This posting was made erroneously to my Federal Tax Crimes Blog. Here is the link to the posting on the Federal Tax Procedure Blog-Tax Court Issues a Precedential Opinion (T.C.) Holding that Form 872-T Does Not Revoke a Proper Form 872 Consent to Extend the Statute of Limitations; Well, Duh!.
My apology for the commotion.
Jack Townsend
In Skatteforvaltningen v. Markowitz, __ F.4th ___ (2d Cir. 2026), CA2 here and GS here [to come], the panel affirms the district court’s judgment based on a jury verdict holding the defendants civilly liable to the Danish government for participating in a massive fraud using the strategy sometimes called “cum-ex”. I have previously written on variations of this fraud here. Basically, through hocus-pocus, the defendants claimed to have stock in Danish companies on their dividend date which would have meant that they would be entitled to the dividends and those dividends would require withholding at 27%. However, under the U.S.-Denmark income tax treaty, dividends paid by a Danish company to U.S “shareholders that are United States pension plans, which are exempt from [U.S.] taxation” (Slip Op. 9) may be refunded to the qualifying shareholders. Basically, the principals or agents of the taxpayers figured out (i.e., created ex nihilo) a way to claim the refunds with respect to stock they never owned and with respect to dividends they never received and thus withholding tax that was never paid in. In this civil suit, Denmark seeks to recover from the participants and beneficiaries of this fraud on its fisc. Denmark succeeded at trial and prevailed on appeal.
The underlying events are straight-forward fraud wrapped in a supposed refund claim. The principal issue on appeal is whether the suit sought liability that was subject to the so-called “revenue rule” where, in Anglo-American jurisprudence, the courts of one sovereign (here the U.S.) will not enforce another sovereign’s tax laws. (See Federal Tax Crimes Blog posts on the revenue rule here.) The panel rejects the defendants’ revenue rule argument. (Slip Op. 20-35.) The Court’s lengthy consideration of that issue is, I think, correct. As suggested above, the refund claims were simple fraud in the guise of refund claims and had nothing to do with Danish tax law. Another way of looking at it is that the transactions were an abuse of the U.S.-Denmark tax treaty which U.S. courts should not be precluded from considering and providing appropriate relief.
In any event, at trial and on appeal, the defendants conceded that they did not own the stocks and hence were not entitled to the refunds they claimed. They instead argued that they had no reason to know that their refund claims were bogus. Without that knowledge, they claimed, their claims were not fraudulent. The jury decided that they had the requisite mens rea for liability. Of course, that argument was bullshit given the red-flags involved and babies-come-in-baskets nature of their claims.
There were other arguments based on trial court evidentiary decisions and sufficiency of the evidence as to two defendants. (Slip Op. 36-61.) The issues are run of mine in criminal cases so I do not discuss them, but I do point interested readers to two discussions:
I have posted my Federal Tax Procedure Book to SSRN, see here. Since it is timely, although a moving target that we sure have not heard the last of yet, I thought I would post the concluding paragraphs of my discussion of Trump v. IRS that Trump and his buddies at DOJ (most notably Blanch, but not only Blanche) tried to use as a pretext to raid the Treasury.
In my opinion, Trump and his buddies new they would draw flack on that package, so they must have realized that the Anti-Weaponization Fund was just a ploy, a stalking horse. The real goal was tax audit immunity which for now has not been taken off the table. I discuss that in the Federal Tax Procedure Book (Student Edition pp. 802-805; Practitioner Edition pp. 1118-1192). I conclude that discussion with some further discussion questions for students of tax crimes and tax procedure:
Exercise for Students: Readers of Chapter Six discussing tax and tax-related crimes should be able easily to spot several tax and tax-related crimes that this conduct might implicate, particularly the ubiquitous defraud conspiracy. One possibility is that anyone participating materially in the audit immunity could be an affirmative act of evasion with respect to the taxes covered or an overt act of conspiracy (offense or defraud). In theory, if that were viable, all of the key players in this drama stand exposed to criminal prosecution. Of course, the prosecutions, if any, will have to be brought by DOJ which Trump can prevent while he is President but can then be brought in the next Administration unless Trump gives sweeping pardons to all who were involved. (I suspect and have read that those actors are counting on such pardons.) One final question worth asking is whether Trump’s control of DOJ and IRS could implicate some conduct for which the Supreme Court has given full or qualified immunity in Trump v. United States, 603 U.S. 593 (2024).
Another Exercise for Students: Tax procedure students should think about the settlement authorities discussed earlier in this text. Thus, generally, only the IRS has settlement authority for cases that it has not yet referred to DOJ. DOJ has settlement authority only for cases that the IRS referred. Facially, it appeared in Trump v. IRS that the IRS had only transferred authority to DOJ over the § 7431 wrongful disclosure suit. There is no indication that the IRS referred all of the matters sweepingly released in the Release Order to DOJ, and there is no indication that the IRS “settled” those claims.
Concluding thoughts: If nothing else, Trump v. IRS and its resulting commotions will occupy tax procedure and tax crimes enthusiasts for a long time.
This blog entry is cross-posted on the Federal Tax Procedure Blog here.
I have published to SSRN my Federal Tax Procedure Book Student and Practitioner Editions. For information on the SSRN links to download see the page on my Federal Tax Procedure Blog “Federal Tax Procedure Book (2026 Editions) (8/4/26)” here.
I am trying to wrap up the 2026 Editions of my Federal Tax Procedure Book but the ongoing drama by DOJ’s $2.77 billion Anti-Weaponization Fund and Trump and related party tax immunity does not permit an easy stopping point. But I have to stop and have just concluded all that will be in the 2026 Editions that I hope to publish later this week. I thought I would post the discussion below (the Student Edition version without footnotes by copy and paste into the blog below) and the Practitioner Edition version with footnotes that can be downloaded here.
2. Examples (Including Trump v. IRS).
A prominent example of this remedy is a suit brought by a Kenneth Griffin, a hedge fund billionaire. An employee of a third party contractor to the IRS, Booz Allen Hamilton, Inc., illegally accessed and disclosed the tax return information of Griffin and others to a news organization, ProPublica, which in turn published some of the tax return information. Griffin sued the IRS under (i) § 7431, alleging violation of § 6103, and (ii) the Privacy Act. The employee was prosecuted and pled guilty, receiving a five-year sentence. Griffin and the IRS settled the civil action resulting in a dismissal with prejudice. All of the terms of the settlement are not available, but apparently there was no monetary consideration and the IRS agreed to and did issue a public apology. Another reputed billionaire brought related action against the employee’s employer, Booz Allen Hamilton, Inc.
A more prominent example arising from the same mass disclosures is a 2026 suit Donald J. Trump filed in his nominal personal capacity for $10 billion damages (asserting both the minimum $1,000 per disclosure with disclosures at $1,000 justifying $10 billion or actual damages of $10 billion) and for punitive damages in an amount not stated. The Plaintiffs included Trump’s sons and The Trump Organization, LLC. (referred to collectively as the Trump Plaintiffs). Before the DOJ filed an answer, the Judge asked the parties to brief whether, given President Trump’s control over the Government parties (IRS and DOJ) and personal interest as Plaintiff, the case met the required Article III case or controversy requirement. The Court also appointed amici to provide independent briefing on that issue. Before the parties presented their briefing but after the amici provided its initial brief, the Trump Plaintiffs moved to dismiss with prejudice under FRCP Rule 41(a)(1)(A)(i) which requires dismissal with prejudice. On May 18, 2026, the Court dismissed with prejudice, noting:
This entry is to alert readers to a recent article on CI. Joseph De Gregorio, Changes to IRS Disclosure Program Should Spark Compliance Checks (Bloomberg Tax 6/26/26), here. The author summarizes some points made by CI Chief Jarod Koopman at the NYU Tax Controversy Forum.
Key points of the article are (I have bold-faced some from the quotes):
1. “The last 18 months produced a detection gap. Approximately 1,700 IRS-CI employees had been reassigned to Immigration and Customs Enforcement operations by September 2025. Abusive tax scheme investigations collapsed 63% in fiscal year 2025 from 92 to 34, the lowest level in a decade. The enforcement budget was cut 8% for 2026, to its lowest inflation-adjusted level since 1988.”
2, The diverted CI agents are “coming back.”
3. Technology should permit CI to do its job more efficiently. “IRS-CI is running large language models inside its own firewalls, on its own data. The output is enhanced pattern recognition across the Form 1099 universe, the Foreign Bank and Financial Accounts database, the beneficial ownership registry, financial institution suspicious activity reports, and the international information exchange pipeline built under Foreign Account Tax Compliance Act and the common reporting standard. Fewer agents, but agents working with a detection instrument the prior enforcement cycle never had.”
4. “Koopman confirmed that approximately 60% of CI’s current work is on tax fraud.”
5. “Reduced IRS-CI headcount doesn’t mean reduced criminal tax risk. It means the risk shifted from broad detection across a wide population to deep detection concentrated on the cases AI-assisted analytics uncovers from third-party reporting, international data exchange, and financial institution data.
6. “When enforcement resources are constrained, agencies prioritize cleaner, more provable cases with strong paper trails. Clients with amended K-1s, offshore account statements, and Form 8938 discrepancies are exactly who the algorithm finds first.”
In United States v. O’Donoghue, ___ F.Supp.4th ___ (D.D.C. 7/22/26), the Court (Judge Colleen Kollar-Ketelly) dismissed 8 Counts of a 9 Count indictment for lack of venue in the District of Columbia. The opinion can be viewed: CL here, TN here, and GS here; the indictment can be viewed: CL here. The dismissed Counts were for tax evasion § 7201 and subscribing to false tax returns § 7201(1) (aka tax perjury). The undismissed count (Count 9) was for making false statements in a proffer session at DOJ offices in D.C., in violation of 18 U.S.C. § 1001.
This post will initially be a notice of the opinion with limited comment. I will post further analysis tomorrow. In the meantime, my initial reading is that it covers a lot of ground familiar to fans of tax crimes (the study rather than the commission). I say that because many leading tax crimes cases make at least a cameo appearance in the opinion.
The overall crux of the case is Constitution Article III’s requirement that trial “of all Crimes . . . shall be held in the State where the said Crimes shall have been committed.” This limits statutes that might otherwise permit venue elsewhere. The Government was relying upon such statutes to permit venue. More on this later. The Government was also arguing that the essential conduct for the 8 dismissed counts included conduct that was outside the U.S., thus permitting venue in the District of Columbia. The Court swatted that down. More on this later.
So, I will just post my quick reactions here and be back with more tomorrow.
1. The lawyering for O’Donoghue appears to have been great! He had good counsel, identified in CL here (click the Parties and Attorneys tab). I haven’t read the briefs but may do so tomorrow or even later—emphasize may because I am otherwise busy). I assume that there was good briefing because I infer that the Court worked from the submissions. Added 7/24/26 4:00pm: The CL Docket Entries are here (some with links to the briefing at ## 79, 89, 95 & 113 (all of . those key documents can be viewed rom the CL docket sheet, except the first; if someone who has joined CL will the CL docket sheet and retrieve it from PACER, it should show up on the CL Docket Sheet; I messed it up on PACER and PACER will not let me retrieve it again.)
2. This may be a pyrrhic victory because, even if the statute of limitations has otherwise closed, 18 U.S.C. § 3288, here, extends the statute of limitations for 6-months after dismissal, assuming an original timely indictment in D.C. Thus, it appears that the Government can seek an indictment in the Western District of Texas which will be a bit of an inconvenience because the undismissed Count (Count Nine, for false statements) will presumably be tried in the District of Columbia, which could add major costs to the Government and O’Donoghue.
Added 7/24/26 1:15pm:
Today, the Supreme Court denied the petition for certiorari in Murrin v. Commissioner (Sup. Ct. No. 25-988), docket here. See Order List dated 6/22/26, here at p.3, The Third Circuit opinion from which Murrin sought cert was Murrin v. Commissioner, 158 F.4th 527 (3rd Cir. 2025), here.
The question presented in the petition here was:
Whether, under 26 U.S.C. § 6501(c)(1), the IRS may assess tax beyond the Code’s three-year limitations period based solely on the fraudulent intent of a third-party, even when the taxpayer herself neither intended to evade tax nor knew of any wrongdoing.
The question as framed by the SG in the Commissioner’s Brief in Opposition here was:
Whether the indefinite limitations period in 26 U.S.C. 6501(c)(1) applies to a false or fraudulent return prepared by a tax return preparer who acted with the intent to evade tax.
In Hunter v. United States, 608 U. S. ____ (2026), SC here and GS here [to come], the Court held (from syllabus):
An agreement [plea agreement] not to appeal a sentence is unenforceable when it would result in a miscarriage of justice—meaning, when it would leave in place the kind of egregious error that would bring the judicial system into disrepute.
That’s a broadly stated general rule that will be fleshed out in its application. However, I will not discuss the holding further. Rather, I focus on a misleading statement made by Justice Thomas in his solo dissent that states a common misconception about plea bargaining.
Justice Thomas states (Slip Op. 39 of pdf, indicating p. 1 of dissent here, emphasis supplied by JAT):
Thanks to the [plea] agreement, Hunter received a 51-month prison term, followed by three years of supervised release, less than 2% of the prison time to which the indictment exposed him.
Added 6/21/26 2:30pm: Justice Thomas makes the statement to explain what the benefit of the bargain was for Hunter in order to support his [Thomas'] claim that the agreement appeal waiver should be binding.
No competent lawyer negotiating the plea would have negotiated against the maximum suggested in Justice Thomas’ statement; rather they would have negotiated against the sentence ranges provided by the Guidelines. I demonstrate with a simple example:
Assume that a taxpayer is convicted of 3 counts of tax evasion, with each count carrying a 5-year maximum sentence. In theory, that might permit "stacking" to reach the maximum sentence of 15 years if convicted of all counts. In fact, the Guidelines Offense Level maximums range from 6 months for tax losses from $2,500 or less to 36 for more than $550 million. Assuming no other adjustments (such as criminal history, etc.), looking at the maximum Offense Level of 36, the Guidelines Sentencing Table maximum range is 188-185 months (about 16 years). Most tax evasion convictions involve tax loss far less than $550 million, so the realistic range is far less than 25 years.
I illustrate with a more realistic tax loss example: Assume tax evasion loss of, say, $20 million (aggregate on 3 counts) producing a Guidelines Offense Level of 26 and Sentencing Table range of 63-78 months (about 5-6 years), again assuming no other adjustments.
In United States v. Chollet, ___ F.4th ___ (4th Cir. 2026), CA4 here and GS here, the 4th Circuit panel (Judges Niemeyer, Thacker, and Rushing) affirmed the conviction of three defendants, specifically rejecting various points that I discuss below.
The defendants were convicted of a variant of a marketed bullshit tax shelter. Two of the defendants—Kohn and his daughter Chollet—were tax lawyers; the third defendant was an insurance broker. I will not get into the specifics of the shams they created for their clients to (i) improperly hide their clients' income and resulting tax liabilities from the IRS and (ii) to make money for themselves as they shared in the false tax savings the taxpayers (clients) claimed. Suffice it to say that the scheme involved meaningless (i) limited partnerships, (ii) fake charitable contributions, (iii) fake royalties and management fees, and (iv) supposed life insurance policies.
Like all bullshit tax shelters and some bogus tax shelters that avoid the bullshit categorization, they work if they are not audited or otherwise discovered in the tax enforcement system. These tax enforcement system discovered and prosecuted these defendants. Most often, such scams result, if discovered, in civil tax and perhaps civil penalties. But, for some reason this one turned criminal. (I dare say that many similar bullshit tax shelters escape tax enforcement scrutiny and therefore “work” because of inadequate enforcement resources; for which I make an editorial comment: more enforcement resources would pay for themselves in multiples and promote fairness in and respect for the system.)
The counts of conviction are described as follows (Slip Op. 5-6]:
After a two-week trial, the jury convicted all three defendants of conspiracy to defraud the federal government, in violation of 18 U.S.C. § 371; Kohn on 16 counts of assisting in the filing of false tax returns, in violation of 26 U.S.C. § 7206(2); Chollet on 13 counts of assisting in the filing of false returns, in violation of § 7206(2); and Simmons on 11 counts of assisting in the filing of false returns, in violation of § 7206(2), and on 5 counts of filing false tax returns, in violation of § 7206(1).
Some of the defendants’ arguments on appeal are of the “Hail Mary” procedural footfault genre. (The best recent example of that genre is the commotion plaguing the IRS and the courts about § 6751(b)’s written manager approval for penalties requirement as courts have flailed around trying to interpret and apply nonsensical statutory text. See e.g., Tax Court Rejects Various Hail Mary Claim, Including APA Claims, to Get Out of Penalty Free (10/8/25), here).
In United States v. Fecondo (3rd Cir. No. 24-1618 5/26/26) (NonPrecedential), CA3 here and GS here the court held that Fecondo’s crimes of conviction under § 7202 for failure to withhold and pay over the employee share (or portion) of FICA does not include the tax loss from the employer’s failure to pay the employer share. (The employee share is sometimes called a "trust fund" tax because deemed to be held in trust to pay over to the IRS.) The employer share is not a tax collected from the employee (by deduction from wages) withheld within the scope of § 7202, the offenses of conviction. Thus, the unpaid employer share could not be included in the tax loss for the counts of conviction. Moreover, the Court held that the employer share is not “relevant conduct” for inclusion in the tax loss under the Sentencing Guidelines. The reasoning is that, although the two portions are related in the sense that they arise from the employer’s payment of wages, failure to pay a direct tax (employer’s share) is not related to the crime of failure to withhold and pay over the employee’s share for relevant conduct purposes. They are related but not relevant conduct to the counts of conviction.
I wonder if that works in reverse. Say the defendant was convicted of tax evasion, § 7201, for failure to pay the employer’s taxes (including the employer share of FICA). Could the failure to withhold and pay over the employee share of FICA be relevant conduct?
In United States v. Garza (N.D. TX 3:22-CR-0390-8), Memo and Order dated 5/6/26, CL here and GS here, the Court dismissed certain counts in a long-standing indictment. (No speedy trial here.) In broad strokes, the indictment alleges that Garza, a Dallas attorney (at least at the time of the charged conduct) and the other co-defendants participated in promoting a “fraudulent tax shelter plan to clients.” (I have a somewhat fuzzy recall that it was Garza’s version of a BLIPS variant.) They were indicted for tax crimes (§ 7206(1)) and for wire fraud crimes, on some counts related to the charged tax crimes.
The Court describes the defendants’ motion as (Slip Op. 2):
Defendants move to dismiss the Wire Fraud Counts under Federal Rule of Criminal Procedure 12 on the grounds that: (1) the Wire Fraud Counts function as an end-run around the willfulness mens rea requirement for tax fraud in violation of the Fifth Amendment’s Due Process Clause; (2) 26 U.S.C. § 7206(2) (“Tax Fraud Statute”) preempts the application of 18 U.S.C. § 1343 (“Wire Fraud Statute”) “when both rest on the same alleged scheme”; and (3) the Wire Fraud Statute, interpreted according to its text and constitutional footing, does not extend to tax fraud against the Federal government and cannot be used to prosecute tax conduct.
The Court granted the motion as to the overlapping charges, dismissing the wire fraud counts related to the tax counts but leaving the false tax return counts. The gravamen of the holding is that by charging basically the same conduct (or at least intertwined conduct) as tax crimes with the high “willfully” standard of proof (the Cheek standard) and as wire fraud, the Government was trying to bootstrap a conviction for that conduct under the lesser burden of proof for wire fraud. The Court concluded, somewhat conclusorily, that the phenomenon violated due process, almost as if she knew a due process violation when she saw it. (To paraphrase Potter Stewart’s famous comment on pornography.) She thus says (Slip Op. 3):
In this case, the Government’s prosecution of the same tax conduct under both the Tax Fraud Statute and Wire Fraud Statute violates the Due Process Clause by enabling the Government to avoid proving every fact necessary to constitute tax fraud.
The opinion is short and, since I
am leaving Charlottesville for D.C., the primary purpose of this blog (at least
as preliminarily offered) is notice. Readers can read the opinion and draw such conclusions as they wish. (I may add to this blog later, likely not until Monday.)
Quick questions:
In Thody v. Commissioner, T.C. Memo. 2026-30, the Court sustained the deficiencies asserted against Thody with respect to the tax loss that had been subject to a restitution order against Thody in an earlier criminal tax prosecution. The Thody opinion can be viewed at TC No. 27415-21, here, at #50 dated 3/30/26 and GS here. Actually, the original notices of deficiency exceeded the amount in the restitution order, but the IRS conceded (pp.. 4-5) the excess deficiencies asserted in the notice of deficiency, so that with that concession, the amounts of deficiencies the IRS asserted in the case equaled the amounts in the criminal restitution order. The Court then sustained the deficiencies based on the evidence.
One unexplained apparent oddity is that the IRS did not make a restitution-based assessment (“RBA”) allowed by § 6201(a)(4). The Court offers no explanation and treats the case as a straight-forward deficiency case permitting Thody to contest the amounts. If the IRS had made the RBA in the same amounts, Thody could not have contested the amounts of the RBA. In that regard, the IRS can assert deficiencies in amounts exceeding the amounts of restitution, whether or not asserted in an RBA. I don’t know why the IRS conceded the excess amounts. The IRS may have known or believed that it could not sustain that excess, so that this would be a normal concession in a deficiency case. But, if as a straight deficiency case, the IRS could have sustained the excess deficiency amounts, there was no reason to concede them. The IRS may have conceded just to move the case to a prompt decision with less hassle. A related question is whether, once the IRS decided to concede the excess before the trial level consideration was concluded, the IRS could have made an immediate RBA which would preclude Thody from contesting the amounts. I am not sure that there is a statute of limitations on RBAs because I have not researched that issue. And I am not sure that the Court would have treated such a belated RBA as mooting the deficiencies case.
Moreover, the Court seems not to have not the distinction between a tax deficiency and restitution. Thus, at p. 3, the Court says that the Government reduced the restitution (not the RBA) to judgment. In doing so, the Court cites in fn. 4 the IRM for the purpose of suits to reduce tax claims to judgment is to extend the statute of limitations. The IRM provision, I think, relates to tax assessments rather than criminal restitution. Of course, if there is a statute of limitations on restitution (likely), the reason to reduce restitution to judgment may be for the same.
Thody made the argument that the payments he had made for restitution should reduce the amount of the deficiencies. The Court (pp. 7-8) did correctly find that Thody’s argument was incorrect. The Court notes that, although in collecting on any resulting deficiencies approved by the Tax Court, the IRS would have to credit the restitution payments against the tax liabilities.
This blog content is cross-posted on the Federal Tax Procedure Blog here.
I start with a caveat: although this posting is on April 1, sometimes called April Fools Day, this is intended as a serious discussion.
In Daugerdas v. Commissioner, 171 F. 4th 924 (7th Cir. 2026), CA7 here and GS here, the Court held that § 6201(a)(4)(A), which authorizes the IRS to assess and collect restitution awarded in a criminal proceeding for unpaid tax, was a stand-alone collection authority unaffected by the payment schedule the district court imposed for the restitution behind the tax assessment. The assessment is sometimes called “restitution based assessment,” and acronymed to RBA which I use here. The holding seems like a straight-forward holding. But there are some issues lurking in the case that tax procedure enthusiasts may enjoy or at least understand.
First, I offer background worthy of note:
1. Daugerdas, a lawyer, is a notorious promoter of bogus tax shelters who was convicted. The Court says (pp. 2-3, emphasis supplied by JAT):
In 2013 a federal jury in Manhattan found Daugerdas guilty of one count of conspiracy to defraud the IRS (18 U.S.C. § 371), one count of mail fraud (18 U.S.C. § 1341), four counts of client tax evasion (26 U.S.C. § 7201), and one count of obstructing the internal revenue laws (26 U.S.C. § 7212(a)). His sentence brought with it an obligation to pay restitution of $371,006,397 jointly and severally with his co-conspirators for the tax losses resulting from the fraud perpetrated on the U.S. Treasury. The district court established a schedule of payments requiring Daugerdas to pay 10% of his gross monthly [*3] income starting 30 days after his release from prison. The Second Circuit affirmed Daugerdas’s convictions and sentence. See United States v. Daugerdas, 837 F.3d 212 (2d Cir. 2016).
I have not tried to break down the components of the restitution amount. Specifically, I have not tried to determine whether the restitution relates to Daugerdas’ tax liabilities (he did make a whopping amount of gross income that he likely attempted to shelter with similar bullshit strategies) or includes in whole or in part the liabilities of other persons reporting on the basis of bullshit tax shelters he promoted with legal opinions and related services. I don’t know that the difference makes a difference in terms of the RBA.
I have written on Daugerdas several times on my Federal Tax Crimes Blog, here (the results are initially by relevance but may be sorted by date).
I have written on the issue of whether the § 6038(b) penalty can be assessed (with the assessment collection tools available) or must be collected by a collection suit in the district court. See posts here. The Tax Court held in Farhy v. Commissioner, 160 T.C. 399 (2023), GS here, that the IRS could not assess and collect but rather must sue and collect. The D.C. Circuit reversed. Farhy v. Commissioner, 100 F.4th 223 (D.C. Cir. 2024), CADC here**, and GS here. The Tax Court stuck to its Farhy holding in cases appealable to Circuits other than the D.C. Circuit under its Golsen rule, feeling that its original Farhy holding was correct.
Today, the Second Circuit has now aligned with the D.C. Circuit, saying that the IRS can assess and collect the § 6038(b) penalty. Safdieh v. Commissioner, ___ F.4th ___ (2d Cir. 2/27/26), CA 2 here, TN here, and GS here. Other than to say that I think the D.C. Circuit and the Second Circuit are correct, I can’t add any discussion not evident from my earlier posts linked above. Thus, this is a notice-only blog.
This post is cross-posted in my Federal Tax Procedure Blog, here.
I write today on facets of Rule 6(e)(2), Federal Rules of Criminal Procedure, relating to grand jury secrecy. Rule 6 may be viewed here. Rule 6(e)(2) in part relevant to this blog entry requires that Government personnel participating in a grand jury investigation “must not disclose a matter occurring before the grand jury." Rule 6(e)(2)(B).
In Kalbers v. Volkswagen AG, ___ F.4th ___ (9th Cir. 1/30/26), CA9 here and GS here [to come], the Court held in a FOIA proceeding that Rule 6(e)(2) applies to documents delivered in response to a grand jury subpoena. Professor Kalbers sought Volkswagen’s Counsel’s responsive “file” containing millions of documents that Volkswagen, a target or subject, delivered in response to the grand jury subpoena. I think the case assumes that at least some portion of the documents were never presented to, summarized, or otherwise considered by the grand jury. The Court held that subpoena-responsive documents are grand jury matters subject to Rule 6(e)’s secrecy requirement and therefore are not disclosable under FOIA. The reasoning appears solid to me. I won’t track the reasoning here because it is not necessary for the focus of this blog entry.
A tangentially related issue to this is whether, in a tax crimes case, DOJ Attorneys have authority to conduct at the same time (i) a grand jury investigation of potential tax crimes; and (ii) an independent DOJ investigation of the same tax crimes? Such a parallel investigation would necessarily involve grand jury matters bleeding into the so-called independent DOJ tax crimes investigation. Is that an improper use of grand jury matters in violation of Rule 6(e)?
I write today on one of the more unusual documents I have seen filed in a criminal tax proceeding. By jury verdict, Richard Brasser and a co-defendant (Gregory Gentner), co-officers of rFactr (a software company) were convicted of 5 counts of § 7202 willful failure to account for and payover trust fund taxes and acquitted of 2 counts each of § 7201 false returns and one count (for Brasser) of §7201 tax evasion. The judge sentenced Brasser and the co-defendant to incarceration. The co-defendant started and completed his prison time. By contrast, Brasser strung out the time to report to prison, but that time fell due this month (January 2026).
As noted Brasser filed usual motions upon conviction but did not succeed. Finally, seeking the third delay to report to prison, Brasser, acting pro se, filed the document I discuss here. The document is at #161 of the docket entries. (The CL docket entries are here.) Unfortunately, the particular document is not available on CL, but I have posted it to my Google Docs here. At the point of filing the document, the issue before the court was whether the court would allow the third requested extension of time to report to prison.
The document is a letter dated June 9, 2024 from Mark Matthews to “Whom It May Concern” with a Re caption “Richard Brasser – Tax Matter.” The body of the motion says that the author had consented to its filing. The body of the letter is important so I copy and past it here (bold-face supplied by JAT):
I am writing on behalf of Richard Brasser to provide important context in connection with his conviction for what are commonly referred to as employment tax violations (while being acquitted of the more commonly known charges of false tax returns and evasion).
I have significant experience within our criminal tax system, having served as an Assistant US Attorney in the Southern District of New York, the Deputy Assistant Attorney General for the Department of Justice's Tax Division where I was responsible for all tax prosecutions nationwide, the Chief of IRS Criminal Investigation (which is the sole investigative agency for suspected federal criminal tax violations), and the Deputy Commissioner of the IRS (or the #2 official), responsible for all examination, collection, and investigation activities as well as taxpayer services. I have been involved in many hundreds of tax investigations and prosecutions for over 30 years on both sides of the table - as a prosecutor, investigator and as a defense attorney. Based on my experience the indictment of Mr. Brasser was entirely unwarranted and one of the most egregious examples of prosecutorial discretion I have ever encountered.
I wrote on the pardon of Roger Ver, the self-described Bitcoin Jesus who fled the country and renounced his U.S. citizenship rather than face justice for his alleged tax crimes. A Sad Event for the Country’s Criminal Tax Enforcement System (Federal Tax Crimes Blog 10/15/25), here. Propublica today offered a reasonably in depth article on Ver’s pardon. Avi Asher-Schapiro & Molly Redden, How “Bitcoin Jesus” Avoided Prison, Thanks to One of the “Friends of Trump” (Propublica 1/22/26), here.
I don’t think I can add anything that would be meaningful to readers of this blog. I just strongly recommend the article.
One minor quibble. The article quotes me but erroneously states that I am a former federal prosecutor. While with DOJ Tax, I was not a prosecutor; rather I handled civil tax appeals and the civil tax trials.
In United States v. Lemay (S.D. N.Y. No. 1:21-cr-00573 Opinion & Order dated 1/2/26), CL here, GS here, TN here, the Court denied the convicted defendant’s Rule 33 motion based on alleged ineffective assistance of counsel. Lemay and another defendant had been convicted in SDNY of conspiracy to defraud the U.S. under 18 USC 371. Lemay was originally indicted for tax evasion as well, but the evasion counts were severed and moved to New Jersey based on his residence. At the end of the SDNY trial, the Court denied a Rule 29 motion for judgment of acquittal. Shortly prior to sentencing, Lemay filed his Rule 33 motion. The Court then held an evidentiary hearing on the motion; this Opinion & Order is the result of the hearing.
The gravamen of the motion is that a person allegedly agreeing to pay Lemay’s costs of defense in the investigation and prosecution caused Lemay to have ineffective assistance of counsel because of conflicts created by that arrangement and the relationship of Lemay’s counsel (2 different attorneys, one in the investigation prior to the indictment and the other in the criminal trial) to other counsel for other targets or defendants with allegedly competing interests under an alleged joint defense agreement ("JDA"). The facts are more detailed and well worth reading by students and practitioners to see the types of problems that might arise when a defendant is represented by counsel being paid by someone else. The Court calls those arrangements “benefactor payments.” Lemay’s counsel (2 of them) targeted in Lemay’s claims denied Lemay’s claims. After holding an evidentiary hearing in which the Court observed and assessed the credibility of Lemay and other witnesses (including the 2 counsel), the Court rejected Lemay’s claims and denied the Rule 33 motion.
Some points:
1. Perhaps the major point is that, for some reason, Lemay’s counsel (2 of them) allegedly failed to meet the requirement of “The Court’s Individual Rule” that stated "[w]henever defense counsel has received, or will receive, a benefactor payment that subjects counsel to a conflict of interest, he or she must immediately inform the Court and request a Curcio hearing." (See Slip Op. 8.) The Curcio hearing is named for United States v. Curcio, 680 F.2d 881 (2d Cir. 1982), here. At a Curcio hearing, the court can consider the nature of any conflicts the attorney may have and advise a defendant of the risks of the conflict, determine that the defendant understands those risks through questioning, and "give the defendant time to digest and contemplate the risks after encouraging him or her to seek advice from independent counsel. For a good discussion of the Curcio hearing in the Second Circuit, see United States v. Arrington, 941 F. 3d 24, (2d Cir. 2019), here.
