Between a Rock and a Hard Place

Today’s column discusses the Tax Court’s jurisdiction to hear stand-alone petitions from taxpayers who have been denied equitable relief under section 6015(f), one of the three spousal relief provisions contained in section 6015. Congress enacted section 6015 in the Internal Revenue Service Restructuring and Reform Act of 1998 (RRA 98) to replace the old “innocent spouse” provisions in section 6013(e).[1] My prior columns have focused on how RRA 98 inserted inappropriate adversarial process checks on the traditional inquisitorial process of tax collection (using the collection “delay” process provisions in sections 6320 and 6330 as a prime example). The jurisdictional controversy over section 6015(f) reveals a different problem: how Congress pinned the Tax Court between the rock of sovereign immunity and the hard place of doing equity. Thus caught between the commands of the head and the demands of the heart-to use historian Perry Miller’s famous dichotomy – the Tax Court has made the wrong choice. This column explains why.

The problem arises from a laudable congressional effort to make the code fairer. In 1998 Congress sought to expand how taxpayers who had signed joint returns could be relieved of the resulting joint liability. Congress charged the Tax Court the responsibility of reviewing IRS application of the law, both in section 6015 and simultaneously in the collection due process (CDP) provisions of sections 6320 and 6330. But the taxwriters goofed up section 6015 and failed to make the Tax Court’s authority commensurate with its responsibility.

Section 6015 compromises two different visions of joint liability relief: one proposed by the HouseWays and Means Committee and the other proposed by the Senate Finance Committee. I’ll have more to say about those competing visions in my next column. Right now, I’ll note that the unhappy result of the compromise is that literal language of the statute pinches off Tax Court review of section 6015(f) determinations when there is no deficiency at issue. Was it deliberate? No matter. It was done. That it was poorly done is of little use to the Tax Court, which sucks the jurisdictional air it breathes through nozzles of statutory grants of authority. Shut those off and the Tax Court is bereft. The irony here is that in its attempt to expand equity, Congress created inequity. That is, taxpayers who seek their day in court through section 6015(e) are barred, but taxpayers who successfully obtain their CDP hearing and present their claim for spousal relief can then obtain Tax Court review, regardless of whether the IRS is trying to collect a self-reported tax or a deficiency. Substantive equity without procedural equity is no equity at all. That, in a nutshell, is why one can be entirely sympathetic to the Tax Court’s creative overreaching in Ewing to inflate its jurisdiction under section 6015. But the Tax Court is wrong on the law and no amount of huffing can make it right.

I am prompted to this subject by the Eighth Circuit’s call for amicus briefs in Bartman v. Commissioner (No. 04-2771) on appeal from T.C. Memo. 2004-93, Doc 2004- 7609, 2004 TNT 67-17 (April 2, 2004). Accordingly, Part I explains the issue confronting the Eighth Circuit in Bartman. Part II discusses the structure and history of section 6015. Part III suggests the proper resolution of Bartman (the Tax Court has no jurisdiction).

I. The Unhappy Petitioner in Bartman

Ms. Bartman and her husband timely filed joint returns for 1994, 1995, and 1996.[2] The IRS accepted the returns as filed and assessed the self-reported liabilities per its authority in section 6201(a)(1). For 1995 the taxpayers failed to pay roughly $12,000 of the tax liability they self-reported on that return. That is, they had a $12,000 underpayment of tax. In 1997 Ms. Bartman

separated from her husband and timely filed an individual return, reporting “$4,865 in income, $264 in income tax withholding and $1,658 as an earned income tax credit.”[3] On her 1997 return Ms. Bartman requested a refund of $1,922. Instead, on June 28, 1998, the IRS set off her $1,922 overpayment against her jointly owed 1995 liability, under the authority of section 6402 and so partially collected that liability.

On February 2, 2001, Ms. Bartman filed a Form 8857, asking the IRS to relieve her of her joint and several liabilities for 1994, 1995, and 1996.[4] On March 4, 2001, she filed a Form 1040X for 1997, again requesting that the IRS refund her $1,922 overpayment for that year because it had been erroneously applied to the 1995 tax liability for which, she optimistically claimed, she was going to be relieved as soon as the IRS granted her request for spousal relief.

On January 29, 2002, the IRS denied her request for spousal relief. Ms. Bartman filed a timely petition asking for Tax Court review and the whole matter went to the IRS Office of Appeals for settlement talks. The appeals officer granted Ms. Bartman relief from the 1994 liability under the authority of section 6015(b) and relieved her of the remaining unpaid 1995 and 1996 liabilities under the authority of section 6015(f). So at least she got that. But the appeals officer would not authorize return of the $1,922 that the IRS had set off against the 1995 liability. Accordingly, Ms. Bartman and the government agreed to submit the refund issue to the Tax Court.[5]

In Tax Court, the government did not argue the jurisdiction issue. In fact, the government’s pleadings conceded jurisdiction.[6] That stage of the case focused on whether Ms. Bartman could get her $1,922 back, and the issue crystalized around whether the January 2, 2001, request for spousal relief, if treated as an informal claim for refund, was made in time to satisfy the section 6511(b) lookback limitation periods. If one considered her claim to ask for her 1997 overpayment, the three-year lookback rule would apply and would allow the refund because her claim was made within three years of filing her 1997 return. But if one considered her claim to be asking for her 1995 overpayment, the two-year lookback rule would apply and would deny the refund because her claim was made later than two years after she made the 1995payment (albeit involuntarily) on June 28, 1998.[7]

On April 5, 2004, Judge Foley decided that section 6511(b)’s two-year lookback rule applied and, accordingly, the IRS was barred from returning her $1,922. That was a straightforward decision and, however unfortunate its result for Ms. Bartman, would be an easy case to affirm on appeal. Despite the long odds, Ms. Bartman appealed to the Eighth Circuit. Not content to simply win, the government on appeal decided to rub salt in the wound and raise a jurisdictional argument. Of course, the government can always play the jurisdictional trump card, even when it concedes the issue before the trial court, because subject matter jurisdiction is widely believed to rest on the sacrosanct concept of sovereign immunity. If Congress has not consented to be sued by giving a federal court authority to hear cases against it, courts generally view it as a subject matter jurisdiction issue.[8]

One might wonder why the government felt compelled to attack the Tax Court’s jurisdiction in this particular case against a woman whose income for 1997 was less than $5,000 and who was represented on a pro bono basis. Why didn’t the government raise the issue in Tax Court in later cases and then take an appeal itself when the Tax Court rejected the government’s argument, as it was sure to do? Several signs point to no more sinister reason than a shift in the IRS’s thinking on the issue, perhaps as a result of debate between the Office of Chief Counsel and the Department of Justice Tax Division’s Appeals Section, or perhaps as a result of new attorneys looking at the issue and the bad memories of the 1997 Senate Finance Committee hearings fading away. First, while the IRS had completely capitulated in response to Fernandez in a 2000 acquiescence, the IRS backed off from that total acquiescence in September 2004 and put the tax community on notice that it was going to reassert the jurisdiction argument.[9] By then, of course, the Tax Court had issued its decision in Bartman, so it was just too late to raise it, except for the fortuity that the taxpayer appealed. Second, few stand-alone petitions are decided by the Tax Court (my quick LEXIS search showed eight in 2003 and four in 2004).[10] So the government has few opportunities to raise this issue to the appeals court and it appears likely that the Bartman appeal was one of the first, if not the first. Finally, because the government is simultaneously pursuing the same argument to the Ninth Circuit in Ewing v. Commissioner,[11] it is difficult to say it is singling out poor Ms. Bartman, not that she probably takes much comfort from that.

Thus it is that the Eighth Circuit is currently presented with the issue of whether the Tax Court had jurisdiction to hear Ms. Bartman’s petition. Before turning to the structure and history of section 6015, I will sketch out the government’s simple two-step argument as to why the Tax Court lacks jurisdiction. First, the Tax Court’s jurisdiction in this case rests entirely on a parenthetical in section 6015(e)(1)(A). Only if (e)(1)(A) applies can the Tax Court review Ms. Bartman’s petition. Second, the plain language of (e)(1) bars Ms. Bartman access to (e)(1)(A). It’s a closed door. Ms. Bartman cannot invoke the jurisdiction granted in subparagraph (A) because she is shut out of the entire paragraph by the language that starts it. I’ll explain each step in the argument in turn.

The first step in the government’s argument is that the Tax Court’s jurisdiction here rests entirely on section 6015(e). The Tax Court usually exercises what is called its deficiency jurisdiction under section 6213(a). That statute allows it to redetermine deficiencies in tax that the IRS proposes to assess against taxpayers. In those cases, the IRS has sent out the infamous notice of deficiency, which is the key needed to unlock the 6213(a) jurisdictional door.[12] Taxpayers who then petition the Tax Court have long been able to raise claims for spousal relief as one defense to the proposed deficiency.[13] But when an assessment is based on a taxpayer’s self-reported liability, there is (by statutory definition) no “deficiency” and so there can be no notice of deficiency.[14] So Tax Court jurisdiction must come from somewhere else. Since RRA “98 that somewhere else can be either the CDP statutes of sections 6320 and 6330, or it can be section 6015(e). I will discuss the CDP connection next month. Ms. Bartman’s case rests on section 6015(e).

Section 6015(e) allows taxpayers in certain cases to petition the Tax Court (and gives the Tax Court jurisdiction over petitions filed by those qualifying taxpayers) when the IRS denies their request for spousal relief under section 6015, even when section 6213(a) would not give the Tax Court jurisdiction over the deficiency. For example, suppose the IRS sent a notice of deficiency to a couple and the couple did not contest the proposed deficiency in Tax Court within the period allowed by section 6213(a). In that case, the Tax Court can no longer take jurisdiction over the controversy under section 6213(a) and the IRS will assess the deficiency and start to collect it. But an individual spouse could, at any time up to two years after the IRS starts collection action, ask the IRS to grant spousal relief under section 6015 and, if the IRS denies the relief or fails to respond within six months, section 6015(e)(1)(A) says that “the individual may petition the Tax Court (and the Tax Court shall have jurisdiction) to determine the appropriate relief available to the individual under this section.” Such a petition is called a stand-alone petition because the section 6015 issues “stand alone” and are not part of any broader attack on the deficiency. In fact, in stand-alone petitions the Tax Court properly denies it has jurisdiction to consider any reasons to modify the petitioning taxpayer’s liability other than reasons authorized by section 6015.[15]

The second step in the government’s argument is that Ms. Bartman is not the kind of taxpayer who can seek court review under section 6015(e) when the IRS denies her requested section 6015(f) relief. Section 6015(e)(1) starts with this sentence: “In the case of an individual against whom a deficiency has been asserted and who elects to have subsection (b) or (c) apply.” The government says that sentence limits the kind of taxpayers who can file a stand-alone petition to taxpayers (1) against whom a deficiency has been asserted and (2) who elect relief under section 6015(b) or (c). The government doggedly denies that Ms. Bartman meets either qualification, because the IRS never proposed a deficiency against her for 1995 and she did not ask for relief under section 6015(b) or (c) for 1995.

The government is correct that the Tax Court had no jurisdiction, but not for all the reasons it asserts. Part II will explain why the structure and history of section 6015 supports reading the statute as allowing only taxpayers against whom a deficiency has been asserted to use the section 6015(e) door to Tax Court review. Part II will also show why interpreting the statute as conditioning court review on a taxpayer invoking some magic words when and history. Any taxpayer who seeks to avoid the collection of a deficiency by asking for any relief under section 6015 can obtain Tax Court review, whether or not the taxpayer recites a certain incantation, as I shall explain further in Part III.

II. The True Legislative History of Section 6015

Section 6013(d) (3) provides the general rule that spouses who file a joint return are jointly and severally liable for all taxes shown on the return (which includes deficiencies and penalties successfully asserted by the IRS). Courts and commentators commonly state that section 6015 provides three exceptions to that general rule.[16] That is not quite accurate. Understanding why it is not quite accurate is important to understanding why the Tax Court lacked jurisdiction in Ms. Bartman’s case. Subpart A explains how the plain language and structure of section 6015 support the notion that only taxpayers who are the subject of a deficiency can invoke Tax Court jurisdiction. Subpart B then reviews the history of the exception to the joint liability rule, starting with the events leading to the original 1971 legislation and the 1984 modification to show how the plain language reading is consistent with prior law. Finally, subpart C takes a very close look at the RRA “98 reforms and the legislative history of section 6015. That history shows exactly where and how Congress failed to give the Tax Court the necessary authority to exercise its review responsibility over stand-alone petitions for review of section 6015(f) relief when the IRS is neither asserting nor pursuing a tax deficiency against the requesting spouse.

A. The Current Structure of Section 6015

As I mentioned above, many courts and commentators simply recite that section 6015 provides three exceptions to the joint and several liability rule. In so doing, they skip subsection (a). For most purposes, doing so presents no problems. But for purposes of determining the meaning of subsection (e)’s grant of jurisdiction to the Tax Court, one should start an examination of section 6015 by noting that subsection (a) provides that “an individual who has made a joint return may elect” only two avenues of relief from the joint and several liability rule, not three: the relief provided in subsection (b) and the relief provided in subsection (c). The import of that language will become clearer after a review of its history, below.

Subsection (b) provides the current form of “innocent spouse” relief to an individual (whether currently married or divorced). It is a liberalized version of the old rules in section 6013(e) but still applies only when it seeking section 6015 relief is unsupported by its structure would be “inequitable to hold the . . . individual liable for the deficiency in tax”(emphasis added).

Subsection (c) provides what is called “separation of liability” relief but only “for any deficiency which is assessed with respect to the return” (emphasis added).

Subsection (e) provides that:

(1) In General.-In the case of an individual against whom a deficiency has been asserted and who elects to have subsection (b) or (c) apply –
(A) In General. – In addition to any other remedy provided by law, the individual may petition the Tax Court (and the Tax Court shall have jurisdiction) to determine the appropriate relief available to the individual under this section (emphasis added).

Subsection (f) provides what is called “equitable relief” because it authorizes the IRS to grant ad hoc exceptions to the general joint and several liability rule. The only requirements are (1) that it be inequitable to hold the individual liable for any unpaid tax or any deficiency, and (2) that “relief is not available to such individual under subsection (b) or (c)” (emphasis added).[17]

The plain language and structure of section 6015 strongly suggests that subsection (f) is a qualitatively different part of the statute. Subsections (b), (c), and (e) are all about deficiencies. They assume there is a deficiency at issue. If there is no deficiency, they simply do not apply; there can be no election. In short, they are not available. Only then does subsection (f) kick in with the “third” form of relief. Only subsection (f) contains the phrase “any unpaid tax.” That term is conspicuously absent from subsections (b), (c), and (e). Its absence is no accident but is part of a very deliberate compromise reached by the conference committee in reconciling the very different provisions proposed by the House Ways and Means Committee and the Senate Finance Committee in RRA 98. That same compromise, however, gave an unintended meaning to the placement of subsection (e) after subsections (b) and (c) but before subsection (f). While that placement might seem like a very strong support for the idea in subsection (e)(1) that only “an individual . . . who elects to have subsection (b) or (c) apply” can file a Tax Court petition, the legislative history shows that was not the original idea.

B. The Old Rules in Section 6013(e): 1971 and 1984 Congress enacted the first spousal relief provision in 1971 in section 6013(e) and made significant modifications to it in 1984. The history of that provision shows how and why it applied entirely to the collection of deficiencies and not to the collection of underpayments of self-reported taxes.

The most important aspect of the original section 6013(e) as it relates to the jurisdiction controversy in Bartman is that it was limited to deficiency situations, in concept, language, and intent. As to concept, the statute covered only omissions of income. An omission of income inherently results in a deficiency. Regarding language, the statute provided that the requesting spouse had to prove “it is inequitable to hold the other spouse liable for the deficiency in tax for such taxable year attributable to such omission.”[18] And as to intent, the 1971 Senate Finance Committee report explained that the bill provided no relief from joint liability from underpayments, but did give some relief from the fraud penalty in underpayment situations.[19]

The practical result of section 6013(e)’s limitation of relief to deficiency situations was that the Tax Court’s jurisdiction was commensurate with the relief Congress enacted. That is, because relief was limited to deficiency situations, spouses could raise the innocent spouse issue during the administrative examination process (including Appeals conferences) and when petitioning the Tax Court to redetermine a proposed deficiency. Accordingly, there was no need to give the Tax Court jurisdiction because section 6213(a) supplied all that was necessary. Section 6013(e) gave courts no independent jurisdiction over innocent spouse claims and taxpayers who tried to petition district courts for injunctions to restrain the assessment or collection of liabilities for which they claimed section 6013(e) relief were shut out of judicial review by section 7421, the Anti-Injunction Act.[20]

Congress amended section 6013(e) in 1984 to expand spousal relief from understatements caused by omitted income to understatements caused by erroneous deductions. [21] While Congress made some significant changes to the scope of relief, the relief still required both innocence and an understatement of income. As with the 1971 legislation, the concept, content, and intent of the 1984 legislation was to address deficiency situations. The statutory language regarding “liability for the deficiency” remained the same, still reflecting the statute’s sole concern with liability for deficiencies and not with underpayments of self-reported liabilities. The House Ways and Means Committee report also explained how the idea behind the statute was still limited to understatements of tax and not underpayments: the present . . . rules . . . are not sufficiently broad to encompass many cases where the innocent spouse deserves relief. Relief may be desirable, for example, where one spouse claims phony business deductions in order to avoid paying tax, and the other spouse has no reason to know that the deductions are phony and may be unaware that there are untaxed profits from the business which the other spouse has squandered.”[22]

Thus, the 1984 legislation did not disturb the congruence of relief and jurisdiction: Taxpayers who sought innocent spouse relief could do so by either prepayment, by contesting the deficiency in Tax Court, or postpayment, through the refund claim process. Because the relief was limited to deficiencies, the Tax Court’s deficiency jurisdiction in section 6213(a) was deemed sufficient. In fact, it was not sufficient, because in many cases the couple was still married at the time the deficiency was assessed and the first time a spouse would want relief from a liability would be during collection, long after the opportunity to obtain prepayment review in Tax Court had vanished.

C. The RRA 98 Reforms

Application of the 1984 reforms, particularly regarding the required innocence of knowledge, was widely criticized as arbitrary and unfair. Some commentators argued for reform of the innocence tests. Others advocated repeal of the joint and several liability rule itself (replacing it with a proportional liability rule). Finally, just as divorce statutes had moved from fault to no-fault regimes, so did some commentators argue that relief from joint liability should have nothing to do with fault but that any member of the marital unit should be able to “opt out” of what he or she had “opted into” when signing the return.[23]

In RRA 98, Congress tried again to make the joint liability regime fairer. But the House and Senate could not agree how. The original House bill and Senate amendment (which was in the nature of a substitution of the entire House bill with a completely new Senate version) were quite different from each other. An examination of their differences and how those differences were reconciled in conference will illustrate just where Congress goofed up the procedure.[24]

The House proposal was a straightforward reform of the innocent spouse provisions. While it created a new section 6015 to house the provisions, subsection (a) basically liberalized the requirements for innocent spouse relief, widened the scope of relief, and introduced the idea of partial relief if a spouse could show ignorance of some – but not all – of the erroneous items causing the understatement. But it still kept the idea of innocence, and it still related only to deficiencies, even crossreferencing the definition of “understatement” in section 6662(d)(2)(A), which parallels the definition of deficiency in section 6211.

The House proposal also contained a separate subsection, subsection (b), which unlike old section 6013(e), provided a jurisdictional basis for the Tax Court to review IRS treatment of innocent spouse claims independent of the deficiency process. Subsection (b) allowed “an individual who has filed a claim under subsection (a) within the period specified in subsection (a)(1)(E)” to “petition the Tax Court (and the Tax Court shall have jurisdiction) to determine such claim . . . .” Thus, a taxpayer who wanted relief under the new section 6015 did so by filing a “claim” and the Tax Court had jurisdiction to “determine such claim.”

The Senate proposal took an entirely different approach to the problem of joint and several liability. The Senate adopted the “opt out” model whereby a taxpayer, whether married or divorced, could elect to basically “undo” a previously filed joint return. Subsection (a) thus completely eliminated any requirement that the spouse be ignorant of anything. Instead, subsection (a)(1) allowed any individual to “elect the application of this section” to undo any joint return. It contained provisions that not only limited “the individual’s liability for any deficiency” but, more importantly, limited “the individual’s liability for any tax shown on the return.” In other words, the Senate’s subsection (a) offered relief from not just understatements but also any part of a self-reported liability, whether previously paid or not. Even if there was no underpayment, one of the signers of the joint return could come to the IRS later and say “undo it.” The payment status just did not matter; either way the electing spouse would be responsible for only his or her proportionate share of the liability. Subsection (a) contained further provisions placing the burden of proof on the electing spouse, creating a limitations period after which no election could be made and allowing the IRS to deny an election if the IRS could prove that the spouse “had actual knowledge of any item giving rise to a deficiency (or portion thereof) which is not allocable to such individual,” in which case the election would be ineffective as to that item. But the Senate version presumed no actual knowledge and burdened the IRS to rebut that presumption.

The next subsection of the Senate proposal, subsection (b), contained the general apportionment rules for the items reported on the return. Subsection (c) contained the general apportionment rules for “any deficiency on a mjoint return.” Subsection (d) was the Tax Court review subsection, and the language was different than the House version.

Like the House proposal, the Senate created a new jurisdiction provision, in its subsection (d), to enable the Tax Court to review the IRS administration of section 6015. Unlike the House provision, the Senate jurisdiction provision allowed “an individual who elects to have this section apply” to “petition the Tax Court (and the Tax Court shall have jurisdiction) to determine the appropriate relief under this section.” So, consistent with the use of “elect” language in subsection (a) relief provision, the Senate jurisdiction provision also used those terms. That was in contrast to the “claim” language used in the House bill’s subsections for relief and court review. Next came a most curious aspect of the Senate bill. The next subsection, (e), was titled “Equitable Relief.” It directed the IRS to create procedures to relieve taxpayers of liability even for underpayments (or deficiencies) properly attributable to the taxpayer seeking relief. And why should they be so excused? Because of innocence! The subsection on equitable relief in the Senate proposal brought back all the language about ignorance and the situational search for equity. What was the purpose of that subsection? Why did it come after the jurisdiction section? What was going on here?

The Senate Finance Committee’s report explains the curiosity this way:

If the electing spouse establishes that he or she did not know, and had no reason to know, of an item and, considering all the facts and circumstances, it is inequitable to hold the electing spouse responsible for any unpaid tax or deficiency attributable to such item, the item may be equitably reallocated to the other spouse.[25]

In other words, the electing spouse could invoke equitable relief under subsection (e) – and the IRS was supposed to apply it – as part of the election made to receive the benefits of the section. It was not a separate election. Because apportionment of underpayments was treated in one subsection and apportionment of deficiencies was treated in another, it was logical to put the equitable exceptions equally applicable to both sets of general rules into a separate subsection. So the reason why there was no separate “election” language for subsection (e) was that there was no need for it: It just applied as part of the election made by the taxpayer under subsections (b) or (c). It allowed the IRS to bend the otherwise applicable rules in those sections when strict application of those rules would be unfair, in the traditional innocent spouse meaning of unfair. If the Senate proposal had become law, it is difficult to imagine the government prevailing in an argument that the Tax Court did not have jurisdiction over a taxpayer’s claims that the IRS misapplied the equitable relief provisions of subsection (e). That is because those provisions, though stated in a separate subsection, were (as the Senate Finance Committee report indicates) part and parcel of the apportionment rules the IRS was supposed to apply in either the underpayment or deficiency situation. The structural placement of subsection (e) in the Senate bill did not affect the scope of the Tax Court’s jurisdiction in the prior subsection. One cannot infer from either the language or structure of the Senate proposal any intent in the Senate proposal to give the Tax Court only partial jurisdiction over IRS spousal relief provisions.

Now let’s look and see what happened in the conference compromise. The conference report summarizes five compromises made between the House and Senate conferees: (1) “The conference agreement follows the Senate with respect to deficiencies of a taxpayer who is [divorced or separated]”; (2) “The conference agreement also includes the provision in the House bill expanding the circumstances in which innocent spouse relief is available”; (3) “Taxpayers, whether or not eligible to make the separate liability election, may be granted innocent spouse relief where appropriate”; (4) “the conference agreement provides the Secretary with authority to grant equitable relief in appropriate situations”; and, finally, (5) “the conference agreement follows the House bill and the Senate amendment in establishing jurisdiction in the Tax Court over disputes arising in this area.”[26] I will look at each in turn.

Regarding the first compromise, the final version of the bill limited the Senate’s proposed proportional relief to spouses who were (a) divorced or something like it and (b) resisting a deficiency. So taxpayers would not simply have the option of opting out of just any joint return – including those that were fully paid as well as those that had an underpayment – as the Senate had proposed. That compromise nuked the “opt out” election for everyone else, including those who wished to avoid joint liability for underpayments.[27] The conference report speaks of that plainly: “The conference agreement does not include the portion of the Senate amendment that could provide relief in situations where tax was shown on a joint return, but not paid with the return.”28 That compromise is reflected in subsections (c) and (d) of section 6015 as enacted. Subsection (c) sets out the eligibility requirements and subsection (d) sets out the general apportionment rules.[29]

The second compromise simply adopted the House bill’s basic reforms to the old innocent spouse rules. Again, those rules applied only in cases of a deficiency. That compromise is reflected in subsection (b) of the enacted version and in Treas. reg. section 1.6015-2. The third compromise made it clear that taxpayers could get one or both types of relief for liability on deficiencies; the election of one type of relief did not preclude application for the other type of relief. That became subsection (a) as enacted.

The fourth compromise contains the goof. Recall that the Senate bill’s grant of equitable relief powers to the IRS was set out in a separate section following the jurisdiction section, but was not itself a separate election by the taxpayer. Instead it was simply a grant of power for the IRS to modify the otherwise applicable apportionment rules to do equity when the taxpayer had elected relief, either from a joint return or from a proposed deficiency. So under the Senate version, any taxpayer who made an election automatically got the benefit of the IRS equitable powers in the equitable relief subsection. Recall that the Finance Committee’s explanation for the grant of equitable powers was so that the IRS could alter the general allocation rules when necessary. Because taxpayers could make an election for both understatement liabilities and underpayment liabilities, the structure made sense. But with the elimination of the Senate’s relief provisions for underpayment liabilities, the equitable powers provision took on a whole new dimension. The conference committee report noted that while the two elections for relief were limited to deficiencies, the IRS’s grant of equitable powers was not so limited and could indeed include relief from underpayments.

The conferees intend that the Secretary will consider using the grant of authority to provide equitable relief in appropriate situations to avoid the inequitable treatment of spouses in such situations [underpayment of self-reported liabilities]. For example, the conferees intend that equitable relief be available to a spouse that does not know, and had no reason to know, that funds intended for the payment of tax were instead taken by the other spouse for such other spouse’s benefit.[30] In that way, the conference committee, which had deleted any potential underpayment relief from what

became subsections (b) and (c), brought it back and stuck it all into what became subsection (f). In so doing, Congress essentially turned subsection (f) into “innocent spouse relief from underpayments.”[31] What had been part and parcel of how the IRS was to administer an election for proportionate relief now became a freefloating command to “do the right thing” when subsections (b) and (c) were “not available.” No longer was it there to permit the IRS to bend the rules in subsections (b) and (c) when equity so demanded, it was now there apparently to allow the IRS to do equity when relief under (b) and (c) was not available because there was no deficiency.[32] The practical effect of doing that was to allow taxpayers to ask the IRS for innocent spouse relief from underpayments of tax when there was no deficiency. That is what the conference committee report’s example is about. Further, that is what is logically left after the conference committee had removed from the table the Senate proposal to allow all taxpayers to opt out of any joint return (whether fully paid or not). Was that compromise the dog or was it the tail? That, my dear reader, is the jurisdictional question in Bartman and is what the final compromise fails to address.

The last compromise was the jurisdiction compromise. In the House version, recall that Tax Court jurisdiction was over the “claim” of the taxpayer, but in the Senate version, it was over the “petition” of a taxpayer “who elects to have this section apply.” The statute as enacted uses the Senate’s language. But the same words now have a different scope. Under the Senate version, there was perfect congruence between the scope of who could petition (a taxpayer who elected to have “this section” apply) and the scope of the Tax Court’s jurisdiction (to determine the appropriate relief “under this section”). But in the final version, that congruence disappeared because the only “elections” allowed were those relating to relief from deficiencies. As noted, the equitable powers covered nondeficiency situations. In other words, the jurisdiction was limited to taxpayers who could make the elections offered by the statute. If the taxpayers could not make those elections, there was nothing to review. Is that what the taxwriters truly “intended”? One can make as plausible an argument either way. Either way, however, it goofed up the statute.

Thus it was that the creature of compromise spawned by the conference committee was a bit of a monster, and by its terms, denied jurisdiction to the Tax Court to fully review all IRS denials of spousal relief. Part III will look at how the Tax Court has tried to avoid that result and why its reasoning fails.

III. What All This Means to Ms. Bartman

Despite the language, structure, and history of section 6015, the Tax Court has held that it has jurisdiction to review denials of spousal relief when there is no deficiency at issue and the requesting spouse seeks relief from a previously reported but unpaid liability. Before getting to the specifics, let’s put the jurisdiction issue in context.

On one hand, regarding Article III courts, the Supreme Court has repeatedly emphasized the strong historical presumption of judicial review of administrative agency discretionary action. The presumption is grounded in the idea of preserving the effectiveness of Article III as a check on abuse of agency power. “[A] survey of our cases shows that judicial review of a final agency action by an aggrieved person will not be cut off unless there is a persuasive reason to believe that such was the purpose of Congress.”[33] Accordingly, while Congress can get rid of that check, it must do so with unambiguous, explicit, and precise statutory language.[34] Otherwise, only in the rare cases in which there is “no law to apply,” or in which court review would be little more than programmatic participation in agency administration will courts declare an agency action unreviewable.[35]

On the other hand, the presumption is reversed as to Article I courts in general, and the Tax Court in particular. Traditionally, an Article I court’s jurisdiction is carefully bounded by the language Congress uses in the governing statute.[36] “The Tax Court’s jurisdiction is defined and limited by Title 26 and it may not use general equitable powers to expand its jurisdictional grant beyond this limited Congressional authorization. It may exercise its authority only within its statutorily defined sphere.”37 The reason for the different treatment of Article III and Article I courts is unclear, at least to me, especially since the Supreme Court has decided that Article I courts are as much “Courts of Law” under the appointments clause in Art. II, section 2, cl. 2 as are Article III courts.[38]

The Tax Court has declared its ability to review IRS equitable relief determinations made under section 6015(f) in two cases: Fernandez in 2000 and Ewing in 2002.[39] The Fernandez opinion stands for the proposition that, so long as the Tax Court has jurisdiction over a petition, it can review the IRS decision to grant equitable relief under section 6015(f). When Fernandez was decided, section 6015(e) read like this, in relevant part:

(1) In General. – In the case of an individual who elects to have subsection (b) or (c) apply –
(A) In General.-the individual may petition the Tax Court (and the Tax Court shall have jurisdiction) to determine the appropriate relief available to the individual under this section.

The taxpayer in Fernandez requested relief from a deficiency.[40] She requested relief under subsections (b),
(c), and (f) and was denied on all counts. She petitioned the Tax Court for review and the IRS argued that the Tax Court could review only the denials under subsections
(b) and (c) because paragraph (1) limited court review, by its terms, to elections of subsection (b) or (c). Special Trial Judge Panuthos properly rejected that reasoning, pointing out that the language in paragraph (1)(A) gave the Court the authority to “determine the appropriate relief . . . under this section” and the term “this section” included subsection (f).

The Fernandez opinion is correct and well within precedent. The Tax Court’s statutory authority to act on a petition properly before it may be broader than its jurisdictional authority. Small doors can lead to large rooms. For example, the Tax Court does not have jurisdiction to hear claims for refunds of overpayments, but if a taxpayer can squeeze through the section 6213(a) door to deficiency jurisdiction, then the Tax Court can determine an overpayment for the year before the Court and can order the IRS to refund it.41 Similarly, although the Tax Court lacks equitable powers to take initial jurisdiction over a case, it does have a surprisingly broad array of equitable powers once a case is otherwise properly before it.[42] Recall that section 6015(e) gives the Tax Court jurisdiction to hear “stand-alone” petitions – petitions that are not made in the course of a timely contest to a notice of deficiency. Once the Tax Court properly has jurisdiction over a taxpayer’s stand-alone petition, the expansive language in subsection (e)(1)(A) authorizes the court to determine the correctness of the IRS decision under all subsections of section 6015.

In contrast to Fernandez, the taxpayer in Ewing did not contest a deficiency but instead asked for relief from liability for an underpayment of self-reported tax. Nonetheless, Judge Ruwe wrote for the Tax Court majority that the Tax Court could take jurisdiction.[43] His basic idea was that the statutory language “in this section” trumped the preceding language that started the paragraph.

Judge Ruwe grounded his opinion on the legislative history to section 6015. According to Judge Ruwe, both the House and Senate bills intended for Tax Court review to be commensurate with the relief provided in the sections, so the compromise patched together by the conference committee must also have had the same intent as evidenced by the conference committee’s use of the term “this area” in its initial description of the jurisdictional compromise that “the conference agreement follows the House bill and the Senate amendment in establishing jurisdiction in the Tax Court over disputes arising in this area.”[44] In so reasoning, Judge Ruwe followed the noble tradition of a court refusing to follow the plain implications of statutory language when to do so would conflict with the general purpose of the statute.[45]

Unfortunately, the true history of the legislation is more than one sentence in a conference committee report. There is simply not enough in the history of this statute to trump the clear meaning of the restrictive language that starts paragraph (1), even if one is inclined to use a statute’s remedial purpose to interpret a statute contrary to its language. The restrictive language of paragraph (1) trumps the expansive language in subparagraph (A), regardless of whether the taxpayer actually made an election for relief under subsection (b) or (c). That is because, as per the legislative history above, the language that starts paragraph (1) was simply the combination of the House and Senate proposals. Each of the two proposals gave jurisdiction to the taxpayer’s invocation of all the relief provided by the proposal, but the final compromise to bifurcate relief, with deficiencies being covered by

subsections (b) and (c) and underpayments being covered by subsection (f), cut the latter subsection off from its former relation to the deficiency relief sections. So while the equitable relief provisions fit into the election made by the taxpayer in the Senate proposal, Congress failed to modify section 6015(e) to reflect the compromise creating subsection (f) as, essentially, a third choice for taxpayers in the nature of “innocent spouse relief from underpayments.”

While both the House and Senate versions of section 6015 made Tax Court review commensurate with IRS actions, there is no compelling reason why, in the final product, Tax Court review had to still be as extensive as IRS authority. That is, while I suggest that the design was a goof I do not suggest the taxwriters did not act deliberately. For example, the IRS has historically had significantly autonomous powers of collection.[46] Particularly as to collection issues, Congress has been historically reluctant to subject IRS collection action to court review.[47] During enactment of RRA 98 the IRS strongly resisted, at every possible turn, imposition of court review. Innocent spouse relief was viewed as a collection matter and the taxwriters repeatedly stated that deficiency relief was the major problem taxpayers had in collection (not just innocent spouses, but all taxpayers).[48] The same conferees who limited subsections (b) and (c) to deficiencies and moved underpayment relief to subsection (f) could well have deliberately decided to give the IRS administrative authority to act in a class of cases without court review. The problem is that there is just no evidence in the legislative history of RRA 98 to suggest either that Congress meant to give the Tax Court jurisdiction over IRS decisions about underpayments or that Congress meant to deny that jurisdiction.

Further undercutting Judge Ruwe’s reliance on the RRA 98 legislative history are the amendments Congress made to section 6015(e) after Fernandez but before Ewing. The amendments, labeled “technical amendments,” became effective on December 21, 2000, and inserted the following italicized language into section 6015(e):

(1) In General.-In the case of an individual against whom a deficiency has been asserted and who elects to have subsection (b) or (c) apply –

(A) In General. – In addition to any other remedy provided by law, the individual may petition the Tax Court (and the Tax Court shall have jurisdiction) to determine the appropriate relief available to the individual under this section.

As technical amendments, the language does not change the prior meaning of the statute. But it does clarify the need for a deficiency.[49] That is why, even though the government’s brief in Bartman incorrectly claims that the language was “effective as if originally enacted,” that mistaken claim is harmless.[50]

Judge Ruwe denied that the amendment affected his earlier analysis. He pointed out that the amendment simply clarified the timing of petitions. That is, the original statute had put the outer limit on when a taxpayer could submit an administrative claim for spousal relief at two years after the IRS starts collection action against the taxpayer for a deficiency. But what was the earliest time the taxpayer could submit a claim? Did the taxpayer have to wait until the IRS sent a notice of deficiency or even until after assessment of a deficiency? As Judge Ruwe properly noted, the conference committee’s report explained that the new language allowed a taxpayer to elect spousal relief as soon as “the IRS states that additional taxes may be owed. Most commonly, that occurs during the examination process. It does not require an assessment to have been made, nor does it require the exhaustion of administrative remedies in order for a taxpayer to be permitted to request innocent spouse relief.”[51]

What Judge Ruwe failed to consider is that everything in the conference committee report spoke about the proper timing during the deficiency determination process. In other words, the amendment was certainly all about timing but was also all about deficiencies. So it simply reinforces the idea that the elections in subsection (b) and (c) are also all about deficiencies. Taxpayers simply are not eligible to elect relief under subsections (b) and (c) unless they are trying to avoid a deficiency.

The taxpayer’s reply brief in Bartman makes a big deal out of the language “in addition to any other remedy provided by law.” That part of the brief appears to suggest that this broad language overcomes the limitation in the initial language of the paragraph that limits court petitions (and hence jurisdiction) to deficiency situations. The argument fails at the textual level. Just as the subordinate language “in this section” cannot overcome the primary language that starts paragraph (1), neither can the subordinate language relied on by the taxpayer. Still, one may ask what other purpose the language could have. To me, the language “any other remedy” logically refers to the petition to redetermine a deficiency authorized by section 6213(a). Without the additional language in section 6015, some misguided IRS attorney might think that a taxpayer who raised an innocent spouse issue before receiving a notice of deficiency would later be precluded from resisting the notice of deficiency in the Tax Court. While the doctrines of claim preclusion and issue preclusion would most likely enable a court to reach the right result, the added language helps clarify the proper outcome. Choosing that door into the Tax Court does not necessarily lock up any other door. But the language adds nothing to the Tax Court’s jurisdiction.

The government’s second argument in Bartman also bears some comment. The government claims that even if the taxpayer is seeking relief from a deficiency, if the taxpayer does not “elect to have subsection (b) or (c) apply,” the Tax Court cannot obtain jurisdiction. In other words, the government claims that only taxpayers who recite the magic words “I want relief under subsection (b) or (c)” can petition the Tax Court. Oh please. Let’s not descend into pettifoggery. That reading should be rejected for the specious elevation of form over substance that it is.

The better reading of statutory election language is that court review is limited to the class of taxpayers who come within the scope of relief afforded by Congress in subsections (b) or (c) – that is, relief from deficiencies – whether because they are innocent spouses under (b) or because they are now sufficiently separated to invoke proportional relief under (c). But they must still come within the scope of the relief provided by subsections (b) and (c). So they must be resisting a deficiency. If they are, the Tax Court has jurisdiction to review the IRS denial of relief and has authority to determine the proper relief for the taxpayer under the entire section, including equitable relief in subsection (f). But if the taxpayer is not resisting a deficiency but is instead seeking relief as to an underpayment of self-reported tax, even if the taxpayer goes through the formal motion of “electing” relief under subsection (b) or (c), the Tax Court has no jurisdiction – at least until Congress writes a better statute. In other words, Congress wrote section 6015 to afford relief to a certain class of taxpayers in subsections (b) and (c) (those being pursued for a deficiency) and shoved all other taxpayers into subsection (f) equitable relief. The relief in subsections (b) and (c) relates to relief from deficiencies. Just as a taxpayer cannot elect to participate in the deficiency process absent a deficiency, neither can a taxpayer elect relief from the deficiency when there is no deficiency. At the same time, a taxpayer who seeks spousal relief during the deficiency determination or collection process is necessarily “electing” relief under either subsection (a) or (b) because that is their purpose. Judge Ruwe is correct to hold in Ewing that “in every case where the taxpayer submits a request to the Commissioner for relief under section 6015, and such request includes a claim for relief under section 6015(f), the Commissioner must first examine both subsection (b) and (c) to determine whether relief is available under those subsections before determining whether relief is available under section 6015(f).”[52]

That better reading of the statute is analogous to how courts determine refund jurisdiction. A court may not take jurisdiction over a refund suit unless the taxpayer has filed a timely administrative claim for relief.[53] The law there has properly refused to elevate form over substance. Therefore, although the Treasury regulations require taxpayers to file their administrative claims for refund of overpaid income taxes on a valid original return or an amended return (a Form 1040X), neither the IRS nor the courts hold taxpayers to the formal requirements. Instead, “[w]hat is essential is that the taxpayer must inform the Internal Revenue Service that a claim for a refund is being asserted, and must provide enough information so that the IRS can adequately examine the merits of the claim.”[54] So the initial jurisdictional question in refund suits is not a search for magic words on a magic form. It is instead a substantive search for meaning because a valid claim for refund may be asserted in a variety of ways, even on a series of documents taken together.[55] The “elect” language in section 6015(e) should be similarly construed. Taxpayers not contesting a deficiency cannot, by definition, elect relief under subsections (b) or (c) and thus cannot, by mouthing the magic words, confer jurisdiction on the Tax Court to review denial of relief under subsection (f). Similarly, taxpayers contesting a deficiency cannot deprive the Tax Court of jurisdiction by failing to mention subsections (b) or (c) in their request for relief from a deficiency. Such an “election” is an inherent part of every request for spousal relief.

[1]Pub. L. 105-206, 112 Stat. 685 (1998).

[2]I take all my facts from the Tax Court opinion unless otherwise indicated.

[3]Reply Brief for Appellant at 15.

[4]For the 1994 tax year, Ms. Bartman sought relief under section 6015(b), (c), and (f) because there was a deficiency for that year, but she sought relief only under section 6015(f) for 1995 and 1996 years. See Brief for the Appellee at 11, 18. The government raises Ms. Bartman’s failure to ask for 6015(b) and (c) relief as an independent ground for depriving the Tax Court of jurisdiction for the 1995 tax year. Id. at 17. I shall explain below how the government misreads the law on that point; what relief a taxpayer actually requests is irrelevant to the jurisdictional issue.

[5]Brief for the Appellant at 2. The appeals officer’s initial decision was based on an issue that the IRS later conceded in another case. The IRS eventually raised the section 6511 issue as an alternative theory of why it was without authority to return what it conceded was money that Ms. Bartman had paid on a liability she did not owe.


[7]The various section 6511 limitation periods form a nasty little maze of rules. And they sometimes deny refunds to deserving taxpayers. But that is inherent in all limitation periods, which form the squarest corners of all laws. For a reasonably good treatment of refund rules in general, and the section 6511 rules in particular, I modestly point you to my chapters (10 and 12) in the looseleaf treatise Federal Tax Practice and Procedure (Lederman and Murphy, Eds.) (LEXIS/NEXIS 2005).

[8]Preferred Risk Mutual Ins. Co. v. United States, 86 F.3d 789, 793 (8th Cir. 1996) (citing cases). For some reason, the government’s brief fails to cite Eighth Circuit authority for this proposition but instead cites a Ninth Circuit case. See Brief for the Appellee at 19. Similarly, courts can raise the issue sua sponte. See Ewing, 118 T.C. 494, Doc 2002-13179, 2002 TNT 106-9 (2002), where the Tax Court raised and resolved the issue itself even though both parties had agreed on jurisdiction in their pleadings.

[9]Fernandez v. Commissioner, 114 T.C. 324, Doc 2000-13039, 2000 TNT 92-10 (2000) (exercising jurisdiction to review section 6015(f) determination when taxpayer was seeking relief from a deficiency); compare AOD 2000-06, 2000-12 IRB 22, Doc 2000- 25046, 2000 TNT 190-14 (acquiescence) with AOD 2004-05, 2004-35 IRB 349, Doc 2004-17329, 2004 TNT 171-5 (acquiescence in result only).

[10]Even fewer involve taxpayers who, like Ms. Bartman, are represented by counsel (2 of the 8 in 2003 and 2 of the 4 in 2004).

[11]118 T.C. 494 and 122 T.C. 32 (2004), appeals docketed Nos. 04-73237 and 04-73699 (9th Cir. July 27, 2004).

[12]The usual metaphor is that the notice of deficiency is the “ticket to the Tax Court.” For a thorough explanation of its multiple functions, see generally Leandra Lederman, “Civilizing Tax Procedure: Applying General Federal Learning to Statutory Notices of Deficiency,” 30 U.C. Davis L. Rev. 183 (Fall 1996).

[13]See Butler v. Commissioner, 114 T.C. 276 Doc 2000-12209, 2000 TNT 84-62 (2000) (where taxpayer had petitioned court in response to a notice of deficiency, court could consider taxpayer’s section 6015(f) claim for equitable relief under its section 6213(a) jurisdiction). That was also true under the predecessor statute, section 6013(e). See, e.g., Krampf v. Commissioner, T.C. Memo. 1983-382 (court had jurisdiction to consider section 6013(e) defense but would not because petitioner had not properly raised the defense in her pleadings in violation of Tax Court Rule 39).

[14]Section 6211.

[15]See Block v. Commissioner, 120 T.C. 60, Doc 2003-2307, 2003 TNT 16-9 (2003) (denying taxpayer’s motion to amend petition to plead issue regarding statute of limitations when taxpayer had filed a stand-alone petition, because “[p]etitioner’s amendment would allow her to go beyond the specific relief contemplated by section 6015 and questions the viability of the tax liabilities from which she seeks relief.”).

[16]See, e.g., Block v. Commissioner, supra note 15, at 65 (“Section 6015 provides qualifying taxpayers with three distinct avenues of relief.”); Sandy Horowitz, “Tax Court Jurisdiction and Equitable Relief Under Section 6015(f) and 66(c),” Doc 2003-12001, 2003 TNT 94-129 (May 15, 2003) (“Congress made relief from joint and several liability more readily available with three avenues of relief: (1) innocent spouse relief under section 6015(b); (2) separation of liability under section 6015(c); and (3) equitable relief under section 6015(f)”). I explain further in Part II.C. below

[17]Subsection (g) concerns credits and subsection (h) authorizes regulations regarding nonrequesting spouses and are not of immediate concern here. The IRS has used its regulatory authority to give taxpayers the ability to request relief directly under section 6015(f) without having to elect relief under either subsection (b) or (c). Treas. reg. section 1.6015-4; see also Rev. Proc. 2000-15, 2001-1 C.B. 447, Doc 2000-2048, 2000 TNT 12-4. As I explain below, that is in keeping with the severing of subsection (f) relief from subsections (b) and (c) relief by the conference committee.

[18]Section 6013(e)(1)(C) (emphasis added), quoted in Kirtley v. United States, 488 F.2d 768 (10th Cir. 1973) (holding that section 6013 provided no independent grounds for jurisdiction).

[19]S. Rep. 91-1537, 1970 USCCAN at 6092 (“The bill also amends the provision imposing a 50 percent penalty when the underpayment is due to fraud (sec. 6653(b) of the code). This relief would apply, for example, where the underpayment resulted from the fraudulent deductions (rather than an omission from gross income) – an example of a situation in which no relief is provided the spouse for the tax liability as such.”) (emphasis added).

[20]Kirtley v. United States, 488 F.2d 768 (10th Cir. 1973).

[21]It was part of the Deficit Reduction Act of 1984, Pub. L. 98-369, 98 Stat. 494.

[22]H. Rep. (Ways and Means Committee) No. 98-432, Oct. 21, 1983, reprinted in 1984 USCCAN 697, 1143. There was no corresponding Senate provision and the conference committee report states that the final bill adopted the House provision. Id. at 1807.

[23]This paragraph draws from Richard C. E. Beck, “The Innocent Spouse Problem: Joint and Several Liability for Income Taxes Should Be Repealed,” 43 Vand. L. Rev. 317 (1990); Note, “The Bonds of Joint Tax Liability Should Not be Stronger Than Marriage: Congressional Intent Behind Section 6015(c) Separation of Liability Relief,” 78 Wash. L. Rev. 831 (2003); Note, “Do You, Elizabeth, Promise to Pay John’s Taxes? I Do: A Review of the Innocent Spouse Provisions and a Proposal for Change,” 1996 Utah L. Rev. 1065.

[24]The text of all versions of RRA 98 can be found at http:// (last visited June 18, 2005).

[25]S. Rep. 105-174 at 57.

[26]H. Rep. 105-599 at 251.

[27]My take on the reasons for that change by the conference committee was that the IRS prevailed in its argument that to allow all divorced taxpayers to make the proportionate election would create an administrative nightmare for the IRS. If one reads the Senate proposal carefully, one sees that the proposal allows everyone to elect proportional relief from the moment they become divorced until “not later than 2 years after the date on which the Secretary has begun collection activities.” That is, the two-year limitation period was the outside limit on the election. There were almost 950,000 divorces in 1998 alone. See (last visited June 18, 2005). No divorce attorney would risk malpractice by failing to suggest to a client that he or she consider electing the proportionate relief. There were tens of millions more taxpayers who would be eligible to make the election the day the bill became law. Faced with that, the conference committee modified the statute so that even divorced spouses could not make the election until after the IRS started to come after them. That, at least, is my take on it and I could well be wrong since I was not a fly on the wall in the conference. But I was part of the IRS team assigned to provide technical support to Treasury’s negotiator, Chris Rizek. So I know that was a huge IRS concern and I know it was discussed at least at the staff level.

[28]H. Rep. 105-599 at 254.

[29]The regulations likewise restrict election of proportionate relief to deficiencies. See Treas. reg. section 1.6015-3 (“Allocation of deficiency for individuals who are no longer married, are legally separated, or are not members of the same household”).

[30]H. Rep. 105-599 at 254.

[31]Note that is still a rejection of the Senate’s idea to allow every taxpayer (even those still married) the ability to undo any joint return (even those fully paid). Instead, it allows the IRS to grant equitable relief similar to innocent spouse relief to taxpayers when there is an underpayment.

[32]This is how the IRS has so treated it in the regulations. Under the regulations, a taxpayer can make either of the two elections or else make a “request” for equitable relief. Treas. reg. section 1.6015-1(a)(1) (“a spouse or former spouse may be relieved of joint and several liability for Federal income tax for that year under . . . three relief provisions”) (emphasis added). That is, instead of the taxpayer being given two choices for relief and the equitable powers going with each choice – the IRS interprets the statute as giving taxpayers three choices for relief. There is no functional difference in the regulations between the labels “election” and “request.” See Treas. reg. section 1.6015- 1(h)(1). Two of those choices carry with them the opportunity for court review; the third does not.

[33]Abbot Laboratories v. Garner, 387 U.S. 136, 140-141 (1967).

[34]It is questionable whether Congress can deny jurisdiction to an Article III court over constitutional claims. See Johnson v. Robison, 415 U.S. 361 (1974) (statute precluding judicial review of any Veterans Administration decision arising “under any law administered by the [VA]” did not preclude courts from considering constitutional claim because the VAdid not administer the constitution).

[35]Heckler v. Chaney, 470 U.S. 821 (1985) (analogizing regulatory discretion to prosecutorial discretion in rejecting plaintiffs’ petition to compel FDA to enforce regulatory statutes against states for using drugs for lethal injections when the drugs had not been approved for that use); Lincoln v. Vigil, 508 U.S. 182 (1993) (refusing to interfere with agency choice on which agency programs to fund out of lump-sum appropriation).

[36]United States v. Testan, 424 U.S. 392 (1976) (emphasizing the limited nature of Court of Federal Claims); Commissioner v. McCoy, 484 U.S. 3 (1987) (Court of Appeals limited by Tax Court’s limited jurisdiction in reviewing the decision of the Tax Court).

[37]Estate of Branson, 264 F.3d 904, 908, Doc 2001-23234, 2001 TNT 174-6 (9th Cir. 2001). See section 7442 (“the Tax Court and its divisions shall have such jurisdiction as is conferred on them by this title”).

[38]Freytag, 501 U.S. 868 (1991). See Leandra Lederman, “Equity and the Article I Court: Is the Tax Court’s Exercise of Equitable Powers Constitutional?” 5 Fla. Tax. Rev. 357 (2001).

[39]Fernandez v. Commissioner, 114 T.C. 324 (2000); and Ewing v. Commissioner, 118 T.C. 494 (2002). The Tax Court also decided in Ewing that it reviewed IRS spousal liability decisions for abuse of discretion on a de novo basis. I will have more to say about that in my next column.

[40]So says the court in Ewing, 118 T.C. at 500.

[41]Section 6512.

[42]Kelly v. Commissioner, 45 F.3d 348, 351 (9th Cir. 1995) (“while [the Tax Court] cannot act, equitably or otherwise, in a case over which it lacks or has lost jurisdiction, the Tax Court can act equitably in a case in which it has jurisdiction”).

[43]Judge Laro wrote a dissent, joined by Judges Whalen and Foley, in which he hit the jurisdictional nail squarely on the head, although with a different emphasis and analysis than mine.

[44]H. Rep. 105-599 at 251.

[45]Church of the Holy Trinity v. United States, 143 U.S. 457 (1892); see generally Kent Greenawalt, Statutory Interpretation: 20 Questions (Foundation Press 1999) at 48-53

[46]Bryan T. Camp, “The Inquisitorial Nature of Tax Administration and the Partial Paradigm Shift in the IRS Restructuring and Reform Act of 1998,” 56 Fla. L. Rev. 1, 20-31.


[48]Id. at 78-132 (analyzing general concerns underpinning the entire RRA 98 statute). In my next column I hope to show how it should be viewed as a liability issue and not a collection issue

[49]Of course, the source of the amendment is most likely the IRS, via Treasury, who wanted the language precisely because it clarified the law. If even a single member of Congress knew what was in this one technical amendment buried in an 800-page appropriations bill, I would be surprised.

[50]The government makes the claim on page 20 of its brief. The new language for section 6501(e) comes from subsection (a) of section 313 of the Community Renewal Tax Relief Act of 2000, which was passed as Appendix G to the Consolidated Appropriations Act, 2001, Pub. L. 106-554, 114 Stat. 2763, and is found on pages 2763A-587 through 2763A-651. Section 313(f) of the act, found at 114 Stat. 2763A-641, provides “(f) EFFECTIVE DATES. The amendments made by subsections (a) and (b) shall take effect on the date of the enactment of this Act [December 21, 2000]. The amendments made by subsections (c), (d), and (e) shall take effect as if included in the provisions of the Internal Revenue Service Restructuring and Reform Act of 1998 to which they relate.” CCH gets it right in its publication. The government just misread the act by ignoring the bolded language and instead applied the subsections (c), (d), and (e) effective dates to subsection (a). Besides, Ms. Bartman filed her petition well after the statute’s true effective date.

[51]H. Rep. 106-1033 at 1023. Of course, the statement that “nor does it require the exhaustion of administrative remedies” is quite confusing. To the extent it suggests that a taxpayer can simply blow off the IRS and file a petition directly with the Tax Court, it runs smack into the language in section 6015(e)(1)(A)(i), which seems to require a taxpayer to ask the IRS first.

[52]118 T.C. at 497-498.

[53]Section 7422(a) (prohibiting suits against the United States until after a “claim for refund has been duly filed with the Secretary, according to the provisions of law in the respect.” See U.S. v. Dalm, 494 U.S. 602 (1990); Mutual Assur. Inc. v. U.S., 56 F.3d 1353, 1356 (11th Cir. 1995) (filing an administrative claim “is a juridictional prerequisite to the maintenance of a tax refund suit”).

[54]Evans v. U.S., 618 F. Supp. 621, 622-23 (E.D. Pa. 1985), aff’d 787 F.2d 581 (3d Cir. 1986) (citations omitted).

[55]U.S. v. Kales, 314 U.S. 186 (1941) (court had jurisdiction because taxpayer’s letter was a valid claim); PPG Industries v. U.S., 94-2 USCT para. 50,617 (N.D. Pa. 1994) (court had jurisdiction because taxpayer’s two letters and numerous conversations with the IRS constituted a valid refund claim taken together).

George H. Mahon Professor of Law, Texas Tech University School of Law