In The Death of Tax Court Exceptionalism, 99 Minn. L. Rev. 221 (2014), we don’t tackle the part of Chenery raised by the petitioners in Ax—that agency reasons must have been raised at the agency decision-making level and not for the first time in court. But we do tackle another core principle from the Chenery decisions: the ordinary remand rule. That is, when a court concludes that an agency’s decision is erroneous, the ordinary rule is to remand to the agency to consider the issue anew (as opposed to the court deciding the issue itself).

These two Chenery principles are intertwined, and the analysis used to determine whether one applies in a given situation is arguably the same for either. Here’s what we had to say about the remand rule and deficiency actions (at 266-67, footnotes omitted):

This [ordinary remand rule] does not mean, however, that the Tax Court must remand every erroneous IRS determination to the IRS for reconsideration. As the ordinary remand rule suggests, it is the “ordinary” rule subject to exceptions for “rare circumstances.” These rare circumstances include when there are minor errors as to subsidiary issues that do not affect the agency’s ultimate decision or when the agency lacks authority to decide the issue. And there is another exception of particular relevance here: when APA § 706(2)(F) applies and “the facts are subject to trial de novo by the reviewing court.” In particular, in explicating the ordinary remand rule and how it is “the proper course, except in rare circumstances,” the Supreme Court has noted that “[t]he reviewing court is not generally empowered to conduct a de novo inquiry into the matter being reviewed and to reach its own conclusions based on such an inquiry.” Logically, then, if the reviewing court is empowered to conduct a trial de novo, the court is not required to remand (though it retains discretion to do so) because de novo review allows the court to take the unusual step of substituting its judgment for that of the agency.

In light of these principles, it becomes clear that the Tax Court need not remand when it conducts a trial de novo. As explained in Part II.A, the Tax Court conducts a trial de novo of an IRS tax deficiency determination in any proceeding in which the underlying deficiency determination is challenged. (That said, as discussed in Part III, in some circumstances policy considerations may weigh in favor of the Tax Court exercising its discretion to remand even some de novo redeterminations to the IRS.) But when the trial de novo provisions of APA § 706(2)(F) do not apply, the Tax Court should adhere to the ordinary remand rule.

In other words, as Les notes in his post, the Tax Court seems to have gotten the outcome right (that Chenery does not apply in deficiency actions in Tax Court), but that’s not because the APA does not apply. That’s the result you get after applying the APA.

In fact, the Tax Court actually did apply the APA in Ax, although it stopped short of acknowledging that it is a reviewing court for purposes of the APA. In particular, the Court relied on APA § 703, which describes the form and venue of judicial review of an agency action, to support the position that it is not bound by APA § 706 in deficiency actions. Section 703 provides that the appropriate form of judicial review of an agency action is “the special statutory review proceeding relevant to the subject matter in a court specified by statute,” or, if none, “any applicable form of legal action.” The Court reasoned that because IRC § 6213 and other sections mentioned by Les describe a “special statutory review proceeding,” APA § 706, which provides procedural defaults, does not apply to redeterminations of deficiency. (As discussed above and as we flesh out more fully in our paper, our position is that APA § 706 does apply to deficiency actions, but that deficiency actions are “trial de novo” proceedings under APA § 706(2)(F), and thus Chenery’s limitations do not apply.)

Notably, although the Court mentions O’Dwyer’soverly broad holding that the Tax Court is not subject to the APA, its reference to the greying case is almost a throwaway in light of the Court’s APA analysis. The Court takes pains to note (at 18) that its decision to look beyond the IRS’s notice of deficiency “does not violate administrative law principles or conventional standards of judicial review.” It is not clear what the court means by “conventional standards of judicial review,” but its later discussion of Mayo indicates that it is working within the framework provided by the APA (albeit within an APA-provided exception to the default provisions on judicial review). There, the Court notes (at 19) that Mayo did not consign judicial review of IRS actions to a “single one-size-fits-all paradigm for standard and scope of review,” but rather preserved (indeed, in our view, mandated) the application of APA § 703 (the Court cites § 704 here, but it is clearly a typographical error).

In its conclusion, the Court carefully notes (at 20) that its decision “in a deficiency case” is not at odds with either Chenery or APA § 706. The Court’s limitation of its reasoning to deficiency cases suggests that the Court may be rethinking the role of the APA in its review of IRS decision-making. Although the Court’s position on the issue of deficiency is entrenched, there is plenty in the law to suggest that its stance is not in error, and while Ax reads like an anti-APA decision at first blush, the Court’s careful use of APA § 703 in relation to § 706 perhaps foreshadows a different result for petitions related to matters other than deficiency. In short, nothing in Ax precludes the Court from moving away from exceptionalism and toward normalization with its sister courts though full-fledged adoption of the APA provisions on judicial review.

At the outset, it is important to underscore that Smith’s worry is already a reality. As we explain in a forthcoming paper, the Tax Court has declared that “[t]he APA has never governed proceedings in the Court (or in the Board of Tax Appeals).” In other words, the Tax Court refuses to apply the APA’s default standard (abuse of discretion) and scope (administrative record) of review. Instead, it considers the default in both contexts to be de novo, and only departs from de novo review when it determines that the Internal Revenue Code suggests more deferential review.

This approach contradicts the text of the APA. As we explain in Part II of our paper, the APA establishes the default standards for judicial review of federal agency action. The APA defines “agency” as “each authority of the Government of the United States, whether or not it is within or subject to review by another agency, but does not include” among other entities Congress and “the courts of the United States” (5 U.S.C. § 701(b)(1)).   The APA judicial review standards apply to any “reviewing court” of agency action (5 U.S.C. § 706). The IRS, an executive agency within the Treasury Department, is plainly an “agency” for purposes of the APA. And while the Tax Court used to be an agency before the enactment of the APA, as of 1969 it is a “court of record” “established[] under article I of the Constitution of the United States” (26 U.S.C. § 7441). For purposes of the APA, it is thus “a court of the United States” (5 U.S.C. § 702; accord 5 U.S.C. § 701(b)(1)) and, for its review of IRS agency actions, a “reviewing court” subject to the APA’s judicial review provisions (5 U.S.C. § 706).

Although the statutory language appears plain, an argument could be advanced that only Article III courts—not Article I courts—can be “reviewing courts” for purposes of the APA. After all, the Constitution vests “judicial power” in life-tenured Article III judges, not term-limited Article I judges. The text of the APA, however, makes no distinction between Article I and Article III courts, and the APA governs other Article I courts—including the Court of Federal Claims—when they review agency action.

The Supreme Court ‘s decision in Freytag is illustrative. There, the Court rejected an Appointments Clause challenge to the Tax Court Chief Judge’s ability to appoint special trial tax judges. The Appointments Clause allows Congress to vest appointment powers of such inferior officers in one of three entities: “in the President alone, in the Courts of Law, or in the Heads of Departments” (U.S. Const., Art. II, § 2, cl. 2). The parties in Freytag advanced competing interpretations: the Tax Court is an executive department where the Chief Judge as Department Head has power to appoint inferior officers or, alternatively, neither an executive department nor a court and thus constitutionally forbidden from appointing inferior officers. The Freytag Court rejected both positions and instead held that the Tax Court is a “Court[] of Law” under the Appointments Clause. Importantly, the Court noted that “[t]he text of the Clause does not limit the ‘Courts of Law’ to those courts established under Article III of the Constitution” and that “[t]he Tax Court exercises judicial, rather than executive, legislative, or administrative, power.”

Similarly, although Kuretski’s conclusion (at 20) that the Tax Court “exercises its authority as part of the Executive Branch” may be in tension with Freytag, it should have no bearing on the APA’s definitions of “agency” and “reviewing court.” Indeed, Kuretski’s reasoning suggests as much (at 20-21): “[W]e conclude that the Tax Court’s status as a ‘Court of Law’—and its exercise of ‘judicial power’—for Appointments Clause purposes under Freytag casts no doubt on the constitutionality of the President’s authority to remove Tax Court judges.” As a matter of statutory interpretation, Congress can define “court” differently for APA purposes than the Constitution does for appointment or removal purposes. “After all,” as Hickman points out, “if the Affordable Care Act penalty can be a tax for constitutional purposes and yet not a tax under § 7421(a) of the Internal Revenue Code (see NFIB v. Sebelius), then the Tax Court surely can bear varying labels for different purposes as well.”

If there were any lingering doubts about the impact of Kuretski on the Tax Court’s relationship to the APA, Judge Srinivasan puts them to rest—albeit in dicta—near the end of the opinion. Citing a district court decision that held that the Tax Court is a “court of the United States” and not an “agency” for purposes of the APA, he notes (at 26) that “Congress, in establishing those entities [the Tax Court and the Court of Appeals for the Armed Forces] as a ‘court’ rather than an ‘agency,’ perhaps also exempted them from statutes that apply solely to executive ‘agencies.’”

Indeed, the Tax Court’s concerns may be heightened in the context of less sophisticated taxpayers who attempt to navigate the system without representation.  The Tax Court’s current de novo approach may appear to best protect the unrepresented taxpayer.  In contrast, by confining review to abuse of discretion and prohibiting consideration of evidence outside of the administrative record, the APA limits a court’s ability to grant relief when it feels such relief may be merited.  Moreover, by following traditional administrative law doctrines, the Tax Court may be required to remand cases to the IRS when it finds error instead of just granting the relief outright.  The Tax Court may be reluctant to adhere to this ordinary remand rule—indeed, the Tax Court currently does not remand matters except in narrow circumstances—when it believes the taxpayer is entitled to relief and remand would unduly delay or, worse, preclude relief because the taxpayer would get lost in the process on remand.

Despite the common-sense appeal of more searching review, in Part III of our paper we advance a novel and somewhat counterintuitive argument:  the APA’s default regime is actually more effective at addressing these concerns by allowing the Tax Court to have a more systemic effect on IRS decision-making.  In other words, the policy objectives of consistency, efficiency, and equity (as well as leveraging agency expertise) are better advanced if the Tax Court only considers evidence in the administrative record and reviews for abuse of discretion to determine whether the IRS has engaged in rational decision-making.  In the event the facts in the record or the reasons contained in the IRS’s underlying ruling are found lacking, the Tax Court should identify those errors and remand to the IRS for reconsideration in light of those errors.

Why would this more deferential approach to judicial review actually be better for consistent, efficient, and equitable tax administration?  Because the status quo provides little comfort for the vast majority of claimants who do not bring the agency’s determination to the Tax Court and therefore receive no judicial review.  The Tax Court has no power to correct mistakes made in these cases, and if one assumes that the IRS has a similarly high error rate in unreviewed cases (and many unreviewed cases involve less sophisticated taxpayers), its incorrect assessments of income and ability to pay directly affect the equitable distribution of the cost of federal government.  By forcing the agency to improve its decision-making process—including better development of the administrative record and more well-reasoned agency rulings—the Tax Court can have a more profound effect on IRS processes and outcomes.

By abandoning tax exceptionalism, the Tax Court also can take advantage of the wealth of judicial experience in other administrative law contexts.  As detailed in Part III.B of our paper, courts have developed judicial doctrines and practices in other administrative law contexts to have a greater system-wide effect on agency decision-making.  For instance, as one of us uncovered in an empirical study on judicial review of immigration adjudications, courts have developed a number of tools that help alleviate the policy considerations against deferential review of agency action by enhancing their dialogue with agencies on remand.  These tools include: requesting notice of the agency’s decision on remand; retaining jurisdiction of the petition; setting a timeline for remand; providing hypothetical solutions; certifying issues for remand; obtaining concessions from the government at argument or otherwise during judicial review; and suggesting the transfer of the matter to another agency adjudicator.

By using these tools and developing more of its own, the Tax Court can begin a richer dialogue with the IRS—a dialogue aimed at not only improving the IRS’s procedures and decision-making in the case remanded but also in the agency overall.  Such dialogue is critical to improving the treatment of less sophisticated taxpayers—many of whom, unrepresented, do not even seek judicial review of the IRS’s adverse determinations.  That the vast majority of unrepresented taxpayers never seek judicial review of adverse decisions should encourage the Tax Court to engage in a richer dialogue with the IRS via the ordinary remand rule and these dialogue-enhancing tools in order to have a more systemic effect on the quality of administrative adjudication that takes place at the IRS.  It’s time as a policy matter for the Tax Court to abandon tax exceptionalism and embrace administrative law.

The Tax Court has declared that “[t]he APA has never governed proceedings in the Court (or in the Board of Tax Appeals).”  In other words, the Tax Court refuses to apply the APA’s default standard (abuse of discretion) and scope (confined to the administrative record) of review provisions.  Instead, the Tax Court considers the default in both contexts to be de novo, and only departs from de novo review when it determines that the Internal Revenue Code suggests a more deferential review standard.(As we explain in Part I of our paper, the Tax Court’s misguided view of tax exceptionalism may be largely a product of its unique history—including that Congress converted it from an agency to a court (in 1969) after the enactment of the APA (in 1946)—and the post-1969 expansion of its jurisdiction to adjudicate tax-related disputes.)

This approach contradicts general administrative law principles that every other federal court applies when reviewing agency actions.  As we explain in Part II of our paper, the APA establishes the default standards for judicial review of all federal agency action.  The IRS, an executive agency within the Treasury Department, is plainly an “agency” for purposes of the APA (5 U.S.C. § 701(b)(1)).  And while the Tax Court used to be an agency before the enactment of the APA, as of 1969 it is an “[A]rticle I . . . court of record” (26 U.S.C. § 7441).  For purposes of the APA, it is thus “a court of the United States” (5 U.S.C. § 702) and, for its review of IRS agency actions, a “reviewing court” subject to the APA’s judicial-review provisions (5 U.S.C. § 706).

That the APA governs the Tax Court does not mean the standard of review is always abuse of discretion or the scope of review is always confined to the administrative record.  Instead, these are the default standards, and the inquiry becomes whether Congress has overridden them.  Indeed, “[r]ecognizing the importance of maintaining a uniform approach to judicial review of administrative action,” the Supreme Court has instructed reviewing courts to “apply the APA’s court/agency review standards in the absence of an [statutory] exception.”  To depart from the APA default, the agency’s governing statute must suggest “more than a possibility of a [different] standard, and indeed more than even a bare preponderance of evidence”; the exception “must be clear.”

To illustrate how the Tax Court should conduct its inquiry into whether the APA default standards apply, we focus on three types of IRS actions reviewed by the Tax Court.  Check out Part II.B of our paper if you’re interested in that analysis.  In a nutshell, by starting with the APA, it becomes clearer which IRS actions are subject to the APA defaults (e.g., innocent spouse and collection due process) and which may not be (i.e., tax deficiency) based on congressional override of the APA defaults in the Internal Revenue Code.

Whether the APA’s judicial-review provisions apply to the Tax Court “is of more than passing interest.  It goes to the heart of the place of the Tax Court in our administrative system,” explained Judge Bybee last year in a dissent from a Ninth Circuit opinion that embraced Tax Court exceptionalism.  Now that the supersized Ninth Circuit has joined the Tax Court’s side of the split—with four circuits on the other side and the Eleventh Circuit somewhere in between—it is likely only a matter of time before the Supreme Court intervenes to declare the death of Tax Court exceptionalism.