First the background of the case. The defendant was a return preparer who, allegedly, falsified tax returns for his clients. As the case went to trial, the charges were 20 counts for Aiding and Assisting, Section 7602(2) for claiming false deductions for the clients. The defendant was convicted of 16 counts and sentenced to 41 months incarceration and 36 months of supervised release.
Rule 404(b) - Bad Acts Evidence
The Court first addressed Rule 404(b) bad or other acts evidence. (Other acts may be a bit of a euphemism, since the whole point is for the prosecutor to bring bad stuff about the defendant to the jury's attention to influence their decision to convict.) The evidence was an IRS agent's testimony that, in a sting operation, upon the IRS agent's statement of concern about the indicated tax payment due, the defendant had offered to use "creative financing" -- really false deductions -- that reduced the indicated tax liability and generated an indicated refund. So the agent testified, the defendant generated this discussion and charged a higher fee for the indicated results.
The trial court admitted the evidence in under FRE Rule 404(b), although there were no charges related to that conduct. The defendant objected on appeal but the Second Circuit affirmed. The analysis is straight-forward and I quote it below (shorn of some case citations and with some quotation marks omitted for better readability). I do call to readers attention that, in a footnote, the Court deals cryptically with a significant difference between the agent's contrived circumstance and the clients' on the charged counts. Here is the discussion (most case citations omitted and case quotation marks omitted for readability):
The Federal Rules of Evidence prohibit the admission of "[e]vidence of a crime, wrong, or other act" — jointly, "other act" evidence — "to prove a person's character in order to show that on a particular occasion the person acted in accordance with the character," but the Rules permit such evidence to be admitted for other purposes, "such as proving motive, opportunity, intent, preparation, plan, knowledge, identity, absence of mistake, or lack of accident." Fed. R. Evid. 404(b). To determine whether the District Court properly admitted "other act" evidence under Rule 404(b), we consider whether: (1) the prior act evidence was offered for a proper purpose; (2) the evidence was relevant to a disputed issue; (3) the probative value of the prior act evidence substantially outweighed the danger of its unfair prejudice; and (4) the court administered an appropriate limiting instruction.
When "other act" evidence is offered to show knowledge or intent in particular, as opposed to other non-propensity purposes such as proof of identity or corroboration of witnesses, such evidence must be sufficiently similar to the conduct at issue to permit the jury to draw a reasonable inference of knowledge or intent from the other act. If the other-act evidence does not provide a reasonable basis for inferring knowledge or intent, its offer for that purpose should be rejected on grounds of relevance. Evidence of other acts need not be identical to the charged conduct to show knowledge or intent pursuant to Rule 404(b), so long as the evidence is relevant in that it provides a reasonable basis for inferring knowledge or intent. There is no necessity for synonymity but there must be substantial relevancy.
Although Cadet's interaction with Agent Green was not identical to the interactions that the taxpayer-clients described in their testimony, n6 Cadet's preparation of a false return for Agent Green was sufficiently similar to the charged conduct to be relevant to — and indeed highly probative of — knowledge and intent. At trial, Cadet contested the willfulness of his conduct by arguing that, in preparing his clients' returns, he either relied on inaccurate information provided by the clients or made mistakes. Evidence that Cadet knowingly and intentionally prepared a false tax return for Agent Green by inserting inflated deductions on the Form 1040 thus provided a reasonable basis to infer that the inflated deductions that formed the basis for the 16 counts of conviction were likewise entered knowingly and intentionally.
n6 For example, the taxpayer-clients testified that Cadet neither suggested preparing a false tax return nor demanded a higher fee for doing so. In addition, Cadet prepared Agent Green's return during a visit on April 4, 2006, contemporaneous with five of the charges on which Cadet was indicted, but subsequent to 14 of the 16 counts on which Cadet was ultimately convicted.
Moreover, the high probative value of the evidence substantially outweighed the risk of unfair prejudice [paraphrasing of FRE Rule 403's term "danger of unfair prejudice], as it was accompanied by a careful limiting instruction by the seasoned District Judge to the jury that "[e]vidence of similar acts may not be considered by you for any other purpose," and "may not [be] use[d] . . . to conclude that because the defendant [committed the acts to which Green testified], he must have committed the acts charged in the indictment." We presume that the jury followed this limiting instruction.I suppose it is in the eye of the beholder as to whether the IRS agent's contrived circumstance was sufficiently similar to the situations of the clients involved in the charged counts (I could argue not, as Cadet apparently did), but I think the trial judge could have gone either way on that or on the residual issue of whether, given the differences, the risk of prejudice would be too great. Either way the trial judge went, I doubt that an appellate court would have second guessed the Judge's Rule 403 call. (Of course, had the trial court refused to admit the evidence, the only party that could complain would be the Government and the Government can't appeal acquittals.)
Sentencing and Restitution
The trial judge did make some clearly bad calls on sentencing and restitution, so the Court handily reversed for correction of the errors. One of these -- the reversal on sentencing -- may prove a pyrrhic victory for the defendant, so that is perhaps worth a comment.
The problem is that the trial judge -- then serving as the sentencing judge -- sentence the defendant on the 16 counts of conviction to a term of 41 months incarceration and three years of supervised release. He imposed that sentence on each count to be served concurrently. The problem is that that sentence exceeded the possible incarceration and supervised release periods, respectively. How so?
First, it is important to look at the indicated Sentencing Guideline's indiicated range for sentencing. The Guidelines range was 33 - 41 months. It seems clear that, in imposing a sentence of 41 months, the Judge sought to impose a high-end Guidelines range which he has considerable discretionary authority to do, providing that he does not exceed the maximum incarceration period he is allowed to impose. Therein lies the problem.
Had the defendant been convicted of a single count of conviction, it is clear that the judge could not have incarcerated him for more than 36 months even if the Guidelines range was higher. Thirty-six months is the maximum incarceration period for a Section 7206(2) conviction. The defendant, however, was convicted of 16 counts. By "stacking' the counts of conviction, the judge could easily impose a significantly greater sentence than 36 months (indeed he could go up to 16 times 36 months, although his practical ability to do so is mitigated by the Guidelines and common sense). But the judge did not stack; rather, he imposed the sentences on the counts of conviction to run concurrently Specifically, for each count of conviction, the judge imposed a sentence of 41 months (exceeded the statutory limit) and a three year supervised release period of 3 years (exceed the statutory limit of one year per count.
So, the Court of Appeals handily remanded for resentencing.
But, the sentence may stand if, on remand, the trial judge simply stacks at least enough to get the sentence he previously imposed rather than imposes the sentences to run concurrently. Alternatively, I suppose, the judge could just sentence to 36 months incarceration and 12 months supervised release on the theory that is close enough.