TaxProf Blog op-ed: House Votes to Dump Chevron and Auer, by Steve R. Johnson (Dunbar Family Professor of Law, Florida State):
On July 12, 2016, the U.S. House of Representatives passed H.R. 4768, the Separation of Powers Restoration Act (SOPRA). The vote was 240 to 171, mostly along party lines. If enacted, SOPRA would amend 5 U.S.C. sec. 706, a judicial review section of the Administrative Procedure Act (the APA). The goal, according to the Judiciary Committee’s report, would be “to overturn the so-called Chevron and Auer doctrines of judicial deference to agency interpretations of statutory and regulatory provisions.” H. Rep. 114-622, at 2 (June 14, 2016); see Chevron U.S.A., Inc. v. Natural Resources Defense Council, 467 U.S. 837 (1984) (deference to reasonable agency interpretations of ambiguous statutes); Auer v. Robbins, 519 U.S. 452 (1997)(deference to reasonable agency interpretations of the agency’s own ambiguous regulations).
SOPRA will not be enacted, at least not in the foreseeable future. The Senate has not scheduled hearings on the measure and is unlikely to pass it. Even were SOPRA to reach the desk of the current President, it surely would be vetoed. See Office of Management & Budget, Statement of Administration Policy: H.R. 4768 (June 22, 2016) (“The Administration strongly opposes [SOPRA.] If the President were presented with H.R. 4768, his senior advisors would recommend he veto the bill.”). The votes necessary to override the veto would not be found.
For that reason, some might be inclined to dismiss the House’s action as purely symbolic, mere political grandstanding. But too ready dismissal might be a mistake. First, symbolic actions may be significant as ways to rally the base and motivate donors. Additionally, by forcing friends and foes to “go on record” with their votes, ultimately doomed bills can be ways of defining the differences between the parties, creating issues to bring before the electorate. President Obama’s nomination of Judge Merrick Garland to replace Justice Scalia was another “it’s just not going to happen” gesture but was nonetheless deemed worthwhile by the Administration.
Second, constellations in the political firmament are not forever fixed. Balances of political power and perceptions of political interest change, sometimes suddenly. It is not out of the question that enactment of SOPRA, a virtual impossibility today, could become possible in the future. For instance, in 1997, who could have predicted the avalanche of changes wrought by the IRS Restructuring and Reform Act of 1998?
Third, the House’s action keeps the kettle boiling in the long running controversy about the wisdom of current deference doctrines. The discussion of Chevron has been vast, of course. The ever evolving Chevron doctrine still has supporters, but a growing chorus urges its abolition. The debate about Auer is more recent and less extensive, but it too has gained steam in recent years. The fact of the House’s action, the arguments advanced in support of SOPRA in the Judiciary Committee’s report, and the arguments offered against SOPRA in the dissenting views in that report will keep the debate fresh.
2. Prejudice Revealed
By way of disclosure, I march in the ranks of those who think Chevron should be put out of its misery as soon as possible. See, e.g., Steve R. Johnson, The Rise and Fall of Chevron in Tax: From the Early Days to King and Beyond, 2015 Pepperdine L. Rev. 14; see also id. at 15 (identifying some of the other administrative law and tax law commentators – including Bryan Camp – who find Chevron seriously flawed). (This article is part of a Pepperdine Law Review symposium on King v. Burwell and related cases.) I feel no greater tenderness for Auer. See, e.g., Steve R. Johnson, Auer/Seminole Rock Deference in the Tax Court, 11 Pittsburgh Tax Review 1 (2013) (applauding the Tax Court’s resistance to Auer). As one might imagine, therefore, I like SOPRA.
III. Some Issues
I will not here rehearse the numerous arguments put forward by the proponents and opponents of Chevron and Auer. The arguments are captured in the articles mentioned in this posting, the cases and commentary cited in the articles, and the majority and minority views in the Judiciary Committee’s report. Instead, here are some observations particular to SOPRA.
First, is SOPRA a vindication of the separation of powers principle or itself a violation of it? There is a current of thought that, when a legislature tells a court how and how not to interpret statutes, that act is an unconstitutional usurpation of the core function of the judiciary. See generally Nicholas Quinn Rosenkranz, Federal Rules of Statutory Interpretation, 115 Harvard L. Rev. 2085 (2002); Jennifer M. Bandy, Note, Interpretive Freedom: A Necessary Component of Article III Judging, 61 Duke L.J. 651 (2011). Drawing the contours of such a doctrine is tricky because of the subtle gradations along the spectrum of influence and control.
Before SOPRA, Congress had generally refrained from such efforts, but many states have enacted statutes prescribing to greater or lesser degree the methodologies courts should employ in construing statutes. See., e.g., Steve R. Johnson, Statutes Requiring Plain Meaning Interpretation, State Tax Notes, Sept. 14, 2009, p. 763. Courts in these states have exhibited a range of reactions to such statutes – sometimes treating them as at least somewhat binding, sometimes giving them only lip service, sometimes politely ignoring them, and sometimes holding them unconstitutional.
Were SOPRA ever enacted, we might see federal case law on this issue. SOPRA might well be defensible against an “infringes on judicial prerogatives” challenge. After all, Chevron was built on the premise (fiction) that statutory gaps are implied delegations of rulemaking power from Congress to agencies. Chevron, 467 U.S. at 843-44. How could it be usurpation for Congress to clarify, through SOPRA, that it intends no such delegation? Similarly, Auer involves the reach of agency interpretations. It is fundamental that “an agency’s power is no greater than that delegated to it by Congress.” Lyng v. Payne, 476 U.S. 926, 937 (1986). Sympathetically construed, SOPRA could be viewed, not as dictating how courts wield their power, but as a withdrawal from agencies of authority to make binding rules through interpretation, rather than through amendatory formal or informal rulemaking.
Second, as passed by the House, SOPRA raises the entrenchment issue. “Generally, a statute creates a legislative entrenchment rule whenever it says that a subsequent statute will be effective only if it is enacted or phrased in a specific way.” Amandeep S. Grewal, Legislative Entrenchment Rules in the Tax Law, 62 Admin. L. Rev. 1011, 1012 (2010). Such rules take many forms. One appears in SOPRA. The last sentence of APA section 706(a), as amended by SOPRA, would provide: “No law may exempt any … civil action from the application of this section except by specific reference to this section.” (This sentence was perhaps inspired by a similar sentence in APA section 559.)
Would this sentence in SOPRA be deemed an impermissible entrenchment? There is a school of thought that no Congress constitutionally can bind future Congresses, thus sentences like those in SOPRA and in current APA section 559 are impermissible. Courts have largely finessed the issue in the cases decided thus far, and commentators, as usual, disagree. Grewal, supra, at 1017-25 (reviewing the authorities). Presumably, courts would also find ways to skirt the issue in a SOPRA context, but it would be interesting to watch their gymnastics.
Third, were SOPRA to become law, it would itself be subject to judicial interpretation. Courts that dislike a statute often prefer to eviscerate it via interpretation rather than declare it unconstitutional. The House tried to foreclose this possibility. As amended by SOPRA, the second and third sentences of APA section 706(a) would direct courts to “determine the meaning or applicability of the terms of an agency action and decide de novo all relevant questions of law, including the interpretation of constitutional and statutory provisions, and rules made by agencies. Notwithstanding any other provision of law, this subsection shall apply in any action for judicial review of agency action authorized under any provision of law.”
The Judiciary Committee report stressed the intention to “guarantee that [the measure’s] results will extend not only to cases reviewed under the APA’s judicial review [sections] but also the assorted ‘mini-APAs’ present in the United States Code” and stressed as well the committee’s desire “to render it as explicit as possible that courts, not agencies, must decide all questions of law.” H. Rep. No. 114-662, at 9-10.
The above language seems to embrace the assorted forms of action under the Internal Revenue Code, which is title 26 of the United States Code. What would be the practical effect? For example, many sections of the IRC contain “in the opinion of the Secretary” or equivalent language, which has been interpreted to impose upon the taxpayer a higher standard of proof than mere preponderance of the evidence. E.g., IRC sections 446(b) & 482. Would the universal de novo standard of SOPRA displace such sections?
This and other questions would arise in tax – and in numerous other bodies of federal regulatory law. Would serious conflicts exist between SOPRA and such bodies of law? If so, there are various standard methods by which courts resolve conflicts between statutes. See, e.g., Steve R. Johnson, The Judicial Instinct to Harmonize Statutes, State Tax Notes, Aug. 30, 2010, p. 599; Steve R. Johnson, When General Statutes and Specific Statutes Conflict, State Tax Notes, July 12, 2010, p. 113. Also relevant is the interpretational tradition that general APA rules yield to more specific rules Congress has crafted in particular areas. E.g., 5 U.S.C. sections 703 & 704; see Steve R. Johnson, Reasoned Explanation and IRS Adjudication, 63 Duke L.J. 1771, 1821-27 (2014). On the other hand, also germane is the Supreme Court’s injunction in Mayo that, absent justification, the same Administrative Law rules should be applied to Treasury/IRS as to other agencies. Mayo Found. for Med. Educ. & Research v. United States, 131 S. Ct. 704, 713 (2011).
My point is that the apparently clear language of SOPRA (and of the committee report for those judges who use such reports) might cause a variety of thorny practical difficulties and that courts – motivated by the desire to avoid such difficulties or simply by aversion to SOPRA – might read the language in surprising ways. Both the judicial Left (for instance in King v. Burwell, 135 S. Ct. 2480 (2015)) and the judicial Right (for instance in FDA v. Brown & Williamson Tobacco Co., 529 U.S. 120 (2000)) sometimes have found it appropriate to contort or ignore clear statutory text, as have judges in numerous other cases, e.g., Bob Jones Univ. v. United States, 461 U.S. 574 (1983); Cabell v. Markham, 148 F.2d 737 (2d Cir. 1945), aff’d, 326 U.S. 404 (1945); Riggs v. Palmer, 22 N.E.2d 188 (N.Y. 1889).
Fourth, would enactment of SOPRA have a spill-over effect on state administrative and tax practice? The states, of course, are not constitutionally bound to follow Chevron, Auer, or other federal interpretational principles. Nonetheless, the influence of the Supreme Court is such that state courts sometimes have emulated federal deference rules. See, e.g., Steve R. Johnson, Chevron Deference to State Tax Agencies, State Tax Notes, Jan. 24, 2011, p. 285; Steve R. Johnson, Deference to Tax Agencies’ Interpretation of Their Regulations, State Tax Notes, May 30, 2011, p. 665. (These articles are part of a series on deference in state and local tax cases.) One wonders whether SOPRA could cause some state courts to rethink their approaches to deference.
Fifth, it is rare for human beings of any station or office to faithfully follow a prescribed evaluative method irrespective of the results it produces. More commonly, people decide what results they want, then commit to whatever methodology or set of rules seems most likely to produce those results, and then de-commit when that methodology fails to achieve those results. Thus, one’s like or dislike of agency deference rules often correlates with whether one’s preferred political party happens to hold the White House at the moment. It is no accident that every single voting Republican member of the Judiciary Committee voted for SOPRA while every single voting Democratic member voted against it.
However, at least as to Chevron, I’m not certain that the sides -- either of them -- have a firm grip on the facts. Chevron is usually thought of as a strongly pro-agency rule, an agency-liberating rule. At one time, it may have been, but it has been notably less so in recent years. With any principle of law, the spirit in which it is applied in more important than its verbal formulation. Accordingly, if the spirit of application changes, a rule can produce different outcomes at different times, even if the language of the rule remains constant. In recent years, the Court has whittled away at Chevron, creating Step Zero exceptions in cases like Brown & Williamson, Christensen v. Harris County, 529 U.S. 576, 587 (2000), United States v. Mead Corp., 533 U.S. 218 (2001), Gonzales v. Oregon, 546 U.S. 243 (2006), United States v. Home Concrete & Supply, LLC, 132 S. Ct. 1836 (2012), and Burwell. Moreover, the Supreme Court and lower courts have conducted both Step One and Step Two analysis in considerably more rigorous and exacting fashion. Thus, in tax, Treasury and IRS positions were rejected – under Chevron – in cases such as Direct Marketing Ass’n v. Brohl, 135 S. Ct. 1124 (2015), Home Concrete, Loving v. IRS, 742 F.3d 1013 (D.C. Cir. 2014), Dominion Resources, Inc. v. United States, 681 F.3d 1313 (Fed. Cir. 2012), Ridgely v. Lew, 55 F. Supp. 3d 89 (D.D.C. 2014), and Altera Corp. v. Comm’r, 145 T.C. No. 3 (2015), among others. See Johnson, 2015 Pepperdine L. Rev. at 21-26.
If, in fact, Chevron – as currently applied -- is not an Emancipation Proclamation for agencies, Republicans perhaps should be less enthusiastic about SOPRA and Democrats should perhaps be less outraged by it. Nonetheless, old perceptions, like old habits, die hard. Moreover, there might be fundraising and other benefits to both parties in continuing to mischaracterize Chevron, even if they realize the current reality of Chevron’s application.
Sixth, whether one views SOPRA as a genuine development or merely political opportunism, the fault can be laid at the Supreme Court’s own doorstep. Many have described the hopeless muddle the Court has made of Chevron doctrine in the third of a century since the case was handed down. E.g., Thirty Years of Chevron v. NRDC and the Administrative Law Review: A Letter from the Executive Board, 66 Admin. L. Rev. 235, 238 (2014)(noting the “strikingly contradictory Supreme Court decisions which continually led academics to question [Chevron’s] relevance”): ABA Section of Taxation Report of the Task Force on Judicial Deference, 57 Tax Law. 717, 719-20 (2004). If animated by candor uninhibited by professional etiquette, one might well describe the Court’s behavior as disgraceful.
And, as nature abhors a vacuum and as predators cull from the herd the weak, when one Branch of government demonstrates itself feckless, the tendency is for the other Branches to encroach. The Judiciary Committee’s report openly acknowledges that dynamic as to SOPRA: “the Judicial Branch for more than thirty years has revealed the difficulty it has faced in cleanly, clearly, simply, and definitively explicating whether, why, how, and specifically when it becomes the Executive Branch’s power to ‘determine what the law is,’ not the courts’, under the Chevron doctrine …. Accordingly, it is appropriate for Congress to overturn Chevron and Auer statutorily, rather than wait for the Court to address the numerous deficiencies.” H. Rep. 114-622, at 9.
Seventh, in most of the legal academy, the word “nuanced” is a term of praise. But the quest for exquisite calibration can produce much trouble. The more exceptions, elements, predicates, and conditions a statute, regulation, or judicial rule has, the more likely the legal system is to trip over its own feet in juggling the features. Much of Constitutional Law is a quagmire because of multiple, ever evolving tests, levels of scrutiny, special rules, and the like. And much of the monstrosity of the IRC is because we too often have made the wrong choice when picking between a generally satisfactory “rough justice” rule that the IRS and taxpayers alike can understand and apply versus an impenetrably complicated rule which is theoretically perfect but which is beyond the ability of mere mortals to grasp and to follow. See, e.g., Steve R. Johnson, The E.L. Wiegand Lecture: Administrability-Based Tax Administration, 4 Nev. L.J. 573 (2004).
So too in the topic at hand. SOPRA is a harbinger of the structure of endlessly evolving deference doctrine collapsing of its own weight. Why do we need multiple levels of scrutiny in Constitutional Law? Why do we need scores of substantive canons in statutory interpretation? Why do we need multiple, nuanced, mercurial doctrines of deference to agencies? Wouldn’t it be much simpler for judges to take all these thumbs off all these scales and give an honest weight? Just give constitutional provisions, statutes, and regulations the fair reading their language and their contexts suggest.
I have urged this straight-forward approach in pieces like Tilted Versus Reasonable Interpretation of Tax Laws, State Tax Notes, October 25, 2010, p. 277, and The Canon that Tax Penalties Should Be Strictly Construed, 3 Nev. L.J. 495 (2003). More importantly, the Supreme Court – during moments of lucidity temporarily dispelling the glamour of nuance – has sometimes acknowledged to desirability of this approach. For example, in one case, the Court refused to apply a preferential canon, saying: “We are not impressed by the argument that, as the question here decided is doubtful, all doubts should be resolved in favor of the taxpayer. It is the function and duty of courts to resolve doubts. We know of no reason why that function should be abdicated in a tax case more than in any other where the rights of suitors turn on the construction of a statute and it is our duty to decide what that construction fairly should be. Here doubts which may arise upon a cursory examination of [the sections] disappear when they are read, as they must be, with every other material part of the statute … and in the light of their legislative history.” White v. United States, 305 U.S. 281, 292 (1938).
The approach of White is solidly grounded in Chief Justice Marshall’s declaration on behalf of the Court: “It is emphatically the province and duty of the Judicial Department to say what the law is.” Marbury v. Madison, 5 U.S. 137, 177 (1803). This clear location of decisional authority and responsibility will, I believe, serve the country better than the kaleidoscope of deference doctrines. Even if only symbolic, the passage of SOPRA by the House is a welcome development.