Saturday, July 6, 2019

Auer Deference and Treasury and IRS Policy Statement on the Tax Regulatory Process (7/6/19)

In the  revised working draft for my Federal Tax Procedure editions (Student and Practitioner) which I will publish in final in August 2019, I have revised the section titled "Deference to Subregulatory Interpretations."  This is generally called Auer deference, after Auer v. Robbins, 519 U.S. 452 (1997).  (Auer deference seems to overlap what was formerly called Seminole Rock deference, after Bowles v. Seminole Rock & Sand Co., 325 U.S. 410 (1945).)

Auer deference may be stated as deference to agency interpretations of agency regulations.  As I analyze it, the agency regulations being interpreted may be legislative regulations which are the law but, like statutes, can be ambiguous and thus subject to reasonable agency interpretation.  The agency regulations more commonly may be Chevron-entitled interpretive regulations that are ambiguous and thus subject to agency interpretation.  In either event, the agency interpretation appears in subregulatory guidance not entitled to Chevron deference.

Kisor v. Wilkie, 588 U.S. ___, 139 S.Ct. 2400 (2019) [Sup Ct Slip Op here; Google Scholar with S.Ct. pagination here], approved a restricted version of Auer deference.  The plurality opinion written by Justice Kagan (concurred in by Justices Ginsburg, Breyer and Sotomayor), had a robust support for Auer deference as restricted.  See e.g., Cass R. Sunstein, Justice Kagan’s Powerful Defense of the Administrative State (BloombergOpinion 6/28/19).  Keep in mind that the plurality portion was written and concurred in by the two administrative law experts on the Court (Justices Kagan and Breyer).  [For earlier discussion on Kisor, see Supreme Court Yet Again Weighs In At the Edges on Legislative and Interpretive Rules (Federal Tax Procedure Blog 6/26/19; 7/2/19), here.]

Kisor approved Auer deference to subregulatory interpretations of regulations but reined in its application.  Under Kisor, Auer deference for subregulatory interpretations (or more generally interpretations not entitled to Chevron deference) may draw deference and thus be the law.

In a tax context, however, the recent Treasury and IRS Policy Statement on the Tax Regulatory Process (3/5/19), here, informed the public that, as a policy matter, regardless of whether subregulatory interpretations might qualify for Chevron or Auer deference, the Treasury and the IRS (and presumable DOJ Tax) would not assert either Chevron or Auer deference for subregulatory interpretations.

This is odd.  Can Treasury and the IRS do that in light of the Supreme Court approval of Auer?  Well, of course, in one sense Treasury and the IRS certainly can avoid asserting deference.  The question is whether, if Chevron deference or Auer deference is otherwise applicable, a court can ignore either form of deference?

Here is the relevant portion of my current working draft on that issue (footnotes omitted):

(viii)  Deference to Subregulatory Interpretations.

What does Chevron and its progeny tell us about the deference to be accorded agency subregulatory interpretations – e.g., in the tax arena, Revenue Rulings and Revenue Procedures?  Subregulatory guidance must be interpretive in order to be exempt from the notice and comment requirement of the APA.

The starting point for the state of subregulatory interpretation is the Supreme Court’s recognition in Mead that notice and comment regulations are not necessarily required for Chevron deference.  Stated otherwise, Mead created a “safe harbor” of Chevron deference for regulations with notice and comment, but left interpretations with less procedural formality open for possible Chevron deference.

I return to so-called Skidmore deference which, if deference at all, is weaker, much weaker, in comparison to Chevron deference.  Skidmore v. Swift & Co., 323 U.S. 134 (1944).  Skidmore said that “courts and litigants may properly resort” to interpretations in agency guidance not entitled to stronger deference (now called Chevron deference) based on their “power to persuade.”  As thus formulated, it is unclear whether deference is the right description, for deference means in the final analysis a court applying a reasonable agency interpretation when the court believes another interpretation is the more reasonable interpretation.  If the agency interpretation has the power to persuade, then the court applies the agency interpretation because it is persuasive and does not “defer” in any meaningful sense to that interpretation.  Courts nevertheless use the term Skidmore deference as if it were meaningful, so I use the term also.  Presuming a court means something in paying homage to Skidmore deference, perhaps it means that a court must give slight tilt in favor of an agency interpretation when it does not rise to the level required for Chevron deference.  So, in a case in which the court will not apply Chevron deference, it might still apply Skidmore deference to a an IRS subregulatory interpretation.

This analysis may be pre-empted by the Treasury and IRS Policy Statement on the Tax Regulatory Process (3/5/19), here.  In that Policy Statement, Treasury and the IRS state as a policy that they would not rely upon Chevron or Auer deference for subregulatory interpretations.  Treasury and the IRS do not address Skidmore deference, so that, in theory, if Skidmore means anything, Skidmore deference might apply to IRS subregulatory interpretations.  I have reasoned above that  Skidmore may not be meaningful as a deference concept.  And that means that all interpretations in subregulatory guidance prevail only if they persuade the court, and deference is an oxymoron.

In versions of this text published prior to the Policy Statement, I reasoned that some IRS subregulatory guidance might be Chevron-entitled because Mead said that regulations were not required for Chevron deference and that Auer deference might apply if the subregulatory guidance interpreted ambiguous regulations.  I noted, however, that the Department of Justice had stated that it would not assert Chevron deference for revenue rulings, the highest level of subregulatory guidance.  The Policy Statement is consistent with that prior policy but extends to Auer deference for all subregulatory interpretations as well..

An issue might be whether an agency (here the Treasury and the IRS) can waive Chevron or Auer deference if the interpretation is otherwise entitled to such deference.  I suspect that whatever might have been the possibilities without the Policy Statement, courts will avoid conferring Chevron deference for subregulatory guidance (which, so far as I am aware, they never given before the Policy Statement even with the theoretical possibility stated in Mead).  I am not so sure that, given the recent decision in Kisor v. Wilkie, 588 U.S. ___, ___ S.Ct. ___, 2019 U.S. LEXIS 4397 (6/26/2019), having a robust defense for Auer deference, albeit in a plurality opinion and significantly cabined in its application, a court will want to ignore Auer deference for subregulatory interpretations of regulations.  Nor am I sure that the IRS will not revise its Policy Statement to allow it to assert Auer deference.  (I think it should, because Auer deference (as now cabined) seems more consistent with sound administration.)

There are other issues for subregulatory guidance.

What about prospectivity for subregulatory interpretations?  The rule has been that those interpretations are retroactive to the effective date of the statute they interpret.  As with the traditional holding for interpretive regulations, the notion is that the interpretation is of the statute and hence the statute effective date is controlling.  (For regulations, of course, there are some limitations in § 7805.)  But, what if giving subregulatory guidance deference–whether Chevron or Auer or Skidmore (if meaningful)–imbues the interpretation with the force of law, does that make the interpretation a legislative rule which must be promulgated as a regulation with notice and comment and must be prospective only?  The plurality opinion in Kisor addresses that issue and soundly rejects it.  (This is another good reason to be wary of the claims that given deference to an agency interpretation–i.e., applying the agency interpretation rather than the court’s own better interpretation–turns what would otherwise be interpretive rulemaking into legislative rulemaking.)

What deference, if any, should be accorded to agency statutory interpretations advanced for the first time in litigation? Chevron is based on the policy that Congress has delegated to the agency the power to make interpretations and that considered agency interpretations, particularly those subject to notice and comment, pursuant to that delegation should be given deference.  I have noted that the process involved in promulgating regulations insures that the regulations interpretations are forced through a process that should result in generally better-considered interpretations.  Other agency interpretations, such as revenue rulings, have some lesser process–but process nonetheless–designed to give some institutional consideration to the positions. I would think that,  under Kisor, interpretations of ambiguous regulations in revenue rulings would qualify for Auer deference (except of course for the effect of the Policy Statement).  But under Kisor, ad hoc litigating interpretations in briefs or brief substitutes would not qualify for Auer deference.

No comments:

Post a Comment