind in 2007 when he filed his 2006 tax return. The court's relevance limitations were therefore appropriate. (And where Gershinzon's testimony was relevant, the court permitted it. For instance, Gershinzon testified about accounting principles; about the purpose of certain tax forms; about how tax professionals disagree on the proper tax treatment of certain transactions; about the differences between income, loans, and advances; and about which records he would consider in determining whether a transfer was income, a loan, or an advance.) 

We turn now to the court's reliability limitations. Recall, the court's determination as to Gershinzon's reliability ultimately resulted in the court's finding Gershinzon unreliable and instructing the jury to partly disregard Gershinzon's testimony—including, importantly, his opinions as to whether the $68,763 pension payment and the 100 checks were loans or income. 

The court found Gershinzon unreliable for multiple reasons. Crucially, the court found that Gershinzon "[wa]s not careful with his assumptions," and those "assumptions have overtaken at least the minimum neutrality an expert is supposed to have." As noted above, one assumption was that, because many of Gershinzon's clients do not understand their W-2s, it apparently followed that a high-ranking public official (with access to advisors) was in the same position. The court also criticized Gershinzon's assumptions because Gershinzon relied on his conversations with Beavers in forming assumptions and ultimately conclusions about the proper tax treatment of the transactions at issue in this case. For instance, in concluding that the $68,763 pension payment was a loan, Gershinzon relied in substantial part on Beavers' statement to Gershinzon that Beavers considered it a loan. Similarly, in opining that the 100 campaign checks to Beavers were loans and advances, Gershinzon also relied on Beavers' statements to him. Even though the court had instructed Gershinzon not to rely on Beavers' statements, Gershinzon explicitly referenced those statements, indicating to the court that Gershinzon could not get Beavers' statements "out of his [own] mind... . He can't follow the rules." Gershinzon was also unable to point to portions of the tax code to support his assertions about the tax treatment of various payments, including the pension-fund payment. Thus, stripped of Beavers' statements, Gershinzon's most important opinions lacked meaningful support. 

Gershinzon also was unable to answer basic questions about the topic of his testimony. Specifically, during the second  voir dire, the government asked Gershinzon about the basis for his opinion that the City of Chicago should have deducted Beavers' lump-sum pension payment from his gross income for 2006. Gershinzon acknowledged that Beavers' pension payment was not mandatory, but was unable to say whether it was "voluntary," because he could not testify to Beavers' state of mind. The prosecutor clarified that this was not a state-of-mind question; rather, a letter gave Beavers three options, and Beavers selected the option that maximized his pension benefits. Nonetheless, Gershinzon would not say that Beavers' choice was "voluntary." After considerable back-and-forth between the prosecutor and Gershinzon, the court eventually said, "[a] step which is not compelled is a step which is voluntary." 

We "give the district court wide latitude in performing its gate-keeping function and determining both how to measure the reliability of expert testimony and whether the testimony itself is reliable." Bielskis v. Louisville Ladder, Inc., 663 F.3d 887, 894 (7th Cir. 2011). In light of the foregoing analysis—which indicates that Beavers' expert relied on irrelevant considerations, made questionable assumptions, and may have lacked expertise about certain germane subjects (including relevant portions of the tax code)—we find that the district court properly exercised its discretion. Moreover, the court still allowed Gershinzon to testify as to the objective characteristics he would look for in determining whether a particular transfer was income, a loan, or an advance. As a result, Gershinzon's analytical methodology was presented to the jury, even if his particular conclusions were not.