Last Known Address for Incarcerated Persons

On October 17, 2014, the Procedure and Administration Division of the Office of Chief Counsel, IRS, sent PMTA-2014-19 (Program Management Technical Advice) to Denise Rosenberg a Senior Program Analyst in SB/SE Collection Policy.  The advice addresses the question of “whether the Federal Correctional Institute address of an incarcerated taxpayer, presumably maintained in the Bureau of Prisons’ website and included on Form 13308, Criminal Investigation Closing Report, should be used as the last known address for purposes of mailing notices required under the Code to the taxpayer’s last known address.”  The advice concludes that “unless the taxpayer provides a clear and concise notification that the place of incarceration should be used as the taxpayer’s last known address, the Service, generally, may use the address on the most recently filed return as the taxpayer’s last known address.”  It does go on to issue the cautionary statement that “where the Service has specific knowledge of the taxpayer’s incarceration and there is a defect in the mailing to the last known address, the Service will be expected to use reasonable care and diligence in ascertaining the taxpayer’s correct address.”

I will talk below about some of the reasoning in the PMTA. The issue has administrative significance to the IRS because the last known address issue bears on the validity of a number of items the IRS mails to taxpayers.  The issue has enough importance that in the book Effectively Representing Your Client before the IRS, we devote an entire chapter to the topic of last known address.  The PMTA does not mention a TIGTA report that provided an in depth look at the issue of the addresses of prisoners and the access of the IRS to those address as well as the use by the IRS of those addresses because of the amount of refund fraud perpetrated by prisoners over the past decade.  Despite the fact the IRS has the addresses of the prisoners and despite the fact that we know that 99.9% of individuals heading into prison do not include as part of their pre-incarceration to do list a filing with the IRS of Form 8822 the IRS takes the position in the PMTA that sending a notice to an incarcerated person is subject to a facts and circumstances test.  The rule should change by statute if not administratively changed.


A New Approach

In the United States .75% of the population is incarcerated at any given moment.  The procedure set out in this PMTA virtually assures that most of the people in that segment of the population will have a notice sent to an address that no longer serves as their primary address.  It unnecessarily calls into question the validity of the notice sent when it would be possible based on data in the possession of the IRS to send a duplicate notice to the location of incarceration in many of those cases.  Adopting a different approach targeted at reaching the population of individuals in prison would not increase the cost to provide notice in a significant way, would afford these individuals due process of receiving notice of proposed action and appears possible given current information reporting to the IRS and its technology.  Rather than relying on rules concerning notifying the IRS of a change of address that almost no one follows even when not facing incarceration, the IRS should use the information at its disposal to make a meaningful effort at notification of the incarcerated population.

Perhaps with study of the situation we would find that those in prison do not need to receive correspondence by certified mail because that impedes the receipt of mail. Perhaps we learn that those in prison need more time to take action and should receive 150 days to file a Tax Court petition instead of 90.  Prisoners, who by and large fall within the low income population served by clinics under IRC 7526, deserve the opportunity to receive notices in time to take appropriate action just as we all deserve appropriate notice.  Sending a letter to the address of their last return when it is known that they are incarcerated and where they are incarcerated seems an inappropriate and unnecessary response to the situation.  While sending two notices adds to the cost, it does so in a way that affords meaningful opportunity for redress if the individual feels the need to take action and reduces later claims that no notice was provided.

TIGTA report

The PMTA does not mention a TIGTA report  issued approximately one month before the PMTA.  The TIGRA report focuses on refund fraud by prisoners but contains interesting data about the information sharing agreement between the IRS and prison authorizes having a direct bearing on the provision of notice to prisoners.  Aside from the notice of address issue, the report is interesting in itself to anyone wanting to know more about the incidence of refund fraud generated by prisoners and the efforts to combat that fraud.  It is a piece of the problem I talked about in my Senate testimony urging Congress to flip the filing season to have the IRS wait to pay refunds until it has the matching data.  If you wanted some support for that idea, the TIGTA report provides some.

In 2008 Congress passed The Inmate Tax Fraud Prevention Act.  This gave the Secretary of Treasury temporary authority to disclose to the head of the Federal Bureau of Prisons tax return information for individuals who filed or may have filed fraudulent returns while incarcerated in Federal prisons.  The act created a time limit of December 31, 2011 on this exception to the disclosure statute.  The act required an annual report to Congress and the IRS filed the first report for calendar year 2009   The report is found in Appendix VI of the TIGTA report linked here.  The Homebuyer Assistance and Improvement Act of 2010 expanded the authority to share prisoner tax return information to the State Departments of Correction since refund fraud does not solely occur among the federally incarcerated.  The United States – Korea Free Trade Agreement Implementation Act requires the Federal Bureau of Prisons and State Departments of Corrections to “provide the IRS with an electronic list of all the prisoners incarcerated within their prison system for an part of the prior two calendar years or the current calendar year through August 31.”  This report requires annual updates.  So, the IRS gets a list of everyone in prison in the United States.  The American Taxpayer Relief Act of 2012 expanded Treasury’s authority so share prisoner tax return information concerning false returns to federal and state prison authorizes and made the law permitting sharing permanent.

While the PMTA does not go into all of the information sharing Congress has created in the past decade, it is clear that the IRS now receives considerable information about individuals incarcerated in the United States. While this information comes to the IRS for the purpose of combating refund fraud, it seems inappropriate to ignore it when it comes time to provide notice to the prisoner.


Despite all of the information going back and forth between the federal and state prison systems and the IRS, the PMTA focuses only on the knowledge of the special agent regarding the specific individual. It concludes that “unless the taxpayer provides clear and concise notification that the place of incarceration should be used as the taxpayer’s last known address, the Service, generally, may use the address on the most recently filed return as the taxpayer’s last known address.”  It then talks about a facts and circumstances test that might override this result “where the Service has specific knowledge of the taxpayer’s incarceration.”  Yet, from the legislation recounted in the TIGTA report on refund fraud, it seems that the IRS has specific information about every taxpayer’s incarceration if the incarceration takes place in the United States.

The PMTA should discuss the relationship between the information that now comes to the IRS regarding prisoners and the obligation of the IRS to notify individuals whose addresses it has obtained because it sought them. The IRS is not merely a passive receiver of the addresses of prisoners but affirmatively sought them in order to combat refund fraud.  It is great that the IRS and the prison authorities work together to combat this scourge on tax administration; however, the IRS needs to at least acknowledge it has this information and discuss why it does or does not provide a meaningful basis for notice to the taxpayer.


In our efforts to combat refund fraud that in the electronic filing era has run rampant among prisoners, Congress has passed a series of laws designed to provide the IRS with detailed information about prisoners. Despite the system of information now provided to the IRS, Chief Counsel’s fails to even discuss this influx of information on prisoners in its PMTA concerning last known address issues for prisoners.  It should acknowledge that the IRS now operate in a new period of relations between the itself and prisoners and not overlook the data coming into the IRS allowing it to know who is incarcerated.  Rather than continue to rely on antiquated provisions of notice developed prior to the close working relationship between the IRS and the prison authorities, the IRS should acknowledge that it now receives information about incarcerated individuals.  Either the information about prisoners comes to the IRS in a usable format allowing the IRS to use that information to provide notice to prisoners when sending out documents such as the notice of deficiency or it should spell out how this information does not allow it to do so.

If the information coming from prison officials does not allow it to identify who is in prison and where they reside, perhaps the IRS needs to go back to the prison officials for more or better information. One can imagine after reading the PMTA and the TIGTA report that on the same day, the IRS will mail a notice to prison officials notifying those officials that John Doe incarcerated in the federal penitentiary in Lewisburg, PA, filed 12 fraudulent refund claims while on the same day it mails a notice of deficiency to the same John Doe in Richmond, VA, the place from which he mailed his last tax return five years ago prior to reporting to the penitentiary.  If the IRS now receives adequate information to enable it to provide notice to them in prison, it should acknowledge it has this information and make use of it in sending notices to this segment of the population.  If the information is insufficient to provide notice to incarcerated individuals, it should explain that in the PMTA and not ignore it.

While the issue here concerns prisoners and maybe we do not care what happens to prisoners, the IRS will get data dumps on other segments of the population. If it has the data for one purpose, it should use that data to provide meaningful notice to individuals and not rely on a change of address form that almost no one uses.

SCOTUS Wong Ruling Holding FTCA Time Periods Subject to Equitable Tolling Probably Affects IRC Time Periods

We are not always quick to analyze new developments but today frequent guest blogger Carl Smith brings us a discussion of yesterday’s decision by the Supreme Court in a non-tax case that could have significant implications in the tax world. Equitable tolling has been the subject of frequent discussion here and moves front and center following the Court’s decision. Keith

In two prior posts last year, from May 1, 2014 and July 2, 2014, I discussed two cases out of the Ninth Circuit – Wong v. Beebe, 732 F.3d 1030 (2013) (en banc), and June v. United States, 50 Fed. Appx. 505 (2013) – that held that two different time periods in the Federal Tort Claims Act (FTCA) (one for filing an administrative claim and the other for filing a district court suit) were not jurisdictional and were subject to equitable tolling.  The Supreme Court granted certiorari to review those holdings.  I pointed out that a Supreme Court ruling that the FTCA time limit for filing in district court was subject to equitable tolling would likely suggest that § 6532(a)’s 2-year-after-claim-disallownce time period to bring a tax refund suit was also subject to equitable tolling, notwithstanding holdings by a number of courts of appeal (though not all) that the 2-year period was not subject to equitable tolling.  On April 22, in a combined 5-4 opinion under the name of United States v. Wong, the Supreme Court has just affirmed the Ninth Circuit as to both FTCA time periods.  Below is a summary of the holding and a bit of a repeat of why I think this holding now renders § 6532(a)’s time period subject to equitable tolling.

In a post earlier this year, I also noted that the Ninth Circuit in Volpicelli v. United States, 777 F.3d 1042 (2015), had held (contrary to several other Circuits) that the 9-month time period at § 6532(c) in which to file a wrongful levy suit was not jurisdictional and was subject to equitable tolling.  Because the DOJ’s request for en banc rehearing in Volpicelli was turned down on April 8, we are early in the 90-day period in which the Solicitor General must consider whether to ask for certiorari of the Volpicelli opinion.  In my view, a cert. petition in Volpicelli would now be futile:  Even if the Supreme Court granted cert. (a big if), between Wong and prior recent Supreme Court case law, the government’s chances of overturning Volpicelli are approaching zero.  Indeed, I would encourage counsel representing taxpayers in either wrongful levy or tax refund suits who have arguments for equitably tolling those statutes of limitations to just assume that any Circuit court opinions previously holding tolling of those periods impermissible are no longer good law and will be overruled, if asked.


Prior to Irwin v. Dept. of Veterans Affairs, 498 U.S. 89 (1990), the Supreme Court had usually (though not always) held that time periods in the United States Code that applied to the federal government were “jurisdictional” and thus not subject to equitable tolling.  In Irwin, the Supreme Court — dissatisfied with its ad hoc rulings on different time periods coming to different results — laid down a new rule for interpreting the waiver of sovereign immunity:  Henceforth, Congress was rebuttably presumed to have intended its waiver of sovereign immunity to allow for equitable tolling of such time periods to the same extent that the time periods could be tolled in suits between private parties.  Irwin involved an employment discrimination suit brought against the federal government — a kind of suit that could also be brought against a private employer. Later Supreme Court cases extended Irwin‘s presumption to suits that could only be brought against the federal government.  Holland v. Florida, 560 U.S. 631 (2010) (time period in which to file for federal habeas relief in death penalty cases is subject to tolling under the Irwinpresumption).  Later cases also clarified that Irwin‘s presumption did not apply to “jurisdictional” time periods, but, at most, only to non-jurisdictional statutes of limitations.  Sebelius v. Auburn Regional Medical Center, 133 S. Ct. 817 (2013) (time period in which to file a dispute over Medicare reimbursement was not jurisdictional, but still was a non-jurisdictional statute of limitations as to which the Irwin presumption in favor of equitable tolling either did not apply or was rebutted).

The jurisdictional grant to district courts to entertain suits under the FTCA is at 28 U.S.C. § 1346(b)(1) (“district courts . . . shall have exclusive jurisdiction” over tort claims against the United States).  The time periods at issue in Wong were both located within 28 U.S.C. § 2401(b), which provides that a tort claim against the United States “shall be forever barred” unless it is presented to the “appropriate Federal agency within two years after such claim accrues” and then brought to federal court “within six months” after the agency acts on the claim. (Like IRC § 6532(a), there is also a provision allowing an FTCA suit to be brought after 6 months if the agency has not ruled on the administrative claim.)  Notably, unlike the time periods in which to file an administrative tax refund claim at § 6511, the time periods in § 2401(b) have no exceptions.  In United States v. Brockamp, 519 U.S. 347 (1997), the Supreme Court had held that the Irwin presumption in favor of tolling was rebutted, in part, because of Congress already including many exceptions into the section, leaving the Court to infer that it was not permitted to add an unwritten judicial equitable tolling exception.  (Interestingly, Brockamp is not cited by either the majority or dissent in Wong.)

The government decided to argue the Wong case essentially by not contending that, if the FTCA time periods were not jurisdictional, the periods were still not subject to tolling.  In effect, the government, as a practical matter, conceded the Irwin presumption would lead to tolling if the time periods were not jurisdictional.  Why did the government do this?  Well, there were no factors in these simple FTCA time periods that have ever led the Court to find the Irwin presumption rebutted.  So, it would have been a pointless argument to make.

In Wong, the Court noted that in its recent cases, it had held that time periods are claims processing rules that are not jurisdictional unless the Court finds a clear indication that Congress intended otherwise.  The Court stated:  “Congress must do something special, beyond setting an exception-free deadline, to tag a statute of limitations as jurisdictional and so prohibit a court from tolling it.”  Slip. op. at 7.  The Court found nothing in the language of statute, its legislative history, or its placement in the United States Code to indicate that the FTCA time periods were jurisdictional.  The Court first noted that it has often found a time period to be non-jurisdictional when, as in the FTCA case, the time limit does not use the word “jurisdictional” and appears separated in the statute from the subject matter jurisdictional grant.  The Court then rejected three arguments made by the government:

First, the government argued that the language of the statute was unusually emphatic — stating that suit would be “forever barred” if the time limits were not complied with.  The Court rejected this argument by noting that the “forever barred” language was common drafting language for statutes of limitations at the time the FTCA was adopted in 1946. “Forever barred” appears in the Clayton Act at 15 U.S.C. § 15(b), but the Court had long before held that the Clayton Act time period was subject to equitable tolling.  American Pipe & Construction Co. v. Utah, 414 U.S. 538, 559 (1974).

Second, the government argued that the FTCA time periods were copied from an earlier version of the Tucker Act at 28 U.S.C.§ 2501(a), and that in John R. Sand & Gravel Co. v. United States, 553 U.S. 130 (2008), the Court had held that the Tucker Act 6-year time limit was jurisdictional. In response, the Court noted that John R. Sand acknowledged that the Tucker Act time periods would not normally today be called jurisdictional under current case law, but for over 100 years, and in multiple cases, the Supreme Court had called the Tucker Act time period jurisdictional — likely leading Congress to believe the time period was jurisdictional when it later legislated.  The Wong Court wrote:

No less than three times, John R. Sand approvingly repeated Irwin’s statement that the textual differences between the Tucker Act’s time bar and § 2000e–16(c) [involved in Irwin] were insignificant—i.e., that the language of the two provisions could not explain why the former was jurisdictional and the latter not. See 552 U. S., at 137, 139 (calling the provisions “linguistically similar,” “similar . . . in language,” and “similarly worded”). But if that were so, John R. Sandasked, why not hold that the Tucker Act’s time limit, like §2000e–16(c), is nonjurisdictional? The answer came down to two words: stare decisis. The Tucker Act’s bar was different because it had been the subject of “a definitive earlier interpretation.” Id., at 138.  [Slip op. at 13]

The Supreme Court in Wong noted that it had never issued an opinion on whether the FTCA time periods were jurisdictional, so there was no similar stare decisis concern here.  The Court was not willing to let the FTCA inherit the Tucker Act stare decisis holding of John R. Sand– even if the FTCA likely arose to fill a gap in the Tucker Act, which previously had not allowed tort suits against the government.

Third, the government argued that at the time Congress enacted the FTCA time periods, prevailing Supreme Court case law seemed to hold all time limits involving the federal government to be jurisdictional.  From this fact, the government argued that Congress’ expectation (even in the absence of any mention in legislative history) must have been that the FTCA time periods should be jurisdictional.  The Court rejected this line of argument as foreclosed by Irwin.  Irwin itself had applied its presumption in favor of tolling to a law drafted before Irwin announced the change in the Supreme Court’s view of the normal Congressional waiver of sovereign immunity.  The Court was not willing to adopt an argument that would have the effect of making the Irwin presumption only applicable to statutes passed after Irwin was announced.

So, what does the Wong case mean for the tax world?  Most obviously, Wong further throws into doubt any Circuit court opinions previously holding that the time periods in § 6532(a) (for refund suits) or (c) (for wrongful levy suits) are jurisdictional or not otherwise subject to equitable tolling.  Like the FTCA time periods, these two IRC time periods are not located near the jurisdictional grant to district courts at 28 U.S.C. § 1346(a)(1) (for tax refund suits) or (e) (for wrongful levy suits).  The language in § 6532(a) and (c) is less emphatic than that in the FTCA — i.e., they do not contain the words “forever barred” or similar words — so the government could not even make the unsuccessful emphatic language argument it made in Wong.  The Supreme Court has never ruled on whether these two IRC time periods are jurisdictional, so the government could not make the John R. Sand stare decisis argument with respect to the time periods.  And the time periods do not have numerous exceptions:  As Volpicelli noted, the wrongful levy time period has only one exception (“if it can even be called that”; 777 F.3d at 1046) — one that extends the period to bring suit if the plaintiff seeks administrative review before filing suit.

Indeed, it is hard to imagine anything the government could successfully argue as to why these IRC tax refund and wrongful levy time periods either are jurisdictional or rebut the Irwin tolling presumption, except for the argument that tax law time periods are not subject to equitable tolling per se. This per se argument has been made before, based on language in Brockamp.  Recently, the Ninth Circuit responded to it in Volpicelli as follows:

The government urges us to place overriding weight on one similarity that § 6511 and § 6532(c) do share: Both are found in the tax code. The government contends this shared feature is significant because the Brockamp Court observed, in the course of explaining why Congress did not intend to allow equitable exceptions to § 6511′s filing deadline, that “[t]ax law, after all, is not normally characterized by case-specific exceptions reflecting individualized equities.” 519 U.S. at 352. The Court may in time decide that Congress did not intend equitable tolling to be available with respect to any tax-related statute of limitations. But that’s not what the Court held in Brockamp. It instead engaged in a statute-specific analysis of the factors that indicated Congress did not want equitable tolling to be available under § 6511. The Court later made clear in Holland that the “‘underlying subject matter’” of § 6511—tax law—was only one of those factors. 560 U.S. at 646 (quoting Brockamp, 519 U.S. at 352). As we have explained, the other factors on which the Court relied are not a close enough fit with § 6532(c) to render Brockampcontrolling here. [777 F.3d at 1046]

If the Solicitor General wishes to bring to the Supreme Court the argument that no tax law time periods are subject to equitable tolling, he should feel free to.  However, I don’t think that is going to fly these days in a court that only recently unanimously stated that “we are not inclined to carve out an approach to administrative review good for tax law only”.  Mayo Foundation v. United States, 562 U.S. 44, 55 (2011).

ATPI Conference This Friday Highlights Tax Administration and Low-Income Taxpayers

This Friday April 24 at 8:45 a.m., the American Tax Policy Institute is live-streaming an all-day conference called “Delivering Benefits to Low-Income Taxpayers through the Tax System.”The conference is in Washington DC at the offices of Skadden Arps and is free and open to all. For those not in DC or unable to attend, the event will be live streamed here.

View the program and also register to attend the conference in person here.

The conference should have a great deal of content addressing tax administration as it relates to lower-income taxpayers. The panels include a morning session moderated by Professor Michelle Drumbl on US administrative efforts to improve delivery of the EITC that has as panelists Dave Williams, the Chief Tax Officer for Intuit, as well as Professor Bryan Camp (who like Dave has written for PT), and Jodi Patterson from IRS, who directs the Return Integrity and Correspondence Services.

I am moderating the  later afternoon panel which looks at international efforts to administer transfer programs through the tax system; Nina Olson will be on that panel as well as Professor Allison Christians from McGill and Steve Vesperman, the Deputy Commissioner of the Australian Taxation Office.

There are two other terrific panels  looking at take up of benefit programs and how families use the benefits. The full program linked above has the times and all the panelists.

Time has slipped away this week and I had hoped to discuss NTA Nina Olson’s testimony a week or so ago before the Subcommittee on Government Operations, Committee on Oversight and Government Reform. The testimony has a detailed discussion of issues relating to the administration of the EITC. I will save thoughts on the testimony until after the conference, though point interested readers to the testimony around page 24 which considers the balancing of taxpayer rights and the goal to reduce error rates. It is easy when thinking about EITC to focus on one of the goals and not the other. The improper payment rate as it relates to EITC is often looked at in a vacuum, without considering for example errors in other provisions in the tax code, the relative low direct costs of administering benefits in the tax system (see the table on page 26 presenting costs and benefits of many programs), and the role of preparers in facilitating claimant error. The NTA testimony addresses some of these issues and I suspect that there will be lots more discussed this week at the ATPI conference.

The NTA testimony discusses some of the experiences of other countries, including that of Australia. Broadening the inquiry to look at international issues is I think a very interesting part of this week’s ATPI conference and I look forward to learning more about the ATO experience.



Penalty Relief and Premium Tax Credit Reconciliation

Today we welcome back Christine Speidel, an attorney with Vermont Legal Aid who directs the low income taxpayer clinic there. Christine has specialized over the past year in tax issues arising from the Affordable Care Act and co-authored a new chapter in “Effectively Representing Your Client before the IRS” on ACA issues.  Keith

This is the first tax season that people who received advance payments of the Premium Tax Credit (APTC) must reconcile those payments on their federal income tax returns. APTC was paid during 2014 based on a person’s projected 2014 income (and other eligibility criteria), which may have been estimated as early as October 2013. Not surprisingly, many people are discovering that they received too much APTC or do not qualify for a Premium Tax Credit (PTC) at all, and they are having to repay all or a portion of it with their tax return. As of late February, H&R Block announced that fifty-two percent of its clients who received APTC had to repay a portion of the subsidy.

Taxpayers who have a balance due because of APTC reconciliation do get some relief. In Notice 2015-09 (IRB 2015-6, 2/9/15), the IRS announced limited penalty relief for 2014 only, for taxpayers who have a balance due as a result of excess APTC. However, Notice 2015-09 imposes several conditions that must be met for penalty relief to be granted. It also sets out procedural hurdles that will be difficult for some taxpayers to overcome.

The rationale behind the penalty relief for taxpayers with excess APTC is not fully set out in the Notice. However, some exchanges made erroneous APTC determinations in 2014 amid technological and operational difficulties. Also, many consumers did not understand how APTC worked. This issue was described in a recent New York Times article. The penalty relief is provided under the authority of Sections 6651(a)(2) and 6654(e)(3).  Thus, one could say that Notice 2014-09 amounts to a blanket finding of reasonable cause for late payment, and a concession that the imposition of the estimated tax penalty would be against equity and good conscience. (There are also administrative penalty waivers such as First Time Abate.)

This post will describe the penalty relief available under Notice 2015-09 and some of the barriers that may prevent low-income taxpayers from accessing the relief. I will then offer some thoughts on improvements that could be made.


Relief under Notice 2015-09


Notice 2015-09 provides relief from two penalty provisions: the penalty imposed by Section 6651(a)(2) for late payment of a balance due, and the penalty under Section 6654(a) for underpayment of estimated tax. Unfortunately, relief is not automatically applied to qualifying accounts, and relief from both penalties cannot be requested at one go.

The substantive criteria for relief are the same for both penalties. The penalties will be abated if a 2014 balance due was caused by APTC and: (1) the taxpayer filed a timely 2014 return (including by a properly extended due date); (2) the return reports excess APTC; (3) the taxpayer is current in all tax filing and payment obligations; (4) if the 2014 tax return was filed after 4/15/15, the taxpayer paid the balance due by 4/15/16; and (5) the taxpayer has requested relief. (See Notice 2015-09 at pages 4-5, and also Publication 974 (Mar. 2015) at page 8.)

Taxpayers will be considered current in their tax filing obligations if they “have filed, or filed an extension for, all currently required federal tax returns.” (Notice 2015-09 at page 5.) More controversial (in the LITC community at least) is likely to be the Notice’s definition of when a taxpayer is considered current in their payment obligations. If a taxpayer has any outstanding balances, the taxpayer must either have a current installment agreement or have entered into an offer in compromise. A taxpayer will also be considered in compliance if a “genuine dispute” is pending regarding the existence or the amount of a liability, so long as the liability has not been “finally determined.” (Notice 2015-09 at page 5). This definition of compliance will exclude many LITC clients from relief.

There are distinct procedures to request relief from each penalty. For the late payment penalty (known as the Failure to Pay or FTP penalty), relief is not requested with the tax return. Rather, the taxpayer must respond in writing to the IRS notice demanding payment of the balance (and charging penalties and interest). The taxpayer’s letter must request relief under Notice 2015-09 specifically. (See Notice 2015-09 at page 6.) In all cases, it seems the IRS will impose the FTP penalty before abating it.

There is a different procedure for requesting waiver of the estimated tax penalty. According to Notice 2015-09, “taxpayers should check box A in Part II of Form 2210, complete page 1 of the form, and include the form with their return, along with the statement: ‘Received excess advance payment of the premium tax credit.’” (p. 6) Theoretically at least, this penalty should never be assessed against taxpayers who qualify for relief.

Barriers to Relief


Two substantive restrictions on eligibility for penalty relief will bar deserving taxpayers from accessing relief. First, the narrow definition of compliance excludes many taxpayers, including people who have prior-year IRS debts in Currently Not Collectible status due to disability, unemployment or other hardship. Second, taxpayers who timely file a 2014 return after 4/15/15 are denied relief unless they pay off their balance by 4/15/16. There is no consideration given to the reason why the taxpayer filed under extension.

The complexity of the procedures for requesting relief is also worrisome. Notice 2015-09 was issued after the filing season had started, and well after most tax professionals had completed their annual continuing education courses. People are not likely to request relief from the estimated tax penalty unless prompted to do so by their preparer or software. I question whether the unrepresented low-income taxpayer population will be able to follow the procedures required to obtain relief from the FTP penalty (or even know to look them up). LITC practitioners will need to review future clients for this issue and request penalty relief for taxpayers when appropriate.

Broader context and ideas for improvement


Some of Notice 2015-09’s restrictions on eligibility for relief are puzzling given the circumstances that the policy is presumably intended to address. The procedural requirements mean that many people will not benefit from it even though they qualify. Other people will not qualify even though their 2014 circumstances relating to APTC are just as compelling.

Penalties should be administered with the goal of improving tax compliance. The IRS takes this approach in Policy Statement 20-1, found at IRM (06-29-2004). Compliance is improved when people believe that the system is fair. (See the National Taxpayer Advocate 2014 Annual Report to Congress, MSP#8, especially pp. 100-101 and n. 47.) One component of fairness is a rational relationship between the problem and the solution.

With all that in mind, there are three things that should be improved about APTC penalty relief.

First, the IRS should not penalize taxpayers who file under extension. There is no rationale given for requiring someone who timely files pursuant to an extension to finish paying the debt by 4/15/16, while imposing no such requirement on someone who timely files on 4/15/15. This requirement will entirely exclude some taxpayers from relief, without good reason.

Some people may not file by 4/15/15 because they still have not received correct 1095-A forms. I have two clients in that situation currently. It is possible that the forms will not arrive before 4/15. Treasury directs taxpayers who know that corrected forms are pending to wait and file their return when the corrected form arrives. (March 20, 2015 press release and accompanying FAQs.) In Notice 2015-30 (scheduled to appear in IRB 2015-17 on April 27), Treasury issued guidance extending penalty relief to taxpayers who are affected by a delayed or incorrect Form 1095-A. (Notice 2015-30 merits its own post and will not be discussed in detail here.) This is a good step, but it does not go far enough. There are other valid reasons for a taxpayer to file on extension. For example, Joe Kristan recently explained why K-1s are often filed as late as September 15.

It makes sense to encourage taxpayers to file timely returns, including by the extended deadline. However, I question whether restricting APTC penalty relief will actually prevent any late filed returns. Also, some of the other restrictions on relief do not seem to have a policy basis other than reluctance to extend any relief to “bad” taxpayers.

Not everyone will be able to repay their APTC promptly. I know of one Vermont taxpayer who must repay over $11,000 in APTC due to a change in circumstances that he did not realize he should report to the exchange. Just because a taxpayer’s 2014 AGI was over 400% of the poverty line, and thus they are ineligible for APTC repayment caps, that does not mean the taxpayer has plenty of cash now. For my clients, AGI is often inflated by cancelled debt. Many of my taxpayers have also taken lump sum retirement distributions that were withdrawn and used for a specific purpose well before tax time.

Second, a prior-year clean slate should not be a necessary condition before one can find that a taxpayer deserves relief from penalties caused by 2014 APTC reconciliation. If this penalty relief policy were permanent rather than applicable only to 2014, it would make more sense to take prior year compliance into account. The government certainly does not want to see taxpayers with significant, unpaid, or late-paid excess APTC year after year. It might slightly encourage taxpayers to gamble or even intentionally try to receive more APTC than they should if penalty relief were available every year. In the first year of this program, however, whether taxpayers have unresolved/unpaid prior-year liabilities does not seem particularly important to the question of whether they deserve relief from APTC-related penalties for 2014.

In general, prior-year compliance is not a requirement for reasonable cause penalty relief. Rather, it is one factor that the IRS considers in determining whether the taxpayer exercised ordinary business care and prudence. See IRM Requesting Penalty Relief (11-25-2011), #6. And that makes sense. The IRS should consider priory-ear compliance in the same way for APTC penalty relief.

The Service could at least specify some CNC closing codes that would be acceptable for purposes of qualifying for relief under Notice 2015-09. Taxpayers who are uncollectible due to residence in a foreign country need not be treated the same as taxpayers who are uncollectible due to unemployment or disability. (See table of CNC closing codes in IRM Currently Not Collectible Procedures (1-1-15)).

Third, the procedural complexity is worrisome. The NTA and other stakeholders have repeatedly criticized IRS for its tendency to impose penalties automatically and correct them later. See the National Taxpayer Advocate 2014 Annual Report to Congress, MSP#8, especially notes 26-33 and accompanying text on pages 97-98. The National Taxpayer Advocate’s 2014 Annual Report to Congress highlighted the IRS penalty regime as Most Serious Problem #8.

Generally, failure-to-file and failure-to-pay penalty relief may be requested in writing or over the phone. The Internal Revenue Manual (IRM) allows consideration of oral requests under reasonable cause or FTA. See IRM (08-05-2014), Unsigned or Oral Requests for Penalty Relief. This section does not currently encompass relief under Notice 2015-09. Relief from estimated tax penalties is not included at all in current oral waiver authority; a signed written request must be submitted. See IRM, #6 (08-05-2014).



When an individual owes tax to the IRS, several different penalty provisions of the Internal Revenue Code (IRC) may come into play. The penalty and interest provisions of the IRC often confound the taxpayers I work with. Many unsophisticated taxpayers are afraid to file a tax return showing a balance due. They frequently fail to understand how important a timely-filed return is.

Penalty relief that must be specifically requested by the taxpayer is less effective than a blanket policy that is automatically applied by the IRS. This is particularly the case for relief that must be requested in writing.

Taxpayers have the right to a fair and just tax system. (IRS Pub. 1) APTC penalty relief should be simplified substantively and procedurally so that it actually reaches the deserving taxpayers for whom it was designed. The penalty relief provided in Notice 2015-09 is a welcome step. But the Notice does not go far enough to help taxpayers confused by a new system. Penalties only promote tax compliance when they are administered in a way that is perceived as fair.


Tax Court Order Rejects APA Claim That IRS Precluded from Asserting Penalty in Answer

One of the issues we have looked at from time to time is the relationship of the Administrative Procedure Act (APA) to procedures embedded in the Internal Revenue Code. It is an issue we are currently working on as we finish the rewrite of Chapter 1 in Saltzman and Book, which among other topics addresses the relationship of the APA to certain IRS administrative procedures that generally fall under the broad administrative law category of informal adjudications (recall that agencies perform two broad functions in the administrative state: adjudicating and rulemaking). That relationship is complex, in part because when you mix the murkiness of administrative law with the labyrinth of tax procedure you wind up often with head scratching questions without any clear answers.

I came across an order from late December 2014 in the case of Illinois Tool Works v Commissioner that involved an interesting APA issue. Illinois Tool Works argued that under the APA the IRS’s failure to assert a penalty in its notice of deficiency meant that IRS was barred from asserting a penalty in pleadings. Judge Lauber took the APA-issue head on in an order denying Illinois Tool Works’ motion to strike the pleadings.

In this post, I will briefly describe the facts, Illinois Tool Works’ argument, and Judge Lauber’s approach to the issue.



The order lays out the facts:

In a notice of deficiency dated February 11, 2014, the IRS determined a deficiency of $70,174,594 for the 2006 taxable year. This deficiency is attributable to respondent’s determination that a transfer of funds to petitioner from a foreign subsidiary constituted a taxable dividend. The notice of deficiency did not assert a penalty under section 6662(d).

Petitioner timely petitioned this Court, contending that the transfer of funds in question constituted a nontaxable return of capital. Respondent filed his answer on July11, 2014. In paragraphs 3 and 7 of the answer, respondent alleges that, “pursuant to the provisions of I.R.C. § 6214(a), an increased amount [is] due from [p]etitioner, on the grounds that petitioner is liable for the accuracy-related penalty under I.R.C. § 6662(a) for the 2006 tax year in the amount of $14,034,919.”

In response to the answer Illinois Tool Works moved to strike the paragraphs asserting the 20% accuracy-related penalty. In its motion to strike it noted that neither exam or Appeals proposed to assert the penalty with respect to that issue, though exam  proposed the penalty with respect to an issue that was resolved.

General Principles

In support of its motion, Illinois Tool Works looked to administrative law principles:

Citing the Administrative Procedure Act (APA) and Mayo Found. for Med. Educ. & Research v. United States, 562 U.S. 44 (2011), petitioner argues that the assertion of a penalty for the first time in an answer is impermissible as a matter of law because such assertion would be inconsistent with what petitioner describes as a prior “determination” by respondent not to assert that penalty. As such, the delayed assertion of the penalty would supposedly be analogous to a disfavored “post hoc rationalization” by the agency. M Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 50 (1983); SEC v. Chenery, 332 U.S. 194, 196 (1947).

SEC v Chenery generally stands for the proposition that the courts are not supposed to allow agencies to argue a new reason for their determination, or justify agency actions based upon arguments or issues that were not properly made below.

The Chenery issue has been popping up in a number of Tax Court matters. Steve mentioned it last year in a Sum Op discussing a Judge Holmes order in Renka v Commissioner; I discussed it in a post last year considering Bergdale v Commissioner, which involved a CDP case where the taxpayer argued that under Chenery the IRS should not be allowed to argue that the determination should be sustained on the basis that the taxpayer failed to submit a Form 656 because that was a new legal theory not made in the CDP determination.

In my Bergdale post, I referred readers to Professor Steve Johnson’s outstanding 2014 article Reasoned Explanation and IRS Adjudication. One of the points Professor Johnson makes (page 1773) is that while it is clear that administrative law principles apply to tax, “administrative law is about nuance, and it must be adapted to the issues, agencies, and circumstances of the particular situation at hand. (citing to the 2004 Supreme Court case of Hamdi v. Rumsfeld where the Court emphasized the importance of context in applying administrative law to military proceedings).

The point Johnson makes is that yes we are in a post-Mayo world where there is no question that tax lawyers ignore administrative law principles at their own peril. But tax procedure is also nuanced, and there are many differing kinds of determinations that IRS makes. Context matters.

I point readers to the article at around page 1823 where he discusses the legal issues surrounding the application of the APA and general administrative principles to deficiency proceedings. I think Professor Johnson sensibly concludes both on policy and legal grounds that the reasoned explanation requirement (and by extension Chenery) should not apply to deficiency determinations. Other determinations (such as CDP cases) involve differing contexts, and the legal and policy justification for applying general administrative law principles or particular provisions under the APA may be much stronger as I have argued in Tax Notes in an earlier article CDP and Collections: Perceptions and Misperceptions [free link not available] criticizing the Tax Court’s conclusion in Robinette that certain provisions of the APA did not apply to CDP proceedings.

The Order’s Resolution of the Matter

Judge Lauber resolves the issue in large part by looking at how the context of the situation does not support precluding the IRS from asserting the penalty in its pleadings.

First he notes that Tax Court Rule 52 provides that the Court may strike any pleading that is insufficient or frivolous. As the Order describes, the Tax Court “does not favor motions to strike pleadings.”

He then goes on to consider how the Chenery argument “clearly proves too much.”

Our Rules explicitly permit respondent to assert an increased deficiency or “new matter” in his answer, Tax Court Rule 142(a), and Congress has specifically granted this Court jurisdiction to hear such claims. Section 6214(a) provides the Court with jurisdiction to redetermine a deficiency greater than that set forth in the notice of deficiency, “and to determine whether any additional amount, or any addition to the tax should be assessed, if claim therefor is asserted by the Secretary at or before the hearing or rehearing.” On petitioner’s theory, such a determination would be impermissible because it would be inconsistent with a supposed prior “determination” by respondent– embodied in the notice of deficiency–that a smaller deficiency was correct or that the new matter should not be asserted. That is clearly not the law. In this and in other respects, the specific procedures that Congress has ordained for this Court in the Internal Revenue Code may differ from the more general rules embodied in the APA.


We are in the early phases of determining how some administrative law principles and parts of the APA apply to IRS determinations. We will see other creative arguments where taxpayers seek to use general administrative law principles to determinations. While Mayo has heralded in a new phase in opening the door in tax cases to those principles, the hard work is just beginning as we struggle to see how and whether those principles will influence the varied types of IRS’s adjudicative determinations.

Judge Lauber’s approach in a case involving a possible penalty imposition in a deficiency case sensibly resolves the matter, as the context here would make applying Chenery in the way Illiniois Tool Works’ argued inconsistent with the specific approach that Congress and the Court have long applied. I am sure, however, that the Tax Court and circuit courts will have much more to say about applying the APA and general administrative law principles.

Summary Opinions for the week ending 3/27/15

How could I not start with John Oliver and Michael Bolton singing about the IRS.  This link is not really great for work, and to say it is sophomoric may overstate the sophistication and maturity.  Sexy singing is at the end of a fairly long clip, which is all pretty funny (on the IRS, “it combines two things we hate, people taking our money and math”).  This is probably the funniest Michael Bolton clip from the month, which is really impressive since it is about the IRS and he recently recreated the Office Space scenes with the character sharing his name – if you liked that movie, you should find the clip.  Equally as entertaining and enlightening were our guest posters during the week ending March 27, 2015.  Peter Hardy and Carolyn Kendall of Post & Schell did a two part post (found here and here) regarding the definition of willfulness in civil offshore enforcement cases.  First time guest poster, Bob Nadler, posted on the recent Sanchez case dealing with an interesting innocent spouse issue that hinged on whether a joint return was actually filed.  Thank you again for the great content.

I also need to thank our guest posters from the last week and a half.  Carlton Smith provided two of the three posts on the Godfrey case, the last of which can be found here and links to the first two.  Godfrey is an interesting case raising a couple issues regarding appropriate notice with collection actions.  We were also pleased to have Prof. Bryan Camp with a three part post on Eight Tax Myths, the last of which can be found here and links the first two.  Both sets of posts were very well received, and both generated a fair amount of discussion.  I would encourage everyone who has not read both sets to do so, and, for those who have, you might consider going back and reading the comments and responses.

To the other procedure: 


  • In a FOIA dump, the Service has released PMTA 2014-015, which discusses the erroneous refund penalty under Section 6676.  The following points are discussed in the memo:

1. Does the Section 6676 penalty apply to refund claims made on Form 1040 and Form 1040X and does it matter whether the Service has paid the claim?

2. Does the nature of the item to which the excessive amount is attributable have any bearing on the penalty?

3. Is the Section 6676 penalty subject to deficiency procedures?

4. Are there any specific taxpayer notifications required for the penalty to apply?

5. Does the ‘reasonable basis’ exception to the Section 6676 penalty have the same general meaning as the reasonable basis exception to negligence found in Reg. 1.6662-3(b)(3)?

 I’m not sure there are any earth-shattering realizations to be found in the IRS response, but some points seem worth noting.  As to the first point, the Service stated the penalty can be imposed even when the IRS does not actually refund the amount requested.  For the second point, the Service discussed the various situations where other penalties would apply (reportable transactions, EIC, etc.).  As to the third question, the Service stated the general rule that the penalty is not subject to the deficiency procedures, but stated that for some refundable credit cases the penalty will have to be assessed pursuant to the procedures.  No court has apparently addressed either point.  The last thing that jumped out at me was that the Service stated the reasonable basis exception under Section 6676 has the same meaning as under the accuracy related penalty provisions found in Section 6662, which is not news, but good reinforcement of the prior position.

  • Harper Int’l Corp v. US is a case we (I) missed in January (see page 13 of this PDF for a more robust recitation of facts and holding).  In the case, the IRS denied a refund request.  On May 2, 2012 the IRS issued a Notice of Disallowance, which stated the taxpayer had two years to challenge the determination.  About a month later, another notice was received by the taxpayer, stating the claim was rejected and another formal Notice of Disallowance would be issued – but it never was.  Taxpayer petitioned the Court of Federal Claims in June of 2014, more than two years after the first letter, but less than two years after the second letter.  The Court of Federal Claims held that although equitable provisions might apply, it was not reasonable for the taxpayer to rely on the second notice (and they failed to comply even if using the date of the second notice because timely mailing was not timely filing for CFC).
  • Another sham(wow) partnership case in CNT Invest., LLC v. Comm’r, where the Tax Court has held that gain recognized in a collapsed step of a multi-step transaction was gross income for determining the extended statute of limitations under Section 6501(3)(1)(A).  Case also confirmed limitations period was the longer of the period found under Section 6229 or Section 6501.
  • Businessweek thinks the IRS sucks.  The reasons are largely outlined by the John Oliver video above.  I’m sure this has generated a lot of scoffs, but I honestly do try to keep this in mind as I sit on hold for 90 minutes.  Maybe it helps me from being a complete jerk to the person who eventually picks up.  Solid chance that person’s day is worse than mine. How much longer before this all implodes? Is that the goal?  Might work.
  • Kardash v. Commissioner was decided by the Tax Court on the 18th, and has a good discussion of transferee liability but a difficult result for taxpayer minority shareholders in a company where the Service found transferee liability for tax due that was the result of theft by the majority shareholders.  This is going to get a little longwinded, sorry.  In Kardash, a concrete company was largely owned by two shareholders, who controlled all aspects of the business.  Two other minor shareholders oversaw sales and operations; neither had any control over the overall management or finances of the company.  During the early 2000s, the company was very successful and the minority shareholders received huge additional compensation.  Unfortunately, during this time, the majority shareholders were plundering the coffers and not paying any taxes ( one of whom is in the clink and the other is no longer with us).  Here is some more background on that sad story.  The finances of this company were apparently open for the taking, as two other employees were jailed for stealing over $5.5MM from it before the IRS got involved.  On audit, for 2003 to 2007, the Service assessed over $120MM in tax, penalties and interest.  The company was insolvent at that point, payment was not possible, and the company and the Service entered into an installment agreement to pay $70,000 a year until the end of time.  The Service reached agreements with the two majority shareholders, but substantial amounts of tax were still outstanding.  The Service then attempted to recoup a portion of the remaining amount from the minority shareholders pursuant to Section 6901(a).  For Kardash, the amount was around $4MM.  There were a host of questions before the Court regarding the IRS’s collection actions against the company and majority shareholders cutting off liability, but what I found interesting was the issue about whether, under state law, the minority shareholders were responsible for the tax due to fraudulent transfers to them by the majority shareholders.

For the fraud, the Court looked to Florida law to determine the extent of the potential transferee liability.  As an initial point, the Court did not aggregate the transfers with those of the majority shareholders (contrary to the Service argument), and instead looked at each payment to the minority shareholders to determine constructive or actual fraud of each payment.  The FL statute provides that if the company did not receive “reasonably equivalent value” for the payments, they may be fraud if: “(1) the debtor was engaged…in a business…for which the remaining assets of the debtor were unreasonably small…;(2) the debtor intended to incur…debts beyond his ability to pay as they became due; and (3) the debtor was insolvent at the time of the transfer or became insolvent as a result of the transfer.”  Kardash argued his work for the company was reasonably equivalent value, and the Court agreed for certain “loans” in 2003 and 2004, which were really advanced on compensation.  For 2005 through 2007, the funds were provided to Kardash in the form of a dividend from the Company.  The Court noted the conflict in cases regarding the treatment of dividends as “reasonably equivalent value” as compensation for work done.  The Court seems to indicate the general position is that dividends are not compensation for services rendered and therefore not an exchange for value.  In the limited cases holding the opposite, the dividend has been directly tied to work provided.  See In re Northlake Foods, Inc., 715 F3d 1251 (11th Cir.) (holding dividend made as tax distribution to pay tax due on s-corp shares); In re TC Liquidations, LLC 463 BR 257 (ED NY 2011) (dividend made to shareholder to repay loans taken out to expand business).  Although I have not read these cases, this seems like a point that could be open to other interpretation in this case.  The dividends here effectively replaced a prior bonus program.  The program was stopped and the company made the loan/advances to the minority shareholders because the company knew the minority shareholders needed that level of compensation.  This was a temporary solution until the dividends were to start.  Since at least a portion was compensation provided in a different form, a finding that it was received in exchange for equivalent value would not seem unreasonable in this case.  The Court did address this by stating the company did not benefit from the dividends as clearly as in the above two cases, but I am not sure I agree.  Had the dividends not be issued to take the place of the prior bonus program and advances, the minority shareholders may have left.  During the period in question, the company was very successful, arguably because of the minority shareholders.  The second reason is that the company and shareholders treated it as dividend income and not compensation.  Although a factor worth consider, I am not sure it has to be dispositive.

The issue of insolvency was reviewed next, with a few pages devoted to the debts and income stream.  The Court relied on the IRS’s expert’s opinion that since there were no tax returns, no buyer would ever pay more than the gross value for the land and tangibles, and the company had no intangibles.  Based on that, the company was insolvent most of, if not the entire time.  Interestingly, the opinion includes the gross revenue, but I don’t think it includes the asset values.  Ignoring the various other ways to value a company, I think this is also open to other interpretation.  I am not sure the conclusion that no one would be willing to buy the company is correct—obviously that would be a substantial risk, but business people often take risk if the reward appears sufficient.  I am also not sure the value of the intangibles was $0, since the revenue for the years in question was north of $100MM, which was substantially more than the hard assets.  Clearly, the company, as a going concern, had some value that exceeded hard assets.  The company may have still been insolvent, I just wasn’t sold on those particular points.  An interesting case, and what seems to be a tough result for some transferees who were screwed by their employer.

  • The Service has issued internal guidance indicating that it will no longer allow taxpayers to enter into installment agreements for post-petition liabilities when the taxpayer has filed for Chapter 13 bankruptcy.  The guidance indicates that this was previously allowed in some jurisdictions, but that the Service believes this potentially violates the BR stay.
  • 2014 data book has been issued by the Service in electronic form, and can be found here.  Lots of interesting stuff.  Looks like 40% of penalties were abated in terms of amount.  Less business returns, but more individual in 2014 than 2013.
  • Barry and Michelle paid an effective tax rate of 18.8% (maybe slightly higher –I’m finding some conflicting reports and too lazy to do the math) for federal income tax purposes.  I think that is a little higher than mine…although we made slightly different amounts.

Eight Tax Myths – Lessons for Tax Week Part III

We welcome back guest blogger Bryan Camp for part three of his three part series dispelling tax myths. In Part III he covers myths 7 and 8.  Post 1 can be found here and Post II can be found here. Keith

This post originally appeared on the Forbes PT site on April 15, 2015.

7. The IRS Abuses Taxpayers

Some myths are based on a kernel of truth and this is one of them. But the abuse is not what you think it is.

Taxpayer abuse happens when a taxpayer is not treated fairly according to his or her circumstances. The key to abuse is the idea of discretion.  On the one hand, if there is no discretion in applying rules, then some folks will inevitably be abused because their circumstances were not foreseen by those who wrote the rule.  On the other hand, giving someone discretion to bend the rules also results in abuse because the humans exercising discretion will inevitably make poor judgments in exercising the discretion.  That’s what it means to be human: to make mistakes.


Here’s a simple example: let’s say we want to give “equal” access to justice and so we have a rule that one must climb 39 steps to reach the courthouse door and file suit.  If you don’t climb those 39 steps, you cannot obtain justice.  If that rule applies to everyone “equally” so that there are no exceptions, then the rule abuses those folks who cannot climb the 39 steps because of some disability.  But if we station a person at the bottom to decide who must climb the 39 steps and who need not do so, we now create opportunities for abuse by that person.  We may put rules in effect by which that person must exercise the discretion but, at bottom, it’s a case-by-case determination.

So it is with the IRS. Both types of abuse exist but it is the second kind that most of us associate with taxpayer abuse.  For example, the gist of the so-called “targeting scandal” is that IRS employees had discretion on how to approve applications for exemptions and abused that discretion in giving extra scrutiny to applications from conservative organizations.  The mythology being built on top of this story is impressive, but my purpose here is to simply note the example.  Interested readers can find a lucid and geekily comprehensive take on the matter by Professor Philip Hackney of LSU.

This abuse of discretion certainly exists because IRS employees often have discretion to decide what actions to take on a taxpayer account, and so if follows that IRS employees sometimes abuse that discretion. You can even find court cases on this, where a court—usually the U.S. Tax Court—finds that an IRS employee abused discretion.  For example, taxpayers generally cannot ask for forgiveness of their tax liabilities, but there are exceptions and certain IRS employees have discretion to compromise the tax liability—i.e. forgive some or all of it—in some situations.  In a case called Szekely v. Commissioner, heard by the Tax Court in 2013, Mr. Szekely asked to compromise his tax debt and the IRS employee reviewing his request refused. The Tax Court ruled that the IRS employee did not treat Mr. Szekely right because the employee rejected the request a mere one day after Mr. Szekely missed a deadline to provide additional information.  Under the circumstances of Mr. Szekely’s situation—detailed in the Court opinion—that refusal was an abuse of discretion. Mr. Szekely got a “do-over.”

While this second idea of abuse—IRS employees abusing discretion—is sometimes true, it is a myth that such abuse is widespread. One has to go back to 1997 to even find allegations of widespread taxpayer abuse.  In was then that Senators Roth and Grassley held hearings—one set in September 1997 and one set in April 1998—hearings for which they had spent over a year carefully gathering the stories of this kind of supposed IRS abuse.  The hearings were high political theater and, as with most theater, were mostly fictional: the dramatic allegations made from behind face screens and voice screens turned out to be almost all false, according to later independent investigations from the General Accounting Office and the Treasury Inspector General’s office.

If you really want a juicy IRS “scandal” you need to go back to the 1940’s, as I have explained in gory detail elsewhere. That scandal resulted in the removal or resignation an extraordinary number of high-level officials, including the Assistant Commissioner in Charge of Operations, the Chief Counsel, the Assistant Attorney General in Charge of the Tax Division of the Department of Justice, and 9 high-level agency employees, 3 of whom were also criminally prosecuted.

In contrast, the 1997 and 1998 so-called scandals resulted in no criminal prosecutions, no forced resignations, no removal from office. The sole individual implicated in the congressional hearings was a mid-level manager.  And yet, of the 20 allegations against this individual—described in an 80 page narrative report with 2,200 pages of attachments reflecting interviews with 140 people, including every revenue officer and manager in the this individual’s office—only 6 violations of policy were substantiated, none involving abuse of specific taxpayers.

And the current “tea party” scandal? There have been two forced resignations and no prosecutions.

The clear-eyed view of our tax system is to see how it is effectively administered by computers, not humans.  Once the myth that humans run the show is exposed, one can more easily see the devastating effects of budget cuts as I have elsewhere blogged. And so we move to Myth #8.

8. The IRS is Run by Humans

Lots of folks work at the IRS. Even after five years of punitive budget cuts, the IRS still employs some 80,000 workers.  It is true that the IRS budget is mostly spent on personnel costs.  It is true that the face of the IRS is human, currently the Commissioner John Koskinen, who appears as able a Commissioner as any since the redoubtable George S. Boutwell held the position in 1862.

But behind those employees, behind that elfin face of Koskinen, lies the heart of our tax administration system: computers. The story of tax administration since WWII is the story of increased reliance on automation, on computers.  And computers run on rules.  Strict rules. The basic job of the humans at the IRS is actually to prevent the abuse that results from rigid application of rules.

The myth that humans run the IRS, that the fiendishly complex Tax Code (not IRS Code!) created by Congress is administered by actual human beings, obscures both what is good and what is not good about the current state of tax administration. Let’s look at two examples.

First, on the tax assessment side, we have already seen in Myth #4 how IRS computers rule the returns processing function. If one in a series of computer filters flags a return, the IRS will not process that return unless, and until, an IRS employee decides to accept it.  Put another way, it is the computer that decides whether to assess the tax reported on a return.  It takes a human to override the computer’s decision.  The good part about this is that the computers can process millions of returns far more efficiently than humans.  The bad part is that if the computer makes an error, the risk falls entirely on the taxpayer because it is up to the taxpayer to see the error and find some human IRS employee to fix it.

Similarly, the IRS has other computer systems that relentlessly propose increases to tax liabilities, increases that will be assessed unless and until a human IRS employee decides to override the decision of the computer. One example is the Automated Underreporter system.  It compares the W-2’s or 1099’s filed by third parties to what the taxpayer reports on the Form 1040.  If the numbers from the third party information returns are more than what the taxpayer has reported on his or her return, the computers follow the rule that the numbers on the W-2’s or 1090’s are correct.  The computers (not humans) send out a notice to the taxpayer.  And if the taxpayer does not respond to the computer-sent proposal in the right amount of time and to the right office with (often) the right-sized envelope, the computers (not humans) automatically print out the increased tax for assessment.  No human being ever reviews these cases before assessment unless and until there is a recognized taxpayer response to stop the computers.  The IRS employs a similarly automated system for what it calls “correspondence exams.”  Here is how the National Taxpayer Advocate described that system:

Once the IRS engages the batch system, cases move through the examination process automatically. Each step in the process has a pre-established period programmed into the system.  Files are not created or examiners assigned to the cases until the IRS receives and controls a taxpayer‘s correspondence.  If a taxpayer fails to furnish the requests documentation precisely within the prescribed period, the case automatically moves to the next phase in the process.  … Because the batch system automatically processes a case from its creation through the issuance of a Statutory Notice of Deficiency and subsequent closing, the IRS has effectively eliminated the need for human involvement in every case in which a taxpayer does not reply in a timely fashion.


We see the same model of computer action on the collection side, in the Automated Collection System. This is the computer system that collects unpaid taxes.  The computer, not humans, matches the W-2 information filed by employers and the Form 1098 information filed by banks to identify where taxpayers work or have assets.  Then the computer, not a human, sends out a Notice of Levy to grab wages or grab a taxpayer’s bank account.  It is the computer, not a human that will file a Notice of Federal Tax Lien (NFTL), which not only makes it difficult for taxpayers to sell their homes but also results in a major hit to the taxpayer’s credit rating.  If the taxpayer wants to undo any of these actions, or if the taxpayer believes the actions were abusive or unfair, the taxpayer must find a human IRS employee and persuade that employee to use discretion to ameliorate the damage.

Just look at the volume of levies and NFTLs in FY14: over 2 million levies and some 535,000 NFTLs.  That’s the work of computers, not humans.  And while the computers might be shutdown if the IRS shuts down, they will keep on working, even if employees must be furloughed. So finding the human to override the computer becomes increasingly difficult.

The clear-eyed view of our tax system is to see how it is effectively administered by computers, not humans. Once the myth that humans run the show is exposed, one can more easily see the devastating effects of budget cuts.  First, fewer and less trained employees means that taxpayer will be increasingly unable to undo the mistakes that computers will inevitably make.  Second, the less money the IRS has to hire and train competent employees, the more reliant tax administration will be on computers.  One sees this in the current IRS plea for expanded math-error authority.  The problem with this expanded authority, as Keith Fogg promises to explain in his typically thoughtful way, is that it will simply shift more work current done by humans to computers, creating more room for computers to make automatic initial decisions against taxpayers that then can only be undone by humans if and when the taxpayer responds.

This concludes my series on myths about tax and tax administration. I have little doubt most of them will persist.  That does not trouble me.  As a teacher, my job is to introduce curious minds to new ideas and give those who are interested tools to pursue their interests further.  I hope at least to have accomplished that.



Eight Tax Myths – Lessons for Tax Week Part II

We welcome back guest blogger Bryan Camp for part two of his three part series dispelling tax myths. Yesterday he covered myths 1, 2 and 3. In Part II he covers myths 4, 5 and 6. Keith

This post originally appeared on the Forbes PT site on April 14, 2015.

4. The IRS Exists

People commonly refer to the IRS as if it were a sentient being, like the Borg from Star Trek. That’s part of the charm when Ted Cruz refers to the “IRS Code.” Double myth-making: the evil Borg writes the tax laws. But more sensible people also fall into the shorthand. Heck, I do it too. Go to yesterday’s post in Myth #3 where I wrote that the IRS “could choose to instead file suit in federal court.” As if the IRS can “choose” to do anything! The anthropomorphic metaphor is so commonplace and convenient that it leads us into mythology.

In truth, the IRS does not exist either in fact or in law. As a matter of fact there are just a bunch of people, organized into offices, each with assigned tasks, assigned functions, and each placed into a hierarchy of review. In other words, it’s a bureaucracy. And it’s what I get paid to teach, to study, to write on and to think about. Lucky me. So it’s the people working in the agency, not some Marvel Comics SuperBeing, who do stuff — bad stuff, good stuff, whatever. It is not the IRS that allegedly gave extra scrutiny to conservative organizations that sought to be exempt from paying tax on their incomes. It was individuals in Cincinnati, or in Washington D.C. who allegedly did that. Having working the bowels of the bureaucracy for eight years, I just snort with derision when I read yet another conspiracy theory about “the IRS.” The whole tax-exempt so-called scandal (more on that tomorrow in Myth #7) is much more about office politics—bickering between field employees and the National Office employees—than partisan politics.


As a matter of law, what we now call the IRS grew out of that same seismic 1862 revenue legislation I discussed in yesterday’s posts. It was at that time that Congress created the position of “Commissioner of Internal Revenue” and made that person the head of an office within the Treasury Department “to be called the office of the Commissioner of Internal Revenue.” The first Commissioner was the indefatigable George S. Boutwell, who took office in On July 17, 1862. He started with just three clerks but by the end of 1862 he had ramped up operations to 3,882 employees, all but sixty scattered throughout the non-rebelling states. The guy was an operator. As these numbers suggest, and as I have written in boring academic articles, tax administration was in large part a field operation involving what one member of Congress denounced as an “army” of tax collectors and another colorfully termed “pygmies.” Over time, this army of pygmies became known as the Bureau of Internal Revenue (BIR). The name changed to the Internal Revenue Service only some 90 years later, as part of a reorganization in 1952 following a real scandal. More on that tomorrow, in Myth #7.

If you go read the tax statutes, you will see something curious: they are written as if one single individual—the Secretary of the Treasury—is responsible for the entire tax system. For example, here is the foundational language Congress uses to empower the collection of tax: “The Secretary is authorized and required to make the inquiries, determinations, and assessments of all taxes…imposed by this title.” The tax statutes do not place duties or obligations or give directions to the IRS and rarely to the Commissioner. Everything is to be done by one individual: “the Secretary.” And so that is where the chain of command starts, with various duties delegated down from the Secretary to the Commissioner to various other offices within the agency. And so grows the myth: the IRS exists.

5. Returns are Rarely Reviewed

USA Today recently told its readers that “the audit rate, the percentage of individuals’ tax returns IRS revenue agents examined either in person or via correspondence, fell to 0.86% last year.”

That’s a true statement as far as it goes, but it does not go very far. It is generally combined with myth #3 (from yesterday) about self-assessment to create a myth that taxpayer returns are rarely reviewed.

In fact, all returns—yep, 100%—are reviewed, scrutinized, inspected, verified, analyzed, checked, checked out, checked over, investigated, looked over, probed, and otherwise studied, before the tax liability is assessed. And remember, an IRS employee makes the assessment not the taxpayer. Most of the review comes during what is known as “returns processing.” While this review is not as extensive as a full-scale audit, it is wrong to think that returns processing review rubber stamps whatever the heck taxpayers see fit to report on their tax forms.

The processing of returns subjects every return to various computer filters to decide whether that return is likely enough to be correct to be accepted as filed. If the computer flags the return, it gets kicked out and does not get processed through to assessment until passing through additional verification procedures. Nina Olson, the National Taxpayer Advocate makes this point nicely in describing the arduous journey that what she terms “every poor little refund return” must take before the IRS accepts it. During the 2013 filing season, says Ms. Olson, 12.3 million returns were kicked out of the processing flow because of potential errors identified during processing—meaning they were sent to “error resolution,” an office that requires taxpayers to present additional information to get their returns processed. All refund returns are run through an Electronic Fraud Detection System and suspicious returns selected by the computers are held until the income and IRS employees can verify withholding.

These automatic filters are by no means perfect. The latest hot issue is that the IRS got scammed to the tune of some $5.8 billion dollars by identity thieves. More on that in Myth #6. Of course, what is not as often noticed is that the filters did catch and stop some $24 billion in fraudulent refund claims.

Returns claiming certain types of credits are subjected to additional review over and above the series of computer filters. According to Ms. Olson, during the 2103 filing season some 358,000 Earned Income Tax Credit returns and 90 percent of returns submitting claims for the adoption credit were kicked out the regular processing stream by computers and given individualized review by IRS employees.

Ms. Olson takes a dim view of how the IRS processes returns because it holds up the refund process for many honest taxpayers. But whether one views the current process as benign or malignant, the fact remains that the process does not simply “accept” returns as filed and blindly or mechanically “assess” what the taxpayer reports.

6. The IRS Wastes Money on Erroneous Refunds

I love my yearly tax refund. I know I am not supposed to, but I do. Theorists argue that my refund represents an interest-free loan to the government and I am supposed to resent that, for reasons grounded in the obsessive individualism of our culture. But even accepting the argument’s premise, the value to me of that yearly manifestation of my forced savings outweighs the value of the interest I would theoretically earn in my checking account. I don’t worry, I be happy.

I am not alone. In FY2013, the IRS made lots of folks happy, sending out some $313 billion income tax refunds to individual taxpayers on gross collections of $1.564 trillion. Some folks get extra happy because their refunds were in error, either because of an IRS mistake or because their tax refund scam worked. No one knows the exact dollar amount of erroneous refunds issued, but the Government Accountability Office recently accepted $5.8 billion as a reasonable guess just as to the amount issued in FY2013 on account of identity theft.

Some folks use these numbers to propagate a myth that the IRS commits significant error in making refunds. Similar to the myth of the IRS Code, this is myth that seeks to fix the blame and not fix the problem.

For example, I was watching IRS Commissioner John Koskinen testify before the House Appropriations Subcommittee on Financial Services on March 18, 2015. There, Representative Tom Graves (R-Ga.) asserted that the amount of erroneous refunds was significant because it was “more than half, I guess, of your entire agency’s budget.” Graves also asserted that the IRS issued the $5.8 billion “knowing that $5.8 billion could go to defense, it could go to so many other needs within our country right now….but instead criminals all across the country, if not across the world, are receiving these tax refunds.” Using these comparisons, Graves was seriously concerned that the IRS was screwing up big time. Notice here the use of Myth #4: an entity called the IRS “knew” the erroneous refunds could go to other purposes.

Graves’ office was so proud of his performance that they posted his part of the hearing on They should not have. The excerpt makes Graves look like a clueless git.

First, Grave’s claim that the money erroneously refunded could have gone to some other program is fatuous. There is no direct linkage between money collected and money spent. Congress first must budget for an expense and then it must appropriate the money for the expense. Once it takes those two actions, the executive branch has to spend the money, and if there is not enough money to spend, the executive branch must borrow the money, again subject to broad controls by…Congress itself. It’s not like someone from the IRS says “hey, we found some extra money here!” and runs over to Congress with a check for Congress to immediately spend on those “other needs,” which—speaking of wasting taxpayer dollars—no doubt include many new bridges to nowhere. You would think Graves would understand that basic budget process, him being on the House Appropriations Committee and all.

Second, and more importantly, this excerpt shows that Graves sucks at numbers, which is pretty sad given his committee appointment. His comparison of the erroneous refund rate to the agency’s budget is nonsensical. As best as I can make out, he is suggesting that the agency gave away half of the money the Congress appropriated to it. Really? Is he really trying to say something like “hey, we gave you a budget of more than $10 billion and you gave half of it away”? If so, that’s really stupid because at the same time the IRS was erroneously refunding $5.2 billion it was properly collecting $1.56 trillion in individual income taxes alone (i.e. not counting corporate income taxes, excise taxes, “death” taxes, or employment taxes). So in Grave’s own terms, the agency was not giving away half its budget; it was collecting some 156 times its budget. If you really want to compare dollars collected to IRS budget, study after study shows that every additional dollar allocated the IRS results in multiple additional tax dollars collected. As the National Taxpayer Advocate reported in 2013: ““For virtually every other spending program, a dollar spent is just that – it increases the deficit by one dollar. But a dollar spent on the IRS generates substantially more than one dollar in return – it reduces the budget deficit.”

A more reasonable approach would start with the question “is the IRS collecting what it should?” I figure that $5.8 billion is about 0.04% of the $1.56 trillion properly collected. So one could as well say that the IRS properly collected 99.96% of what it would have collected if it had not sent out these refunds. Yes, $5.8 billion is a lot of moola, but it’s a 0.04% drop in the tax collections bucket.

Another reasonable approach would be to ask “is the IRS refunding what is should”? That is, measure IRS performance here by comparing erroneous refunds to all refunds made. In FY13 the IRS issued $314 billion in total refunds. I figure that the $5.8 billion in erroneous refunds is about 1.8% of all refunds made. So Graves is pounding on the IRS for correctly refunding 98% of all refunds instead of….what? 100%? Even measuring the IRS performance against what a perfect agency would do is far more reasonable than the budget baseline Graves hurls around.

By almost any measure, the IRS does a great job in getting the right refund to the right taxpayer in a timely fashion. But errors happen. Worse, given the huge amounts involved, even small error rates add up to large numbers. It remains true that the $5.8 is a large number and even a blemish can become cancerous, so the IRS management is appropriately concerned about this situation. It has diverted more than 3,000 employees out of its increasingly small workforce from other tasks to deal with the problem. But in the hearings I watched, Graves is so set on perpetuating the myth of IRS screw-up that he completely ignores the suggestions patiently and repeatedly offered by Commissioner Koskinen on how to fix the problem.

Tomorrow: Myth #7 (The IRS Abuses Taxpayers) and Myth #8 (The IRS is Run by Humans).