Complications from Extensions and Unprocessible Returns

Bob Probasco returns today with an important alert on overpayment interest. Christine

A few months ago, Bob Kamman flagged a number of surprising phenomena he’d observed recently, one of which related to refunds for 2020 tax returns.  Normally, when the IRS issues a refund within 45 days of when you filed your return, it doesn’t have to pay you interest.  That’s straight out of section 6611(e)(1).  Yet, taxpayers were getting refunds from their 2020 returns within the 45 day period and the refunds included interest.  As Bob explained, that was a little-known benefit of COVID relief granted under section 7805A.  Not only were filing and payment dates pushed back but also taxpayers received interest on refunds when they normally would not have.

On July 2, 2021, the IRS issued PMTA 2021-06, which raised two new wrinkles—one related to returns that were not processible; the other related to returns for which taxpayers had filed extensions.  How do those factors affect the results Bob described?

read more...

Interest on Refunds Claimed on the Original Return

The general rule is that the government will pay interest from the “date of the overpayment” to the date of the refund.  Normally the “date of the overpayment” will be the last date prescribed for filing the tax return, without regard to any extension—April 15th for individuals—unless some of the payments shown on the return were actually made after that date.  But there are three exceptions applicable to refunds claimed on the original return: 

  • Under section 6611(b)(2), there is a “refund back-off period” of “not more than 30 days” before the date of the refund, during which interest is not payable.  In practice, this period is significantly less than 30 days.  See Bob Kamman’s post and the comments for details.
  • Under section 6611(b)(3), if a return is filed late (even after taking into account any extension), interest is not payable before the date the return was filed.
  • Under section 6611(e)(1), no interest is payable if the refund is made promptly—within 45 days after the later of the filing due date (without regard to any extension) or the date the return was filed.

Section 6611(b)(3) and (e)(1) affect when overpayment interest starts running, while section 6611(b)(2) affects when it ends.

Other Code provisions, however, may adjust this further.

When Is the Return Considered Filed?

This is where “processible” comes in.  Section 6611(g) says that for purposes of section 6611(b)(3) and (e), a return is not treated as filed until it is filed in processible form.  As the PMTA explains, a return that is valid under Beard v. Commissioner, 82 T.C. 766 (1984), may not be processible yet, because it omits information the IRS needs to complete the processing task.  IRM 25.6.1.6.16(2) gives an example of a return missing Form W-2 or Schedule D.  Those are not required under the Beard test, but the return cannot be processed “because the calculations are not verifiable.”  The PMTA also mentions a return submitted without the taxpayer identification number; still valid under the Beard test but the IRS needs the TIN to process it.

As a result, under normal conditions, if a valid but unprocessible return is filed timely, and a processible return is not filed until after the due date, the latter is treated as the filing date for purposes of section 6611(b)(3) and (e).  The return is late, so the taxpayer is not entitled to interest before the date the (processible) return was filed.  And the 45-day window for the IRS to issue a refund without interest doesn’t start until the (processible) return was filed.

And Now We Add COVID Relief

The previous discussion covered normal conditions, but today’s conditions are not normal  As we all recall, the IRS issued a series of notices in early 2020 that postponed the due date for filing 2019 Federal income tax returns and making associated payments from April 15, 2020, to July 15, 2020.  The IRS did the same thing for the 2020 filing season, postponing the due dates from April 15, 2021, to May 17, 2021.  (Residents of Oklahoma, Louisiana, and Texas received a longer postponement, to June 15, 2021, as a result of the severe winter storms.)

This relief is granted under authority of section 7508A(a)(1), responding to Presidential declarations of an emergency or terroristic/military actions.  For purposes of determining whether certain acts were timely, that provision disregards a period designated by Treasury—in effect, creating a “postponed due date.”  The acts covered are specified in section 7508(a) and potentially, depending on what relief Treasury decides to grant, include filing returns and making payments.  For the 2019 filing season, the IRS initially postponed the due date for payments and then later expanded that relief to cover the due date for filing tax returns.

(Yes, the section 7508A provision for Presidentially declared emergencies piggybacks on the section 7508 provision for combat duty.  As bad as the disruptions caused by COVID have been, I prefer them to serving in an actual “combat zone.”  Hurricane Ida, which I am also thankful to have missed, may be somewhere in between.  All get similar relief from the IRS.)

Section 7508A(a)(2) goes beyond determining that acts were timely.  It states that the designated period, from the original due date to the postponed due date, is disregarded in calculating the amount of any underpayment interest, penalty, additional amount, or addition to tax.  No penalties or underpayment interest accrue during periods before the postponed due date, just as they don’t accrue (in the absence of a Presidentially declared emergency) before the normal due date. 

That’s the right result, and section 7508A(a)(2) is necessary to reach it.  Section 7508A(a)(1) simply says that the specified period is disregarded in determining whether acts—such as payment of the tax balance due—are timely.  It doesn’t explicitly modify the dates prescribed elsewhere in the Code for those acts.  The underpayment interest provisions, on the other hand, don’t focus on whether a payment was timely.  Section 6601(a) says underpayment interest starts as of the “last date prescribed for payment” and section 6601(b) defines that without reference to any postponement under section 7508A.

What about overpayment interest on a refund?  Section 7508A(c) applies the rules of section 7508(b).  That provision tells us that the rule disregarding the period specified by Treasury “shall not apply for purposes of determining the amount of interest on any overpayment of tax.”  Great!  We won’t owe interest for that designated period—April 15, 2020 to July 15, 2020 for the first emergency declaration above—on a balance due, but we can potentially be paid interest for that period on a refund.

Section 7508(b) also states that if a postponement with respect to a return applies “and such return is timely filed (determined after the application of [the postponement]), subsections (b)(3) and (3) of section 6611 shall not apply.”  That certainly sounds reasonable with respect to 6611(b)(3); it’s a de facto penalty of sorts for filing late, but if you filed your 2019 tax return by July 15, 2020, that should be treated as though you had paid by the original due date of April 15, 2020. 

Section 6611(e)(1) is a little bit different.  It’s not a de facto penalty for late filing; it’s essentially (1) a recognition that processing returns and issuing refunds takes some minimal amount of time and (2) possibly a small incentive for the IRS to do that expeditiously.  But when Treasury grants relief after a Presidentially declared emergency, any refund issued from a timely return will include interest, unless some other provision overrides it.  That’s what Bob Kamman’s Procedurally Taxing post pointed out and explained. 

And Now, the PMTA

It appears that the PMTA is correcting earlier analysis.  I haven’t been able to track down any earlier guidance, but the PMTA refers to “our previous conclusions” and those differed from the conclusions in the PMTA.  The conclusions in the PMTA seem correct.  Summarized:

  1. If the IRS determined a postponed due date under section 7508A, a return filed by the postponed due date means section 6611(b)(3) and (e) do not apply.  The return need only be valid under the Beard test to be timely and trigger the provisions of section 7508A; it doesn’t need to be processible.  This is what the “previous conclusions” got wrong, apparently thinking that whether a return was timely for purposes of section 7508A depended on whether it was processible.  But section 6611(g) defines “timely” only for purposes of section 6611(b)(3) and (e), not section 7508A.  It doesn’t matter whether the return was timely filed for purposes of 6611(b)(3) and (e); section 7508A negated them.  The default rules below don’t apply.  Interest will run from the date of the overpayment and there is no exception for a prompt refund.
  2. If the IRS determined a postponed due date under section 7508A, but a valid return is not filed by the postponed due date, it doesn’t meet one of the requirements for sections 7508A(c)/7508(b)—that the return is timely filed by the postponed due date—so the default rules below apply. 

Default rules, if there was no section 7508A relief from Treasury, or the taxpayer did not file a valid return by the postponed due date.  Sections 7508A(c)/7508(b) have no effect, so sections 6611(b)(3) and (e) are in effect and the return is “filed” only when it is processible pursuant to section 6611(g).  Thus:

  • If a processible return is filed by the original due date (or extended due date if applicable) it is not late under section 6611(b)(3), so interest runs from the date of the overpayment.  If it is filed after the original due date (or extended due date if applicable) it is late under section 6611(b)(3), so interest runs from the date the processible return was filed.
  • Whether or not the processible return is filed late, no interest is payable at all if the IRS makes a prompt refund under section 6611(e).  The 45-day grace period starts at the later of the original due date (even if the taxpayer had an extension) or the date the return is filed. 

Note that a return filed after the postponed due date (July 15, 2020) but on or before the extended due (October 15, 2020, if the taxpayer requested an extension) is not timely for purposes of section 7508A.  As a result, sections 6611(b)(3) and (e) are not disregarded.  But it is timely for purposes of 6611(b)(3).  Interest runs from the date of the overpayment rather than the date the return is filed, but the taxpayer may still lose all interest if the IRS issues the refund promptly.

Although the PMTA doesn’t mention it, the “refund back-off period” of section 6611(b)(1) will apply in all cases.  That rule is not dependent on whether a return is processible or filed timely, and it is not affected by a Presidentially declared emergency.

Two Final Thoughts

First, I think the PMTA is clearly right under the Code.  That the IRS originally reached the wrong conclusion, in part, is a testament to the complex interactions of the different provisions and the need for close, attentive reading.  I double-checked and triple-checked when I worked my way through the PMTA.  This was actually a relatively mild instance of a common problem with tax.  Code provisions are written in a very odd manner.  They’re not intended to be understandable by the general public; they’re written for experts and software companies, and sometimes difficult even for them.  I suspect that people who work through some of these complicated interpretations would fall into two groups: (a) those who really enjoy the challenging puzzle; and (b) those who experience “the pain upstairs that makes [their] eyeballs ache”.  My bet is that (b) includes not only the general public and most law school students but also a fair number of tax practitioners.  Which group are you in?

Second, that (relative) complexity leads to mistakes.  I assume that these interest computations have to be done by algorithm.  There are simply too many returns affected to have manual review and intervention for more than a small percentage.  An algorithm is feasible but will require re-programming every time there’s a section 7508A determination, with changes from year to year.  Even when the law is clear, there is a lot of opportunity for mistakes to creep in.  (We’re seen some of those recently in other contexts, e.g., stimulus payments and advanced Child Tax Credit.)  Whether by algorithm or manual intervention, particularly given the change from the original conclusions, there’s a good chance that some refunds were issued that are not consistent with the correct interpreation and may not have included enough interest.  Is the IRS proactively correcting such errors?  If the numbers are big enough, it might be worth re-calculating the interest you received—it’s easier than you may think—and filing a claim for additional interest if appropriate.

Disclosure of Collection Activity with Respect to Joint Returns

It’s the annual season for the reports from TIGTA mandated by Congress in 1998 and never unmandated.  So, each year TIGTA dutifully expends its resources on the problems Congress was concerned about in 1998, whether or not anyone is concerned about those issues today.  Some of the issues on which TIGTA writes its reports show that the IRS has persistent problems which, year after year, it cannot seem to fix.  One of the areas on which TIGTA reports annually and which the IRS cannot seem to fix is disclosing information on joint returns.  I wrote about this topic in 2020 and I wrote about this topic in 2018 when the annual TIGTA reports were released.  I probably sound like a broken record by writing on the topic this often, but the IRS needs to train its employees so they understand how the law works. 

Congress recognized that in certain situations the collection of a taxpayer’s liability is tied to payments potentially made by others.  In these situations, prior to the change in the law creating an exception to the disclosure laws and allowing the IRS to provide information to jointly liable parties, it was impossible to obtain information about payments from those jointly liable parties.  The TIGTA report shows that it can still be a practical impossibility to obtain this information, even though Congress opened the door allowing those jointly liable to learn of payments made, or not made, by the jointly liable person.

read more...

The most recent TIGTA report on IRS compliance with the law allowing taxpayers access to information on joint accounts suggests that taxpayers still struggle to obtain this information.  The IRS continues to experience difficulties when it sets up mirrored accounts.  Read the prior post from 2020 linked above if you are unfamiliar with mirrored accounts.  These accounts create difficulty for the IRS and taxpayers alike who do not understand when they exist and how they operate.  Here is the primary finding of this year’s report, which sounds much like the primary finding from each of the last five years:

We reviewed judgmental samples of 124 Accounts Management case histories and 20 Field Assistance case histories documented in the Account Management Services system related to joint filer collection information requests in the W&I Division.7 Based on our review, we determined that employees did not follow the joint return disclosure requirements in 26 (21 percent) of the 124 Accounts Management customer service representatives’ history files and three (15 percent) of the 20 Field Assistance individual taxpayer advisory specialists’ history files. The 29 cases were errors because employees did not provide the requested collection activity to the divorced or separated spouse as required by law. In most cases, employees incorrectly stated that they could not provide any collection activity on the other joint taxpayer, such as whether the other taxpayer made a payment or the current collection status, because the taxpayers were no longer married nor living together. As a result, these 29 taxpayers or their representatives were potentially burdened with additional delays in resolving their respective tax matter. This has been a recurring issue for the last five years and we have made recommendations for the respective IRS business units we have reviewed to update the IRM as well as to provide additional training to their employees. The IRS should continue to address this issue in its respective business unit IRMs that provide guidance to employees who may respond to taxpayer inquiries about a joint return matter.

We also observed that 10 (38 percent) of the 26 cases with disclosure errors in Accounts Management and all three of the cases with disclosure errors in Field Assistance had “mirrored accounts.” Mirroring a joint account sets up two accounts, one for each of the taxpayers. Establishing two separate accounts provides the IRS a means to administer and track collection activity unique to each of the taxpayers. Each taxpayer remains jointly liable for the entire debt; i.e., mirroring an account does not divide the liability in half. Because joint filer taxpayers remain jointly liable, the same collection information, when requested, on mirrored accounts should be disclosed to both taxpayers as would be disclosed on any other jointly filed return, except when the request is for unrelated personal information.

In addition to looking at case files, TIGTA interviewed IRS employees, with 8 out of 24 responding incorrectly regarding information that could be provided to a divorced spouse when the question involved a “regular” account and 15 out of 24 responding incorrectly when the question involved a mirrored account. 

Problems not only occurred when asking the employees questions regarding information that should be disclosed to the former spouse.  TIGTA asked the IRS employees questions that made it clear the employees would provide information they were not authorized to disclose, including:

providing information about the other spouse’s location, name change, or telephone number; information about the other spouse’s employment, income, or assets; the income level of the other spouse at which a currently not collectible module would be reactivated; or the bankruptcy chapter filed by the other spouse.18 When asked questions about a taxpayer who was divorced or separated, five employees (21 percent) of the 24 interviewed responded that they would disclose some of these prohibited items about the other spouse.

This year’s responses continue to point to disclosure of information regarding ex-spouses as a weak point for the IRS.  The law allowing some disclosure of information was enacted 25 years ago.  Almost all of the persons surveyed would have started work at the IRS after the law went into effect.  This is not a case of changes in the law creating confusion.  Yet, confusion continues to persist.

It’s not clear if the problem is that the employees need more training or better training.  The persistent existence of problems in this area which TIGTA identifies year after year should cause the IRS to change its method of training employees so that it can ensure compliance with the law both for the benefit of ex-spouses seeking information and employees trying to keep from violating disclosure laws.

Reliance on the Return Preparer, Too Good to Be True?

We welcome back guest blogger James Creech. Today James discusses a recent opinion by Judge Goeke examining a taxpayer’s reasonable cause defense. Reasonable cause is a frequent topic on PT, but this case involves a provision we rarely discuss: the ASED extension for failure to notify the government of certain foreign transfers. Christine

On June 28 the Tax Court released a 71 page decision in Kelly v Commissioner T.C Memo 2021-76. In Kelly, the IRS sought 6 years of income tax deficiencies and Section 6663 fraud penalties, with accuracy-related penalties in the alternative. For three of the years, the IRS needed an expanded statute of limitations to make its assessments. As an alternative argument as to why the expanded statute of limitations was appropriate for 2008 and 2009, even if there was not fraud, on the eve of trial the IRS raised the issue that the Taxpayer had not timely filed Forms 5471 for KY&C, a corporation he owned that was in part owned by another controlled entity based in the Cayman Islands.

read more...

How does the failure to file Form 5471 extend the ASED for Mr. Kelly’s personal income taxes? Section 6501(c)(8)(A) provides:

In the case of any information which is required to be reported to the Secretary pursuant to an election under section 1295(b) or under section 1298(f), 6038, 6038A, 6038B, 6038D, 6046, 6046A, or 6048, the time for assessment of any tax imposed by this title with respect to any tax return, event, or period to which such information relates shall not expire before the date which is 3 years after the date on which the Secretary is furnished the information required to be reported under such section.

Luckily for Mr. Kelly, the extension is much narrower if 6501(c)(8)(B) applies:

If the failure to furnish the information referred to in subparagraph (A) is due to reasonable cause and not willful neglect, subparagraph (A) shall apply only to the item or items related to such failure.

As Judge Goeke explains, “ if reasonable cause for the failure to file Forms 5471 exists, then under section 6501(c)(8)(B) only the adjustments related to KY&C would remain open under the statute of limitations.”

While the facts are complicated, and involve a company named after Dr. Evil’s company from the Austin Powers movies, the Tax Court held that the fraud assertions were not sustained. The Court then evaluated if the failure to file the 5471’s held the statute open or if, as the Taxpayer claimed, there existed reasonable cause for the failure to file because he had reasonably relied upon his CPA.  

The three-prong test to see if Taxpayer had a reliance defense is

  1. the adviser was a competent professional with sufficient expertise
  2. the taxpayer provided necessary and accurate information to the adviser, and
  3. the taxpayer relied in good faith on the adviser’s judgment.

Applying this test, the Court found that Mr. Kelly had reasonable cause based upon reliance on his tax return preparer. In a rather detailed analysis, the Court referenced the return preparer’s impartiality and lack of disciplinary record, contemporaneous emails where the Taxpayer disclosed the existence of the Caymen ownership to his preparer, and most interestingly the level of imputed knowledge the Taxpayer should have had about the return:

Respondent contends that it was not enough for Mr. Kelly to inform [his tax preparers] S&C that KY&C was a foreign entity, and he implies that Mr. Kelly should have advised Mr. Scott [of S&C] that Form 5471 was required.

The Court rejected this position.

The failure to file the Forms 5471 does not present an obvious tax obligation which was negligently omitted from information that a taxpayer provided to the return preparer. Mr. Kelly, through his staff, provided the necessary information to S&C, identified KY&C as a foreign corporation, and stated that he was unsure of the reporting requirements. Having done this, Mr. Kelly reasonably relied on S&C to prepare his returns properly. While it could be argued that S&C should have done more to ascertain Mr. Kelly’s filing obligations, it was reasonable for Mr. Kelly to rely on S&C do so. A taxpayer need not question the advice provided, obtain a second opinion, or monitor the advice received from the professional. Boyle, 469 U.S. at 251.

The Court’s citation of United States v. Boyle here, in a taxpayer victory, is an ironic twist. As Les discussed in a recent post,

Boyle essentially stands for the position that taxpayers have a nondelegable duty to be aware of tax deadlines. An agent’s incompetence or willful misconduct will not excuse the taxpayer from delinquency penalties.

Here, the Court seemed to impose a catchy “too good to be true” standard when it came to the Taxpayer’s knowledge:

At trial, Mr. Scott credibly described the reasons that his firm failed to prepare Form 5471 for KY&C. No facts suggest that the failure was the result of a conflict of interest or a “too good to be true” situation for either year. … We hold that Mr. Scott’s lack of prior experience with Form 5471 was not fatal to a finding of Mr. Kelly’s reasonable reliance on him or S&C.

The Court also recognized that when it came to filing Mr. Kelly’s tax return, informational returns do not impact the economics of the return. After all, most taxpayers only know how much money they made during the year, whether that be $35,000 or $3,500,000. If the only thing missing is an informational Form or a Statement, taxpayers, especially taxpayers who do not have a tax background, do not have the requisite knowledge to recognize such an error and insist the return preparer correct the return.

Time will tell if the “too good to be true” standard catches on as the knowledge requirement for a taxpayer’s review a return for correctness. As even simple returns grow more complex it would be useful to have a more definable standard that taxpayers can use to frame their standard of care when it comes reviewing returns and reliance on return preparers.

Refund Claims and the Specificity Requirement

A trio of cases have recently seen the IRS raise the specificity requirement in claiming a refund.  In opposition to the motions filed by the IRS, the taxpayers have raised waiver as a defense.  The results are interesting and instructive.  Of course, you don’t want to plan for having to defend against a failure to meet the specificity requirement but understanding how to defend against this argument can be useful.

When a taxpayer seeks a refund, the taxpayer must give the IRS a fair opportunity to administratively determine whether it should grant a refund.  If the claim filed by the taxpayer does not point the IRS in the right direction with enough information to allow to it understand the issue being raised by the taxpayer, then the taxpayer’s claim may fail the specificity test.

In Premier Tech v. United States, No. 2:20-cv-890-TS-CMR (D. Utah 2021); Intermountain Electronics v. United States, No. 2:20-cv-00501-JNP (D. Utah 2021); and Harper v. United States, 847 F. Appx. 408 (9th Cir. 2021) the courts grappled with the claims filed by the taxpayers.  The IRS objected that the claims lacked specificity and that the complaints did as well.

read more...

In Intermountain Electronics, the court held that the IRS had waived the specificity requirement by processing Intermountain’s refund claim, so the court had jurisdiction.  Even though it ruled favorably on the claim itself, the court found that the taxpayer’s complaint was not sufficient to state a claim under Rule 12(b)(6) under the Twombly/Iqbal plausibility pleading standard.  Carl Smith has written about the pleading issues raised by Twombly/Iqbal previously here and Patrick Thomas has written about it here.  The court allows the taxpayer to amend the complaint to make it more detailed and not just spout legal requirements from the research credit statute.

The court states without discussion that the refund claim filing requirement of 7422 is jurisdictional — a point currently pending in the Brown case before the Federal Circuit and a point raised in the cases of Walby v. United States, 2020 U.S. App. LEXIS 13711 (Apr. 29, 2020), which joined a panel of the Seventh Circuit in Gillespie v. United States, 670 F. App’x 393, 394–95 (7th Cir. 2016) questioning this position.  The issue and these cases have previously been discussed here and here.  Although it fails to mention Walby and Gillespie, the court discusses the Angelus Milling Co. v. Commissioner, 325 U.S. 293 (1945) opinion of the Supreme Court from 1945, where the Supreme Court held that the specificity requirements for a refund claim are waived if the IRS reviews and denies the claim on the merits because those requirements are only regulatory and therefore subject to waiver.  The district court in Intermountain notes that the taxpayer appears to have done all that was required under IRS forms by filling out the Form 6765 required for claims of research credits and attaching that form to its amended returns.  While the court is skeptical that the taxpayer had to attach a lot more unrequested information than the Form 6765 to create a valid claim, it does not decide that issue, but holds that the IRS waived the specificity requirements, even if the claims were initially insufficient. 

The court writes, as to the facts: 

As a result of the amended returns, the IRS initiated an examination to determine Intermountain’s eligibility for the credits. Over the course of an approximately five-year examination, the IRS requested and received substantial amounts of information from Intermountain. The IRS ultimately disallowed Intermountain’s claim for credits for both years, concluding both that much of the work performed by Intermountain employees did not constitute qualified research and that Intermountain failed to adequately substantiate its claims through the evidence it provided to the IRS. See ECF Nos. 26-3, 26-4, 26-6.  

The court notes that this case is virtually on all fours with a 9th Cir. opinion from earlier this year also involving research credit claims, Harper v. U.S., where the 9th Cir. held that due to a long audit of a Form 6765 attached to a refund claim, the IRS had waived the specificity requirements.  In both cases, the IRS received tons of documents from the taxpayers while the refund claim was audited, and the formal claim disallowance did not mention a specificity requirement defect, just that the taxpayer hadn’t shown entitlement to the claimed refund (which appears to be a denial on the merits).  (A concurring opinion in Harper would have held that the suit was valid because the taxpayer had made an informal claim, not that there had been a waiver.)

In Premier Tech provides a similar opinion decided a day earlier than Intermountain by another judge in the D. Utah, where it appears the judges must have coordinated their opinions. Premier Tech also involves a section 41 research credit refund claim.  Again, the taxpayer filled out a Form 6765 and attached it to the refund claim, and the IRS argues that alone was insufficient to meet the specificity requirement.  Unlike in Intermountain, though, while the IRS also audited the refund claim, the IRS never issued a notice of disallowance to Premier Tech, so suit was brought based on the rule regarding the passage of six months without such notice.  As a result, there is no argument in Premier Tech that the IRS waived the specificity requirement (the Angelus Milling argument).  Thus, the judge in Premier Tech is forced to decide the issue that was dodged in Intermountain — whether the refund claim met the specificity requirement.  The Premier Tech court holds that the specificity requirement was met and, unlike in Intermountain, that the complaint properly states a claim on which relief could be granted.  (In Intermountain, the judge held that the complaint did not state a claim, but allowed the taxpayer to amend the complaint to provide more detail.)  The reasoning in Premier Tech for finding the claim specific enough reads very similar to the reasoning expressed in Intermountain (but which was dicta there):  

Nevertheless, the United States argues that the amended return is not sufficient because Premier did not attach additional documents addressing every single element in 26 U.S.C. § 41, such as describing the research conducted, explaining how that research worked to develop a business component, detailing on whose wages and what supplies the money was spent, and proving the amount spent on research in the prior three tax years. But the United States offers no authority for imposing such a requirement. Form 6765 does not ask taxpayers to provide any of these details. If the IRS wants more information about the research tax credits, the IRS could require that information on Form 6765. It does not, and the IRS cannot now say its own forms are not sufficient to constitute claims for refunds. That would lead to absurd and patently unfair results for taxpayers. Furthermore, under the government’s position, no tax return claiming tax credits for increasing research activities could possibly constitute a claim for a refund. This is directly contradicted by 26 C.F.R. § 301.6402-3(a)(5).

In cases in which the taxpayer provides a wealth of detailed information to the IRS as part of the claims process and in cases in which the IRS rejects the refund claim on the merits, these cases show that the Department of Justice will struggle when it seeks to stop the litigation pointing to defects in the claim itself for failure of the claim to provide the IRS with enough information to adequately consider whether to grant the refund.  Even though DOJ may have correctly identified a poor refund claim, if the IRS acted in such a way that shows it understood what the taxpayer sought in requesting the refund, then courts will struggle to turn the taxpayer away on this basis.  The best practice is to carefully state the basis for the refund claim but where the facts show that the taxpayer provided enough information for the IRS to understand what the taxpayer wanted, the case will likely survive the types of challenges DOJ makes in these cases.

Tax Court Announces Return to In-Person Trials

On August 27, 2021, the Tax Court issued a press released stating that it anticipates returning to in-person proceedings starting with the winter trial sessions for 2022.  In the announcement, the Tax Court also states that it will continue to hold trials remotely where appropriate.

All of the Tax Court calendars scheduled for the Fall 2021 calendars are remote.  Until the August 27 announcement it was unknown when Tax Court judges would begin holding in-person trials again.  It was also unknown if the Tax Court would continue to offer remote proceedings as an option.  In issuing Administrative Order 2021-1 the Tax Court terminated Administrative Order 2020-2 which replaced in-person proceedings with remote proceedings and set a path forward for post-pandemic trials at the Tax Court.  By adopting a rule that allows for both in-person and remote proceedings, the Tax Court follows another Article 1 court, the Court of Veterans Appeals, and provides maximum flexibility for the parties and itself to conduct future proceedings.

read more...

The announcement creates some rules for requesting remote proceedings.  The rules contemplate that starting with the Winter 2022 Tax Court calendars the default setting for a trial will return to in-person proceedings.  Judges and trial clerks will start buying their plane tickets to the 73 cities where the court holds trials outside of D.C.  While in-person proceedings return to the normal Tax Court setting, the announcement provides that either party can request a remote proceeding by filing a motion up to 31 days prior to the first day scheduled for the trial session on which the court sets the case.

The announcement provides that a petitioner could request a remote proceeding at the time of filing the petition or any time between that date and the 31st day before the trial calendar start date on which their case appears.  Although not specifically stated, I assume respondent (the IRS) could request a remote proceeding at the time it answers the case or any time prior to the 31st date before the calendar.  The Court included with the announcement a sample motion that the parties could use to make the request.  It would seem that parties could start making the request now if they knew they wanted a remote proceeding.

Whether to grant the motion for a remote proceeding is at the judge’s discretion.  The announcement doesn’t say which judge would grant the motion if the party makes the motion before assignment of the case.  Typically, Tax Court cases remain unassigned until they appear on a calendar.  Orders issued in unassigned cases usually come from the Chief Judge’s chambers.  Once a case appears on a calendar, the judge assigned to that calendar takes control of the case.  If the party waits for the assignment of a calendar to request a remote proceeding, the judge in charge of that calendar will make the decision whether to put the case for a remote proceeding.

If a case is removed to a remote proceeding, the court indicates that it will seek to place the case on a stand-alone remote trial session “that is at least 5 months away.”  This sounds like a party seeking a remote proceeding after issuance of a calendar will receive a continuance of sorts.  This may be why the court requires the motion be made 31 days in advance.  Motions to continue made in the last 30 days before a trial session are presumed to be for purposes of delay.

The announcement does not make clear whether the trial judge assigned for the in-person calendar would be the same trial judge who would handle the remote calendar.  Depending on how that works, it is possible that filing the motion could not only allow the party more time to get ready for trial but allow the party to obtain a new judge to try the case.  Maybe that’s not such a big deal in the Tax Court as it might be in other courts.  By and large the trial judge in the Tax Court is not very outcome determinative but there could be times when a petitioner or the respondent might want a different judge than the one assigned to the calendar.  This could provide a path to that result if the judge in charge of the remote session is not the same as the judge at the in-person session. 

The announcement does make clear that the presiding judge for a calendar not only makes the decision regarding whether to allow the remote proceeding when the motion is filed after the issuance of the calendar but also can decide whether to handle the remote trial themselves at an agreed upon time or put the case over to the remote calendar.  In addition to the normal considerations regarding how much time a judge has invested in a case, I imagine the size of the trial location could also play a role in this decision.  Small venues that typically have only one trial calendar per year do not normally have a full complement of 100 cases on a calendar.  It may not make sense to create a virtual calendar for those locations that consists of only one or two cases and the presiding judge may default to retaining jurisdiction.

The remote trial sessions will all begin at 1:00 PM ET.  This allows for the sessions to begin at the same time across the country and is a departure from the normal 10:00 AM starting time in the time zone where an in-person trial session is held.  The parties in Hawaii might need to get up a little early under this system but the court has adopted this rule to reduce confusion.

The announcement states that the public will have access to remote proceedings.  I expect the public would have the same type access it has to remote proceedings today.  The court also indicates that it will continue to coordinate with low income tax clinics and local bar sponsored pro bono programs to provide assistance for petitioners who end up on a remote calendar.  I expect the court will do that it much the same way it has done with remote calendars over the past year.

The court’s announcement will undoubtedly get Chief Counsel’s office thinking about the policies it wants to adopt regarding remote proceedings.  When will it make a request for remote proceedings?  When will it oppose remote proceedings?  Sometimes for workload reasons it shifts cases from one office to another requiring it to send attorneys from one city to another at some expense.  Will it request remote proceedings in those situations as a cost savings measure?  Will it request remote proceedings even if the petitioner wants to conduct the trial in person?  Will it object to motions made 31 days in advance of trial as a tactic to obtain a continuance?  It will have lots to consider.

The announcement moves the Tax Court into a new phase.  It now becomes a court that does not need to travel quite as much and can schedule cases remotely.  Will the trial judge with only one case left on the North Dakota calendar exert some pressure on the parties to try the case remotely and save the judge two days of travel time?  Will there be other factors that influence the court in granting or denying requests for remote proceeding?  So far, the switch for remote or in person has been an on/off switch.  The court was either in person or remote but not optional.  It will be interesting to see how the parties and the court adapt to the optional world and where the Tax Court ends up 10 years from now.

District Court Finds That IRS Failed to Adequately Notify Taxpayer Before it Contacted Third Party

In US v Vaught a federal district court in Idaho declined to enforce a third party summons due to the IRS’s failure to notify a taxpayer of its intent to contact a third party during the course of the audit.  

In this post I will discuss Vaught and provide some context for the summons dispute.

read more...

In 2015 IRS began an investigation of Stanley Crow into penalties associated with his promotion of installment sale transactions. The IRS suspected the transactions were tax shelters subject to registration and disclosure requirements.  At the start of the examination it sent Crow a Publication 1, Your Rights as a Taxpayer. That publication gives generic information about the audit process, and also informs taxpayers that the IRS may contact third parties during the course of an examination.

Prior to the Taxpayer First Act, under Section 7602(c)(1), enacted as part of the Restructuring and Reform Act of 1998, IRS was required to provide a taxpayer with “reasonable notice in advance” before it contacted financial institutions, employers or other third parties.  Pre-TFA there was litigation as to what constituted reasonable advance notice. In many of the cases, the IRS argued that sending Publication 1, at the start of an exam sufficed for these purposes. 

In JB v United States, which I discussed in Ninth Circuit Rejects IRS’s Approach to Notifying Taxpayers of Third Party Contacts, the Ninth Circuit held that sending the Publication 1 was insufficient notice, though it did not categorically hold Publication 1 can never constitute “reasonable notice in advance.” The court was skeptical though stating that it was “doubtful that Publication 1 alone will ever suffice to provide reasonable notice in advance to the taxpayer, as the statute requires.”

Shortly after JB v US, TFA did away with the squishy reasonable standard and requires the IRS to provide notice to the taxpayer at least 45 days before the beginning of the period of third-party contact, which may not extend longer than one year.  In Keith’s post on the major tax procedural developments of 2019, he discusses the TFA change, and the earlier TAS legislative recommendations that highlighted the problems with the “reasonable advance notice” standard under pre TFA law.

Back to Vaught. The case involves an exam and third-party summonses that were issued in 2018 in connection with the IRS’s examination of Crow and his suspected shelter promotions activities. Two of the summonses were served on Steve Vaught and Alpha Lending, LLC, where Vaught was a key executive. Alpha had a business relationship with Crow, having served as a lender or escrow agent in the installment sales transactions Crow and his company promoted. Vaught/Alpha did not appear or produce the IRS’s requested records. The government filed a petition to enforce the summons, and Crow intervened in the case and filed a motion to quash based on an alleged violation of the advance notice requirements.

As I mention above, the summonses were issued before TFA, so the case involves the old reasonable advance notice standard. As the district court explained, in JB the Ninth Circuit added some meat to the reasonable notice standard:

[T]he Ninth Circuit held the phrase “reasonable notice in advance” means “notice reasonably calculated, under all the relevant circumstances, to apprise interested parties of the possibility that the IRS may contact third parties, and that affords interested parties a meaningful opportunity to resolve issues and volunteer information before third-party contacts are made.” 916 F.3d at 1164 (citing Flowers , 547 U.S. at 226). In so holding, the Ninth Circuit highlighted the purpose of the notice requirement is to protect the taxpayer’s reputational interest by giving the taxpayer an “opportunity to resolve issues and volunteer information before the IRS seeks information from third parties, which would be unnecessary if the relevant information is provided by the taxpayer himself.”

The JB standard requires district courts to examine the totality of the circumstances, “balancing of the interests of the State’ in administering an effective auditing system against the ‘individual interest’ in receiving notice of the potential third-party contact and an opportunity to respond.”

Crow challenged both related aspects of the notice requirement:  that the IRS did not provide pre-contact notice and also that the IRS did not provide a reasonable meaningful opportunity to resolve issues and volunteer information before third-party contacts were made.

The government argued it did provide sufficient notice, pointing to the Pub 1 it sent at the start of the exam in November of 2015 as well as supposed oral communications between IRS agents and Crow in December of 2015. At that meeting IRS revenue agents purportedly said that they said that the IRS may contact third parties during the course of the examination of Crow.

The district court found that the IRS failed to satisfy the reasonable advance notice standard, emphasizing the time between the generic notice and the actual contact IRS made with the third party:

Here, the IRS issued the Vaught Summonses in January of 2018, twenty-six months after it sent Crow Publication 1 on November 17, 2015. Similarly, in J.B., two years elapsed between the date the IRS sent Publication 1 to the taxpayers and the date the IRS sought records from a third party. As in J.B , this Court cannot find the IRS satisfied its “administrative duty” of giving Crow a meaningful opportunity to provide relevant documents involving the Alpha Companies by generally informing Crow, over two years before, that it may “talk with other persons” in the course of its investigation

As to the purported oral communication from IRS agents failing to tip the scales toward reasonable notice, the court focused on the absence of specific information concerning the nature of the needed information in the agent’s affidavit:

Neither the Government’s brief, nor Allred’s affidavit, offer any details regarding what London and Allred said about potential third-party contacts on December 16, 2015….[T]he Government does not provide any specific information regarding how the IRS purportedly notified Crow of potential third-party contacts on December 16, 2015. For instance, what did London or Allred say about third-party contacts on this date? Did they reference any specific third parties or types of businesses they may contact if Crow did not provide information himself? Did they give any hint that Crow should produce documents involving the escrow companies or lenders that SCCC used in its installment sales transactions?

Taken together the court concluded that the IRS failed to provide reasonable advance notice. As the opinion notes, IRS could have done more to act consistently with Crow’s legitimate privacy interests. To that end, the opinion discusses (at Crow’s suggestion) steps the IRS could have taken, including renewing its request for information from Crow closer in time to the contact (which was over two years from both the generic notice IRS provided and revenue agent conversation) and specifying that it would contact third parties if Crow did not provide the information it wanted. As a final measure, the court noted that FOIA-obtained IRS case notes erroneously concluded that Alpha and Vaught were not third parties for purposes of the notice requirements. This suggested perhaps that the IRS did not provide additional notice because of the mistaken belief that the contacts did not trigger notice requirements.

Conclusion

The court ultimately concluded that the violation of the advance notice requirement meant that the IRS failed to satisfy the fourth Powell summons requirement, that the IRS follow all administrative requirements in the issuance of the summons.  While Vaught involves pre TFA law, it is an important opinion in at least two respects. First, there may be summons enforcement cases still percolating under the pre TFA notice rules. Second, and perhaps more important, the opinion reflects a perspective that emphasizes that taxpayers have legitimate privacy and reputational interests. When there is the scent of shelters or allegedly improper taxpayer conduct at times IRS may fail to adequately weigh or even consider the interests of taxpayers. Vaught should serve as a reminder that privacy and reputational interests are at stake even when there is a taxpayer suspected of engaging or promoting aggressive transactions.

What are the next steps here? I have not dug into the filings but I assume that if the IRS has not received the information it could reissue the summonses, ensuring that it complies with post TFA notice requirements. Challenges to summons enforcement typically toll the SOL on assessment, so perhaps the taxpayer victory is short lived.

Setting Aside a Settlement

Several years ago, a settlement reached by the Villanova clinic with an Appeals Officer was set aside when the AO’s manager would not accept the settlement recommendation.  Every settlement with an AO or a Chief Counsel docket attorney must receive approval from their manager.  Usually, the AO or the Chief Counsel attorney makes explicit statements about the limitations of their authority.  However, when time permits, these individuals also usually discuss a proposed settlement with their manager so that the formal submission of the settlement does not result in a surprise to the taxpayer and the employee.

Following the unpleasant surprise created by the rejection of a settlement that resulted from months of discussion with the AO, the clinic researched when a settlement could bind the government.  The research did not lead us to the conclusion that we could bind the government in this instance, despite the fact that the AO had led us on for some time.  I wrote a three part blog post series, linked here, here, and here, about our case and an article on the binding nature of settlements in general.  Les wrote a subsequent post involving a different case that also raised the binding nature of a settlement.

The recent 9th Circuit decision in Dollarhide v. Commissioner, 18-71722 (9th Cir. 2021) raises this issue in the context of a stipulation of settled issues.  It is a case worth noting.

read more...

In the Tax Court it regularly occurs that the parties reach a settlement at the last minute.  Certainly the Tax Court is not unique in having parties reach a last minute settlement.  What does create more tension in Tax Court settlements is the circuit riding nature of the court.  It only comes to one of the 74 cities in which it sits every few months or every six months or once a year.  Continuing a case to allow the parties to wrap up a settlement could throw the case back on the general docket.  It could also allow the unscrupulous petitioner or representative to appear to settle only to back out when the court leaves town, not to return for quite some time.

To prevent parties from backing out and to cause them to show that they really did have a settlement, the court regularly requests parties to last minute settlements to file with the court a stipulation of settled issues.  Aside from the fact that the Tax Court travels, settlements can take some time because the issues need to be turned into tax computations.  By filing with the court a stipulation of settled issues, the parties essentially leave open the possibility of a Rule 155 argument on the consequence of the computation of the issues but otherwise commit themselves to a settlement.  The trial judge on the calendar typically retains jurisdiction over the case.  In the routine case in which a stipulation of settled issues is filed, the computation occurs within a relatively short time and a decision document is submitted to the court shortly thereafter.  Judges typically give about 30 days from the time of filing the stipulation of settled issues for this to occur, though continuances sometimes occur when something causes a delay.

The Dollarhides entered into a stipulation of settled issues in their Tax Court case.  After doing so, they declined to sign a decision document.  This sounds similar to what happened in the Dorchester Industries case, the seminal Tax Court case regarding the binding nature of certain agreements.  When the Dollarhides declined to sign the decision document, the IRS moved to enforce the settlement agreement and the Tax Court agreed with the IRS.

The Dollarhides appealed the enforcement and the 9th Circuit reviewed the decision for abuse of discretion.  The 9th Circuit found:

The Stipulation of Settled Issues, on which the Tax Court’s order granting the IRS’s motion for entry of decision is premised, says nothing about the key issue in this case: whether the Dollarhides were barred by the statute of limitations set out in 26 U.S.C. § 6511(b)(2) from receiving a refund for tax year 2006. The Dollarhides contested application of the statute of limitations bar in the Tax Court and continue to do so on appeal.

The Commissioner now concedes that there was no conclusive settlement agreement between the parties with respect to whether the Dollarhides were due a refund for tax year 2006. Because there was no settlement agreement between the parties with respect to this disputed issue, it was an abuse of discretion for the Tax Court to grant the Commissioner’s motion and enter a judgment enforcing the parties’ purported settlement of this issue. See Bail Bonds, 820 F.2d at 1547. We thus vacate and remand on this ground and do not reach the Dollarhides’ remaining arguments on appeal.

This is a somewhat shocking result that both the IRS and the Tax Court would miss the fact that a major piece of the settlement of the case was missing.  Since the IRS conceded that this piece was missing, we do not get an opinion from the 9th Circuit that parses the language of the settlement. 

While it’s possible to give the Tax Court a stipulation of settled issues that does not settle all of the issues in a case, when the parties do that they usually make it clear that the stipulation is in partial settlement of the case and does not resolve all issues.  I would have expected the parties to do that here and cannot say why the stipulation would have left the IRS and the Court with the impression that everything was settled.

Certainly, one lesson here is to make clear in submitting the stipulation of settled issues what issues, if any, are reserved by the parties.  Here, the Dollarhides avoid having the settlement foreclose them from making the refund argument but on appeal they faced the daunting task of overcoming an abuse of discretion standard.  Another lesson is that the possibility exists to challenge a decision based on a stipulation of settled issues.  In most cases taxpayers will lose, but the Dollarhide case shows that success is possible.

Is Economic Hardship the Antidote for Knowledge in an Innocent Spouse Case?

A pair of innocent spouse cases just came out, one granting relief, Grady v. Commissioner, T.C. Summ. Op. 2021-29, and one denying relief, Rogers v. Commissioner, No. 20-2789 (7th Cir. 2021).  Neither case reaches a surprising result but the cases do continue trends.  In this post I hope to not only provide some background on these two cases but to also explore the trends that have emerged in innocent spouse cases.

read more...

In the Grady case, a case tried under the small tax case procedures, the Tax Court details a litany of issues that the non-requesting spouse (the ex-husband) caused during the marriage.  In the end, the Tax Court finds that the petitioner knew that the tax liability was not being paid so the knowledge factor is negative but essentially all other factors were positive, including economic hardship.  The Court states that:

While her knowledge when she signed the 2007, 2009, 2010, and 2011 joint Federal income tax returns that the tax due would not be paid weighs against her entitlement to section 6015(f) relief, generally knowledge is only one of the factors and knowledge alone is not determinative of the Court’s decision. See Minton v. Commissioner, T.C. Memo. 2018-15 (granting relief despite the taxpayer’s admitting to knowledge of a balance owed); Demeter v. Commissioner, T.C. Memo. 2014-238 (granting relief despite finding that the taxpayer knew or had reason to know that her ex-husband would have difficulty paying the tax liabilities). Therefore, in considering Ms. Gans’ entitlement to relief under section 6015(f), her knowledge is only one factor among many to be taken into account. As the Court has noted, no one factor, in and of itself, is determinative. See Stolkin v. Commissioner, T.C. Memo. 2008-211; Beatty v. Commissioner, T.C. Memo. 2007-167; Banderas v. Commissioner, T.C. Memo. 2007-129.

As regular readers of this blog know, we believe, and have discussed here and here, that the Tax Court treats knowledge as a super factor in many cases.  Knowledge alone did cause Mr. Jacobsen and Ms. Sleeth to lose their innocent spouse cases despite four (Jacobsen) and three (Sleeth) positive factors. The fact that, even in this case where knowledge is the only negative factor, the Court spends a paragraph explaining that knowledge alone is not determinative, provides insight into the power of the knowledge factor.

The Rogers case continues the unbroken string of losses for taxpayers appealing IRC 6015 cases.  Since the change in the law in 1998 placing the innocent spouse provisions in IRC 6015, no taxpayer has won an appeal from an adverse Tax Court decision.

In Rogers, the 7th Circuit affirms the Tax Court’s holding that the wife of a shelter promoter isn’t entitled to innocent spouse relief.  The court noted that this was not the first visit to the 7th Circuit by one or both members of the marital unit:

Married since 1967, John and Frances Rogers filed joint federal income tax returns for many years. They underreported their tax obligations many times over, and the misreporting was the product of a fraudulent tax scheme designed by John, a Harvard‐trained tax attorney. The fraud did not elude the Internal Revenue Service, though, and the many subsequent collection and enforcement proceedings in the U.S. Tax Court have not gone well for the Rogerses. Our court has affirmed the Tax Court’s rulings every time.

Before us now is another appeal by Frances challenging two Tax Court decisions denying her requests for what the Tax Code calls innocent spouse relief. Our review of the record shows that the Tax Court took considerable care assessing Frances’s pleas for relief, in the end denying them largely on the basis that she was aware of too many facts and too many warning signs during the relevant tax years to escape financial responsibility for the clear fraud perpetrated on the U.S. Treasury. While the tragedy of what Frances has endured over the years is in no way lost on us, we are left to affirm, for the Tax Court got it right.

In one respect, the 7th Cir. disagrees with the Tax Court as to a factor — the substantial benefit factor does not weigh against relief in this case.  But, interestingly, the 7th Cir. never cites or discusses the Rev. Proc. factors.  It limits its discussion to how the Rogers facts compare to a prior 7th Cir. opinion from 1996, Reser, which, of course, involved 6013(e).  The most the 7th Cir. will do is cite a reg. under 6015 concerning significant benefit for purposes of (b), 1.6015-2, that actually derives from language in the Committee reports from 1971 for enacting 6013(e).  The committee reports can be found at H.R. Rep. No. 91-1734, at 2 (1970), and S. Rep. No. 91-1537, at 2 (1970), 1971-1 C.B. 608. The 7th Cir. focuses entirely on the knowledge issue (both for purposes of (b) and (f) relief) as grounds for denying relief.  If there were no other factors negative for relief, though some positive or neutral factors, this would make Rogers a case similar to the Jacobsen case decided by the 7th Cir. two years ago.

Interestingly, the Grady case presented only one negative factor, knowledge, and multiple positive factors, but the Tax Court granted relief.  That’s the exact same situation as in Jacobsen, but the case leads to a different result.  Carl Smith has done a fair amount of research and thinking on this issue.  He concludes that the reason why Grady won while Jacobsen didn’t is that, although Jacobsen had four positive factors for relief, he did not put in the evidence to establish financial hardship, which Grady did.  Research of innocent spouse cases shows that proving financial hardship serves as the only way to guarantee that the taxpayer wins an innocent spouse case where knowledge is a negative factor.  Lack of significant benefit, marital status, and compliance with return filing obligations are not enough to outweigh knowledge in some Tax Court opinions.  Note that, in Sleeth (from the 11th Cir. this year), Ms. Sleeth was also said not to have proved financial hardship, and her case also involved only one negative factor (knowledge), and three positive factors (the ones in the prior sentence). Jacobsen’s positive factors included those from Sleeth, as well as an additional fourth positive factor — for his bad health.

As mentioned above, the Rogers 7th Cir. opinion did not cite or discuss the Rev. Proc. that was applicable.  That seems significant, since the Tax Court almost always discusses each of the Rev. Proc. factors.  In 2011, Carl Smith wrote a Special Report for Tax Notes entitled “Innocent Spouse:  Let’s Bury that Inequitable Revenue Procedure“.  In the article, he called for the courts to return to deciding the equitable factor under common law — using opinions involving 6013(e) and 6015, not the Rev. Proc. factors.  While using the factors of the Rev. Proc. seems appropriate for the IRS in administratively evaluating cases, it seems less appropriate for courts which need not be bound by the IRS’ views of appropriate equitable factors.

In some ways the courts, particularly the Tax Court, seem to apply their own thinking, yet cloak the decisions in the factors of the Rev. Proc.  While the Rev. Proc. may say that knowledge is no longer a super factor and while the Tax Court may say it is applying the Rev. Proc., the outcomes suggest that the court has its own equitable barometer which still places significant weight on knowledge.  If the Tax Court weighs knowledge more heavily, then taxpayers must look for something to countervail knowledge or potentially lose even where they have many positive factors. In cases where knowledge is the only negative factor and there are three or more positive factors (one of which is lack of significant benefit), the taxpayer usually wins, but the taxpayer always wins if one of the positive factors is also financial hardship.  You can find the list of cases where knowledge was the only negative factor in the Jacobsen brief filed by the Harvard Tax Clinic in the appeal to the 7th Circuit.