NOT RECOMMENDED FOR PUBLICATION File Name: 25a0187n.06
Case Nos. 23-1763/1843
UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT FILED Apr 07, 2025 ) KELLY L. STEPHENS, Clerk UNITED STATES OF AMERICA, ) Plaintiff-Appellee, ) ) ON APPEAL FROM THE v. ) UNITED STATES DISTRICT ) COURT FOR THE EASTERN CARL L. COLLINS, III, ) DISTRICT OF MICHIGAN Defendant-Appellant. ) ) OPINION
BEFORE: CLAY, WHITE, and NALBANDIAN, Circuit Judges.
NALBANDIAN, Circuit Judge. After a federal jury convicted attorney Carl L. Collins,
III, of willfully making false tax returns, he hired a new expert to criticize the case that the
government had just put on. After holding a hearing on the new evidence, the district court
concluded that Collins was looking to rehash arguments that he had already made. So the court
denied Collins’s motion for a new trial. On appeal, Collins argues that the district court erred in
that denial and in admitting certain “other acts” evidence at trial. Finding none of his arguments
persuasive, we affirm.
I.
Collins was a successful personal injury lawyer who practiced for over twenty years in
Michigan. Aside from his law firm, Collins owned three companies: Alpha Living, LLC (an
assisted living facility), MedCity Rehabilitation Services, LLC (a physical therapy company), and
-1- Nos. 23-1763/1843, United States v. Collins
First Third, LLC (a real estate rental company). Alpha Living and MedCity were treated as
corporations and had to file corporate tax returns. But Collins reported First Third and his law
firm on schedules attached to his personal income tax returns.
Collins’s convictions stem from the alleged mismanagement of his various Interest on
Lawyer’s Trust Accounts (IOLTA). An IOLTA is a specialized client-trust account for attorneys
to hold client or third-party funds during litigation. When a party wins or settles a case, the paying
party places the funds in the attorney’s IOLTA. The attorney should then distribute the funds to
themself, third parties, and the client. The attorney pays himself by removing earned fees and
reimbursement for litigation expenses from the IOLTA and placing it in a business operating
account. The Michigan Rules of Professional Conduct require timely disbursement of funds and
immediate removal of earned fees. Lawyers in the state may be subject to disciplinary proceedings
for comingling client and personal funds in an IOLTA.
The Michigan Attorney Grievance Commission disciplined Collins for his mismanagement
of several IOLTAs. At first, he used a single IOLTA as both a personal and business account.
Each time he won or settled a case, he would deposit the payment into an IOLTA and pay his
clients from this account. But he would also use the account for personal expenses, savings, paying
employees, and litigation expenses. In an interview with the Commission in 2012, Collins claimed
to have two IOLTAs at Comerica Bank––one that he used as an IOLTA and one that functioned
as an operating account. But the investigation revealed that he held four IOLTAs, including one
at Bank of America. Collins told the Commission officials that Bank of America accidentally
opened the account as an IOLTA, but he intended it to be a business operating account. During
the investigation, Collins attended a seminar by the Michigan Bar on how to follow trust
accounting rules. The seminar specifically instructed lawyers not to use their IOLTAs to conceal
2 Nos. 23-1763/1843, United States v. Collins
income from the IRS. In 2015, the Commission formally disciplined Collins for improper IOLTA
use.
Over the years, Collins hired various accountants and office support staff to help file his
taxes and manage his office finances. They each followed the same method for calculating
Collins’s income: adding up all deposits made to the operating account. None of them reviewed
deposits to his IOLTAs while calculating income. Collins’s tax preparers explained that he would
deposit his fees and expense reimbursements into the operating account from his IOLTA. So
deposits into that account should reflect his income. The tax preparers arranged the returns at
Collins’s direction and then he reviewed and signed them. Collins followed a cash-basis-
accounting method. So the tax system recognized his income only when the cash was received,
rather than when it was earned.
In 2014, the IRS audited Collins about his 2011 and 2012 taxes. The IRS requested various
records from Collins, including bank statements for all business and personal accounts. Collins
repeatedly did not attend scheduled meetings with the IRS and did not provide the requested
documents. Eventually, Collins told the IRS auditor that he kept all his bank accounts at Comerica
Bank and that he had no personal accounts. Collins did not mention his Bank of America IOLTA
and told the auditor that he did not pay for personal or business expenses out of the IOLTA. He
also said there was no money flowing between the law offices and Alpha Living.
In 2015 while reviewing documents associated with Collins’s real estate company, the
auditor noticed a bank account number that she did not recognize from the accounts Collins told
her about. From there, she discovered Collins’s Bank of America IOLTA. Soon after, the case
was referred to IRS-Criminal Investigations. In 2021, a grand jury indicted Collins of thirteen
counts of tax crimes, including several counts of making a false tax return and willful failure to
3 Nos. 23-1763/1843, United States v. Collins
file a return. The allegations covered Collins’s 2012 to 2018 tax years. But this appeal only
concerns certain personal and Alpha Living tax returns for 2012, 2015, and 2018.
In 2012, Collins reported a gross income of $1,674,010 on his individual tax return. He
reported $1,729,804 in total expenses, resulting in a loss of $55,794 and no taxable income. He
reported no income from company dividends. Collins later filed an amended 2012 tax return that
increased his taxable income by $186,763. He explained the changes on the amendment form as
“adjustment to revenue, rental additions, loss on sale of rental property, and removal o[f] aircraft
expenses and travel.” R.84, Trial Tr., pp.72–73, PageID 579–80. Because of the changes, Collins
paid $42,686 in total taxes.
In 2015, Collins reported gross income of $1,293,780 and expenses of $1,234,631. So he
reported a net profit of $59,149. For Alpha Living, Collins reported $218,577 in gross receipts.
And the company’s total reported taxable income came to $16,132. Finally, in 2018, Collins
reported $1,158,923 in gross receipts on his individual tax return. He reported $234,178 in profits
and a total taxable income of $147,024.
At trial, the government relied on the testimony of IRS Fraud Examiner Kelly Williams to
outline Collins’s underreported income. Like Collins, she followed the cash-basis-accounting
method. But rather than focusing solely on Collins’s operating account, Williams also reviewed
deposits into his IOLTAs. These deposits included dividends and distributions from Alpha Living.
Her calculations determined Collins’s income based on the current year’s fees that he earned. She
then subtracted disbursements to third parties and compared that amount to the gross income
Collins reported on his tax returns. But these calculations did not consider the amount of money
in his IOLTAs at the beginning of the year––only funds that went in or out of the account during
the year. Williams calculated Collins’s underreported income as follows:
4 Nos. 23-1763/1843, United States v. Collins
Count 1: 2012 Personal Return: $623,944.69 Count 9: Amended 2012 Personal Return: $578,934.69 Count 10: 2015 Personal Return: $860,318.83 Count 11: 2015 Alpha Living Corporate Return: $351,296.06 Count 12: 2018 Personal Return: $297,323.38
R.96, Trial Tr., pp.61, 68, 92, 118, 126, PageID 1546, 1553, 1577, 1603, 1611.
Collins relied on the expert testimony of Certified Public Accountant Dana Kaufman.
Kaufman critiqued the government’s methodology for calculating Collins’s underreported income,
saying it disregarded “outstanding checks or monies that may be due to third parties, other costs,
shared attorney fees” and the like. R.97, Trial Tr., p.76, PageID 1726. These third parties would
have been responsible for paying taxes on this money when Collins distributed it. Thus, according
to Kaufman, this tax liability should not have been attributed to Collins.
The jury disagreed. It convicted Collins of five counts of willfully making a false tax return
in violation of 26 U.S.C. § 7206(1). The counts covered Collins’s 2012 individual tax return, 2012
amended individual tax return, 2015 individual tax return, 2015 Alpha Living, LLC corporate tax
return, and 2018 individual tax return. But the jury acquitted Collins of seven counts of willful
failure to file a return in violation of 26 U.S.C. § 7203.
After the jury verdict, Collins moved for a new trial under Federal Rule of Criminal
Procedure 33. Collins urged the court to set aside his convictions and grant a new trial because
the government used a flawed methodology as to the central issue of underreported income. He
claimed it would be a miscarriage of justice to allow the convictions to stand. Yet in the motion,
he admitted that the government’s evidence was not “patently incredible.” R.129, Memo. Br. in
Supp. of Mot. for New Trial, p.2, PageID 2754 (internal quotation marks omitted). Instead, he
argued that Williams’s testimony was “so fatally skewed” and “so crucial to the prosecution’s
case . . . as to justify relief.” Id.
5 Nos. 23-1763/1843, United States v. Collins
Collins hired a new team of experts, including Andrew Perun, to analyze IRS Fraud
Examiner Williams’s testimony. The experts’ report found three flaws in Williams’s testimony:
(1) it assumed there were no outstanding obligations associated with the deposits, (2) it assumed
all services related to each settlement occurred in the same year the settlement deposit occurred;
and (3) it assumed the defendant received approval to recognize funds associated with each deposit
in the same year the deposit occurred. The new experts conducted their own analysis of Collins’s
tax returns and found that Collins had underreported his income in three years. But he overreported
his income in two years. And there were three years in which Collins overpaid his taxes due,
including on his 2012 amended return. Collins claimed this evidence undercut the government’s
argument that he willfully cheated the IRS and called for a new trial.
The district court held an evidentiary hearing to review the new report. Perun testified for
Collins and argued that the government improperly used a cash-basis-accounting method for
calculating Collins’s income. He claimed this method disregarded funds in Collins’s IOLTA that
he had not yet earned because it belonged to third parties. Instead, Perun advocated for a hybrid
approach. Williams returned for the government and critiqued Perun’s alternative methodology.
The district court denied Collins request for a new trial, viewing Collins’s motion as
“effectively request[ing] a new trial so that he may offer a more compelling expert.” R.143, Op.
& Order, p.7, PageID 3042. But the district court noted that Collins did not contend that
Williams’s testimony caused reversible error or a violation of Collins’s substantial rights. He also
did not allege his own expert witness was incompetent or otherwise unqualified. And much of his
original expert’s testimony at trial was consistent with the new expert’s report and drew on the
same logic about the flaws in the government’s methodology. So there was no basis for granting
a new trial.
6 Nos. 23-1763/1843, United States v. Collins
The presentence report found a loss to the IRS of $1,787,860 and recommended that
amount of restitution. Initially, the district court sentenced Collins to eighteen months’
imprisonment, one year of supervised release, and ordered him to pay $1,787,860 in restitution as
a condition of supervised release. But then, Collins and the government stipulated to a reduced
amount of restitution. The parties agreed that Collins would pay a total of $661,032, with credit
for the $102,796 he had paid since the proceedings began. The parties also agreed that the
“adjusted figure [was] premised on the Court’s determination that the government’s tax loss
calculations are correct” even though Collins disputed that fact. R.154, Stipulation, p.2, PageID
3549. Despite this stipulation, Collins did not waive his right to challenge the government’s
calculations.
The district court issued its final amended judgment on September 11, 2023, ordering the
stipulated restitution amount of $661,032.1 Collins appealed.
II.
Collins appeals his conviction for two reasons: (1) he argues the district court improperly
admitted highly prejudicial evidence of his Michigan bar disciplinary proceedings and (2) he
alleges the district court erred in denying his motion for a new trial. Appellant Br. at 30–31. Both
arguments lack merit. We consider them in turn.
A.
1.
The Federal Rule of Evidence 404(b) prohibits evidence of any “other crime, wrong, or
act” to “prove a person’s character in order to show that on a particular occasion the person acted
1 The court first issued an amended judgment accepting the parties’ stipulated restitution amount on August 21, 2023. But because of an error on the form, the parties stipulated again to an amendment to the amended judgment. 7 Nos. 23-1763/1843, United States v. Collins
in accordance with the character.” Fed. R. Evid. 404(b)(1). But this is not a complete bar.
Evidence that implicates Rule 404(b)(1) may still be admissible “for another purpose, such as
proving motive, opportunity, intent, preparation, plan, knowledge, identity, absence of mistake, or
lack of accident.” Fed. R. Evid. 404(b)(2).
Courts use a three-part test to decide whether otherwise prohibited evidence is admissible
under Rule 404(b)(2), and that test comes with its own standards of review. United States v.
Barnes, 822 F.3d 914, 920 (6th Cir. 2016). First, we scrutinize for clear error the district court’s
factual determination that the other act occurred. Id. Second, we review de novo the admissibility
of the act for a proper purpose other than character. Id. Third, we review for an abuse of discretion
the court’s determination that the probative value of the evidence was not substantially outweighed
by its prejudicial affect. Id.
2.
Collins challenges the admission of evidence that the Attorney Grievance Commission
disciplined him for IOLTA-related violations. The government counters that the evidence was
either admissible as intrinsic to the charged offenses or properly admitted extrinsic evidence under
Rule 404(b)(2) relating to Collins’s intent. Without opining on whether the evidence was intrinsic,
we agree with the government and district court that the evidence was admissible under Rule
404(b)(2) as extrinsic evidence of Collins’s intent.
The parties do not dispute that there is sufficient evidence of the Grievance Commission’s
investigation and disciplinary actions against Collins. So we focus on the last two prongs of the
Rule 404(b)(2) test: whether the other acts evidence was admitted for a proper purpose and whether
the evidence’s prejudicial effect substantially outweighed its probative value. Barnes, 822 F.3d
at 920. On the first, we have found that the use of other acts to show specific intent is a proper
8 Nos. 23-1763/1843, United States v. Collins
purpose under the rule. Id. at 921–22. To determine whether the evidence is probative of intent,
“we look to whether the evidence relates to conduct that is substantially similar and reasonably
near in time to the specific intent offense at issue.” Id. at 922 (quoting United States v. Bell, 516
F.3d 432, 443 (6th Cir. 2008)).
Once that bar is cleared, the district courts “have broad discretion to make determinations
of admissibility based on considerations of relevance and prejudice.” Id. at 923 (cleaned up). So
we will leave the lower court’s ruling on the third prong undisturbed “unless we are left with the
definite and firm conviction that the district court committed a clear error of judgment in the
conclusion it reached.” Id. (cleaned up).
Because the Grievance Commission’s investigation and disciplinary actions are relevant to
Collins’s intent to conceal his income, the evidence was admitted for a proper purpose under Rule
404(b)(2). The government had the burden of proving that Collins acted “willfully.” 26 U.S.C.
§ 7206(1). So the government must prove that Collins “voluntarily and intentionally violate[d] a
known legal duty.” R.104, Jury Instr., p.21, PageID 103. Collins’s attendance at a seminar
specifically instructing lawyers not to use their IOLTAs to evade the IRS shows that he knew of
his legal duty not to conceal his income in an IOLTA. And the fact that Collins concealed the
same IOLTA from both the state bar and the IRS during the same two-year period is probative of
whether Collins knew that the account revealed critical information about his income. So the
Grievance Commission’s investigation and disciplinary action provide evidence of “conduct that
is substantially similar and reasonably near in time to the specific intent offense at issue.” See
Barnes, 822 F.3d at 922 (internal quotation marks omitted).
Still, on the third prong, Collins argues that this evidence was unfairly prejudicial. But
following a hearing on the issue, the district court concluded that the evidence was “not of the type
9 Nos. 23-1763/1843, United States v. Collins
to inflame the passions of the jury or cause them to ignore the evidence,” in part because it was
financial and transactional. R.65, Op. & Order Den. Def. Mot. in Lim., p.19, PageID 381. The
court also predicted that the evidence would be “only a modest portion” of the government’s case
in chief. Id. This turned out to be true. Only two government witnesses out of seventeen discussed
the Grievance Commission’s investigation in depth. And these witnesses testified for less than
two hours of a twelve-day trial. Yet because the evidence was “substantially similar to and
intertwined with the alleged conduct,” the district court found that it had “significant probative
value which [was] not outweighed by its prejudicial effect.” R.65, Op. & Order Den. Def. Mot. in
Lim., p.19, PageID 381.
Collins’s only response is that the evidence may have suggested to the jury that improper
IOLTA use equates with filing a false tax return. But given the district court’s consideration of
the type and amount of this evidence offered in comparison with its probative value, we are not
left with a “definite and firm conviction that the district court committed a clear error of judgment
in the conclusion it reached.” Barnes, 822 F.3d at 923 (cleaned up). So the district court did not
err in admitting evidence of Collins’s Michigan bar disciplinary proceedings as extrinsic evidence
of his specific intent to falsify his taxes.
B.
We review a district court’s denial of a new trial motion for abuse of discretion. United
States v. Farrad, 895 F.3d 859, 885 (6th Cir. 2018). When assessing the motion, the district court
may assess the credibility of witnesses and the weight of the evidence. United States v. Hughes,
505 F.3d 578, 593 (6th Cir. 2007). Our review of that assessment is “limited to determining
10 Nos. 23-1763/1843, United States v. Collins
whether it was a clear and manifest abuse of discretion” to credit the witness’s testimony and weigh
the evidence as the district court did. Id.
Collins moved for a new trial under Federal Rule of Criminal Procedure 33(a). Under the
rule, “[u]pon the defendant’s motion, the court may vacate any judgment and grant a new trial if
the interest of justice so requires.” Fed. R. Crim. P. 33(a). The rule does not define the “interest
of justice.” United States v. Munoz, 605 F.3d 359, 373 (6th Cir. 2010). But we have found that a
district court may grant a new trial when the jury’s verdict was against the manifest weight of the
evidence or where a legal error occurred that affected a defendant’s substantial rights. Id. The
rule also explicitly provides that a defendant can request a new trial based on newly discovered
evidence. Fed. R. Cim. P. 33(b)(1).
Collins did not base his motion on any of these established grounds for relief. He expressly
told the district court that his claim was neither based on newly discovered evidence nor a claim
that the verdict was against the weight of the evidence. Instead, he claims it would be a
“miscarriage of justice” to allow the verdict to stand. Appellant Br. at 32. He argues that the
district court erred in denying his request for a new trial because the government’s testimony at
trial was flawed. He claims his post-trial expert report reveals that the government’s methodology
for finding his underreported income was “such an unreliable and superficial approach to income
calculation [that it] cannot rightly result in conviction.” Id. at 30.
Generally, district courts may grant a motion for a new trial when the verdict is against the
weight of the evidence, exhibits reversible error, or involves a violation of established rights.
Munoz, 605 F.3d at 374–76. We have not resolved whether there are other situations, when these
issues are not present, where the interest of justice would require a new trial. Id. We need not
11 Nos. 23-1763/1843, United States v. Collins
answer this question today because we have made clear that a new trial is an “extraordinary remedy
for ‘extraordinary circumstances.’” United States v. VanDemark, 39 F.4th 318, 322 (6th Cir. 2022)
(quoting United States v. Burks, 974 F.3d 622, 625 (6th Cir. 2020)). Whatever the basis, a district
court should only order a new trial if “the verdict exceeds the bounds of reasonableness.” Id.
(internal quotation marks omitted). And this court should only reverse that ruling if it was based
on a clear and manifest abuse of discretion. Id.
No matter how Collins frames his motion, it does not meet the high bar needed for a new
trial. The trial court’s conclusion that Collins post-trial expert did not undermine the
reasonableness of the verdict was not a clear and manifest error. The verdict was based on
arguments substantially similar to those presented by Collins post-trial. So the court did not abuse
its discretion in concluding that the new testimony would not sway the jury’s verdict.
To recap, Collins’s post-trial expert, Perun, argued that the government inflated Collins’s
purported income. He alleged that the government merely added up money in Collins’s accounts
without regard for their availability to Collins. In doing so, he argued that the government
improperly relied on a cash-basis-accounting method rather than a hybrid method to find Collins’s
income. Perun alleged three flaws in the government expert’s testimony: (1) it assumed there were
no outstanding obligations associated with the deposits, (2) it assumed all services related to each
settlement occurred in the same year the settlement deposit occurred; and (3) it assumed the
defendant received approval to recognize funds associated with each deposit in the same year the
deposit occurred. Because of these flaws, Collins argued that the government’s calculation of his
income was unreasonable and substantially incorrect.
But Perun’s testimony only expounded on arguments already made at trial. Collins’s
defense counsel cross-examined Williams several times at trial on how she considered funds
12 Nos. 23-1763/1843, United States v. Collins
potentially owed to third parties. He also questioned her on outstanding funds in Collins’s IOLTA
and how the beginning and ending balances in the accounts affected her calculations. Williams
clarified at trial that these balances were not relevant because income for a cash-basis taxpayer is
based on the income received and expenses incurred in the current year. Likewise, the defense
asked how the timing of when Collins earned the funds versus when he spent them affected her
determination of his annual income. But since Collins was a cash-basis taxpayer, Williams
explained that her review related only to the target-year income and expenses. She was also
specifically asked about how she handled funds that Collins received in his IOLTA but lacked
authority to disburse until later. Again, she reiterated that for a cash-basis taxpayer, the funds were
income when Collins’s fee was available and determinable rather than when he could remove it
from the IOLTA.
The district court did not abuse its discretion in crediting Williams’s testimony that
Collins’s income was determined by when the money was deposited in his IOLTA. If Collins
removed expenses from his IOLTA the next year, such as to pay third parties, it would have been
an expense for that year.
Collins relies on the U.S. Tax Court’s opinion in Isaacson v. Comm’r to say this is incorrect
and money kept in the IOLTA cannot be income to him. Appellant Br. at 33 (citing Isaacson v.
Comm’r, 119 T.C.M. (CCH) 1107 (2020)). Isaacson confirms that “a taxpayer does not recognize
as income funds the taxpayer holds in trust for the benefit of another.” 119 T.C.M. (CCH), at *10
(citing Ford Dealers Advert. Fund, Inc. v. Comm’r, 55 T.C. 761, 771 (1971), aff’d, 456 F.2d 255
(5th Cir. 1972)). But that is the extent to which it helps Collins. The tax court then explained that
only “when a lawyer reporting on a cash basis . . . complies with the requirements placed on trust
accounts by the relevant rules of professional conduct” are “the funds not income to the lawyer.”
13 Nos. 23-1763/1843, United States v. Collins
Id. (emphasis added). When a lawyer does not comply with the relevant rules, and instead
comingles client and attorney fees, the comingling obviates the timing of when the lawyer is paid
versus when the lawyer holds the funds in trust for his clients. Indeed, in Isaacson, the lawyer
comingled his fee and his clients’ funds and repeatedly used the funds for his personal enjoyment.
Id. at *9. So when the lawyer attempted to claim that the money in his account was not income,
the tax court rejected this claim and found the money to be income in the year he received it in the
account. Id. at *12. Treating the funds in this way accords with the general principle that a
taxpayer must report income during the earliest year that he actually or constructively receives it.
Id. at *10.
Collins did not follow the requirements placed on trust accounts by the Michigan Rules of
Professional Conduct. The Michigan Attorney Grievance Commission disciplined Collins for
comingling his fees and clients’ funds in his IOLTA. He used the account for both personal and
business expenses outside his legal work, like paying employees. Like the taxpayer in Isaacson,
Collins’s personal use of his IOLTA converted the account’s money into his income when he
received it. Id. at *10; see also Canatella v. Comm’r, 113 T.C.M (CCH) 1549, at *5 (2017)
(analyzing individual deposits to determine an attorney’s income in his IOLTA when he did not
follow the labeling, anti-commingling, and record-keeping requirements for an IOLTA).
Collins’s cross-examination of Williams at trial sought to reveal the exact flaws in the
government’s methodology that his post-trial expert later claimed made the jury’s verdict
unsustainable. But the jury considered and rejected these arguments. So it was not a clear and
manifest abuse of discretion for the district court to do the same.
Collins also presented his own expert witness at trial, Kaufman, whose testimony mirrored
Perun’s post-trial testimony, even if less comprehensively. At trial, Kaufman testified that money
14 Nos. 23-1763/1843, United States v. Collins
in Collins’s IOLTA may have been earned in years before the income tax year that the government
attributed it. He also discussed third-party funds in Collins’s IOLTA and argued that the
government’s analysis substantially overstated Collins’s income. Perun even agreed that although
his team had done “a more thorough accounting” than Kaufman’s trial testimony, he agreed with
Kaufman’s conclusions. R.142, Hr’g Tr., p.21, PageID 2970. Collins made arguments at trial,
like Perun’s, questioning the government’s methodology. But the jury still convicted Collins of
five counts of willfully making a false tax return. So the district court concluded “that the outcome
at trial would not have changed even with [the post-trial expert’s] more detailed analysis.” R.143,
Op. & Order, p.11, PageID 3046.
The one area where Kaufman’s trial testimony diverged from Perun’s post-trial testimony
was that Perun actually attempted to re-calculate Collins’s income. Kaufman testified that Collins
was a cash-basis taxpayer, and that the government made several improper assumptions in
calculating Collins’s income on a cash-basis. Perun took this a step further and tried to re-calculate
Collins’s income while accounting for the same improper assumptions Kaufman identified. Perun
testified that these calculations corrected the government’s “flawed” and “improper” “cash-basis
approach” by accounting for the fact that an IOLTA includes funds that belong to third parties.
R.142, Hr’g Tr., pp.9–10, PageID 2958–59. So all the defense experts—both at trial and post-
trial—agreed on the alleged flaws in the government’s cash-basis methodology. But Perun
performed calculations to correct for those alleged flaws, while Kaufman did not.
Given the substantial similarities between Collins’s arguments at trial and post-trial, the
district court found that Collins “effectively request[ed] a new trial so that he may offer a more
compelling expert.” R.143, Op. & Order, p.7, PageID 3042. This was not error. Indeed, we have,
in a tax case, rejected a defendant’s attempt to “relitigate his case under a new theory by hiring a
15 Nos. 23-1763/1843, United States v. Collins
new lawyer and a new expert to reinterpret the legal significance of evidence that was either
presented or available at trial.” Barrow v. United States, No. 96-1687, 1997 WL 31427, at *2 (6th
Cir. Jan. 27, 1997) (unpublished table decision). To be fair, Perun’s post-trial testimony provided
a much more detailed analysis of the government’s methodology than Kaufman’s did at trial. But
detail does not create novelty. And Collins may not use a Rule 33 motion to relitigate issues
already raised and rejected at trial. So the district court “did not abuse its discretion in denying
[Collins] a new trial on the basis of this evidence.” See id.
III.
Lastly, Collins urges us to remand this matter for a hearing to decide the proper amount of
restitution. He argues that the government inflated the amount that he underreported his taxes and
offers his post-trial expert’s analysis to argue that the district erred in imposing a restitution amount
based on the government’s evidence. We review a district court’s legal authority to order
restitution de novo, but the amount ordered for an abuse of discretion. United States v. Vandeberg,
201 F.3d 805, 812 (6th Cir. 2000). The government has the burden to prove the amount of the loss
by a preponderance of the evidence. Id. at 814.
The parties do not dispute that the district court could order restitution. So our review is
limited to whether the district court abused its discretion in ordering Collins to pay $661,032 in
restitution. Still, considering the district court’s authority for granting restitution helps in figuring
out if it abused its discretion in doing so. Federal restitution statutes do not authorize restitution
for tax crimes under Title 26. United States v. Kilpatrick, 798 F.3d 365, 391 (6th Cir. 2015). But
the law gives courts wide discretion to order restitution as a condition of supervised release. Id.
16 Nos. 23-1763/1843, United States v. Collins
So the court may impose restitution “for the full amount of the victim’s loss” as a condition of
supervised release. U.S.S.G. § 5E1.1(a)(2). When finding the total loss resulting from a tax
conviction, “all conduct violating the tax laws should be considered as part of the same course of
conduct or common scheme or plan unless evidence demonstrates that the conduct is clearly
unrelated.” U.S.S.G. § 2T1.1 cmt. n.2.
The presentence report determined the IRS’s loss based on Collins’s convictions and the
relevant conduct related to his convictions. The relevant conduct included Collins’s false
individual tax returns each year from 2010 to 2017, his false Alpha Living returns each year from
2010 to 2015, and his false MedCity returns each year from 2010 to 2013. It concluded that Collins
is responsible for a tax loss of $3,131,407. He had made payments toward that loss but still owed
$1,787,860. In the end, the parties stipulated to $661,032 in restitution, which is the amount
reflected in the amended judgment. In an earlier stipulation, Collins agreed that the adjusted figure
was “premised on the Court’s determination that the government’s tax loss calculations are
correct” even though he disputed those calculations and did not waive his right to challenge them.
R.154, Stipulation, p.2, PageID 3549.
Collins now challenges those calculations. He argues that the district court’s order of
restitution was in error for the exact reasons he argued for a new trial: because his post-trial expert
undermined the credibility of the government’s methodology for calculating his underreported
income. For largely the same reasons the district court did not abuse its discretion in denying
Collins a new trial, we hold that it did not abuse its discretion in crediting IRS Fraud Examiner
Williams’s testimony over both of Collins’s experts on the issue of the restitution.2 See United
2 Looking at two specific years, Collins makes one additional argument that his restitution amount is erroneous because it requires him to pay more than he owed. And he says that the government admitted as much in its stipulation. He contends that the government initially double- 17 Nos. 23-1763/1843, United States v. Collins
States v. Behnan, 554 F. App’x 394, 400 (6th Cir. 2014) (upholding a district court’s restitution
order against a similar challenge).
Because the district court did not err in crediting the government’s calculations and
Collins’s restitution is still far below the total loss, a remand to decide the amount of restitution is
unnecessary.
IV.
For the reasons stated, we AFFIRM.
counted dividends for tax years 2012 and 2015, but for restitution, the government realized its error and only double-counted dividends for 2015. The dividends at issue came from checks for Alpha Living that Collins deposited directly into his Bank of America IOLTA. In 2012, Collins deposited two checks totaling $663,905 into his IOLTA. But he reported no income from dividends on his tax return. Williams included these checks as dividends in Collins’s total gross income. In 2015, Collins deposited another two checks from Alpha Living into his IOLTA, this time totaling $361,467.23. He then transferred $5,000 back to Alpha Living. So Williams determined that Collins had $356,467.23 in unreported dividends income. The real problem stems from Collins’s decision to set Alpha Living up as a ‘C Corporation’ for tax purposes. Had Collins handled these checks properly, he would have deposited them into Alpha Living’s account and then given himself a dividend from Alpha Living. And they would have been income to both Alpha Living (as gross receipts) and Collins (as a dividend)––resulting in the ‘double counting’ Collins complains about. But Collins instead deposited the checks into the Bank of America IOLTA. Collins argues that the government essentially admitted to double counting when it lowered the restitution amount. Appellant Br. at 39. But the parties merely stipulated and agreed to pay a lower amount. This was not an admission by the government that the loss was improperly calculated. And, indeed, it inured to Collins’s benefit. In any event, the district court did not abuse its discretion in concluding that Williams did not double-count the dividends. 18