An Empirical Defense of Auer Step Zero
Since 1945, the Supreme Court has given binding respect to a federal agency’s interpretation of its own regulation unless the agency’s construal is incorrect.1 This principle is commonly known as either Auer2 or Seminole Rock3 deference, named after the two cases most often associated with the doctrine’s canonical formulation4—that an agency’s regulatory construction is of “controlling weight unless it is plainly erroneous or inconsistent with the regulation.”5 Although this strong form of deference has existed for more than seventy years, it has become the subject of an intense academic debate only in the last two decades and the polemics have created practical consequences. In 2013, for example, Chief Justice John Roberts announced that the Supreme Court has “some interest in reconsidering” the doctrine.6 In the current 115th Congress, the House of Representatives has passed legislation that would replace Auer deference with de novo review7 and the Senate is considering a bipartisan companion bill that would replace Auer deference with non-binding judicial respect.8
The purpose of this study is to inform the ongoing and consequential debate over Auer deference with empirical data. To this end, I assessed the entire population of U.S. Courts of Appeals decisions from 1993–2013 that employed Auer to review an agency’s regulatory interpretation. To enrich this investigation of Auer with a comparative analysis, I created datasets reflecting every U.S. Court of Appeals decision from 1993–2013 that employed the other two primary forms of deference in administrative law: binding Chevron9 deference and nonbinding Skidmore10 respect. The intent behind conducting a controlled comparison of multiple doctrines in the same courts during the same duration is to afford greater depth of analysis and thereby allow the drawing of stronger inferences.
Part I of this Note assesses four categories of argumentation swirling about Auer deference, beginning with proponents who assert that Auer is appropriate because agencies, and not courts, are better equipped to make policy by interpreting legal texts with the force of law given agencies’ relative advantages in subject matter expertise and political accountability.11 Critics counter that the Auer doctrine encourages poor rule drafting and procedural shortcuts by combining both rule writing and rule exposition authority in the Executive Branch.12 The third category of participants in the Auer debate is composed of those who warn that reforming the doctrine could lead to unintended consequences, including a decrease in participatory policymaking, less precise regulations, and unmanageable administrative burdens.13 The fourth perspective regarding Auer is provided by empirical analyses, the results of which have been inconsistent: one study suggests Auer is a form of super-strong deference, another claims it is no stronger than other forms of deference, and another indicates that the government’s win rate under Auer has diminished in recent history.14
Part II explains the methodology of this study. I used database searches to identify and review the entire population of U.S. Courts of Appeals decisions from 1993–2013 that employed the Auer framework. Because the respective populations of 1993–2013 circuit court cases relying on Chevron and Skidmore deference were too great to practicably review for this study, I used a simple probability method to create samples from which I could draw inferences about the population as a whole. Thus, I collected an original dataset of variables attendant to 1,047 published federal courts of appeals decisions: 416 for Auer, 392 for Chevron, and 239 for Skidmore. Across these decisions, the U.S. Courts of Appeals reviewed 1,120 discrete textual interpretations by regulatory agencies. For each interpretation, I recorded identifying information—case name, case citation, and agency involved—and whether the government’s interpretation was accepted by the court. I also assigned each interpretation to one of twelve categories of administrative procedure. This study contributes to a growing body of empirical analyses of deference regimes as applied by Article III courts,15 yet its methodology differs from its predecessors in three important ways. First, this is the first analysis to systematically compare population-level statistics across multiple deference regimes as employed by U.S. Courts of Appeals.16 Second, this study covers twenty years, a longer timespan than previous studies.17 It is therefore more representative of the courts’ behavior. Finally, this analysis provides a more refined investigation of administrative processes than prior empirical studies.18
Part III sets forth the results of the study and draws three significant inferences from the data regarding the ongoing debate over Auer deference. First, the data indicate that over the twenty-year period, Auer deference was indeed stronger than Chevron. From 1993–2013, the federal government prevailed in 74% of cases when the court invoked Auer and in 68% of cases when it invoked Chevron.19 However, the data also lend support to the thesis that the strength of Auer in the circuit courts has narrowed in the wake of Supreme Court decisions that constrained the doctrine.20 Comparative data contradict the likelihood that the government’s diminished win rate under Auer is part of a larger trend across all deference doctrines. After 2006, the government’s win rate fell significantly when it invoked Auer, whereas its win rate under Chevron and Skidmore remained flat and slightly increased, respectively.21 Also, at any time scale, the government’s win rate when courts invoked Auer and Chevron was significantly greater than when it invoked Skidmore.22 This makes sense given that the former are binding doctrines, whereas the latter is nonbinding.
The second significant result demonstrates the unexpected breadth of administrative process associated with Auer deference. To date, the Auer debate has focused on the extent to which the doctrine encourages policymaking through informal procedures that deprive the public of notice and participation. Inherent in this controversy is an assumption, sometimes explicit, that virtually all the interpretations that benefit from Auer deference are found in informal issuances. In fact, the results of this study indicate that courts give Auer deference to interpretations falling across the continuum of administrative procedure in a surprisingly balanced manner. There are, for example, significantly more interpretations subject to Auer that result from formal adjudications than from nonlegislative rules.23 Overall, federal courts of appeals applied the Auer framework more to interpretations resulting from administrative processes that carry the force and effect of law than to interpretations that do not.24 This surprising result directly bears on the argument set forth by reform skeptics. These skeptics claim that an unintended consequence of reforming the doctrine could be to encourage agencies to exercise a form of discretion under SEC v. Chenery Corp.,25 known commonly as “Chenery II discretion,”26 to shift policymaking from rulemaking to adjudication.27 This argument is based on the understanding that an agency’s discretion to choose between adjudication and rulemaking is a substitute for Auer; however, the data suggest that these two doctrines are complements, which should depress the likelihood of unintended consequences resulting from reforming Auer.
The third significant result of this study is the quantification of the administrative burden incurred by replacing or reforming Auer. Using this study’s dataset, it is possible to create a simple model that estimates the administrative burden of either replacing Auer deference with Skidmore respect or reserving Auer for interpretations stemming from administrative procedures that carry the force and effect of law.28 In this manner, the study tests the claim made by Auer’s defenders that reforming the doctrine would unduly hinder administrative and judicial efficiency. As this study demonstrates, fully rejecting Auer and replacing it with residual Skidmore deference would have resulted in an estimated fifty-one fewer agency regulatory interpretations surviving judicial review in the circuit courts from 1993–2013,29 or about one interpretation per circuit court every five years.30 These results belie claims that disrupting the doctrine would lead to chaos in regulatory agencies and federal courts.
After assessing the four viewpoints in the Auer debate in light of the empirical data in this study, I conclude that the Supreme Court should reform Auer, not reject it. Virtually all criticism leveled at the doctrine would be addressed if the Court limited Auer deference in the same fashion it does Chevron deference—by limiting the judicial respect accorded to interpretations resulting from formal administrative procedures that carry the force and effect of law. For starters, such reform would address the hole in the institutional argument advanced by Auer’s proponents. If Auer is right for the same reasons that Chevron is right, Auer should be treated the same, if for no other reason than uniformity. Reforming Auer would also placate Auer’s critics by closing the loophole they allege the doctrine creates in the Administrative Procedure Act (APA).31 Finally, as discussed, the data deflate the concerns expressed by skeptics of Auer reform to the extent the data suggest that agencies’ Chenery II discretion is a complement of, not a substitute for, Auer. All told, reforming Auer is a modest check accompanied by minimal administrative burden that would achieve doctrinal uniformity and answer unaddressed criticisms.
Subscribe to GLJ1. See Bowles v. Seminole Rock & Sand Co., 325 U.S. 410, 414 (1945).
2. Auer v. Robbins, 519 U.S. 452, 461 (1997).
3. Seminole Rock, 325 U.S. at 414.
4. See Aaron L. Nielson, Beyond Seminole Rock, 105 GEO. L.J. 943, 945 (2017) (“At least as it has come to be understood, Seminole Rock deference—also commonly called Auer deference—commands
courts to defer to an agency’s interpretation of its own ambiguous regulations.”); see also Decker v. Nw. Envtl. Def. Ctr., 133 S. Ct. 1326, 1338 (2013) (Roberts, C.J., concurring) (“The opinion concurring
in part and dissenting in part raises serious questions about the principle set forth in Bowles v. Seminole Rock & Sand Co. and Auer v. Robbins.” (citations omitted)). Throughout this Note, the author uses
these terms interchangeably, as they have been used in the scholarship.
5. Seminole Rock, 325 U.S. at 414.
6. Decker, 133 S. Ct. at 1339 (Roberts, C.J., concurring).
7. See Regulatory Accountability Act of 2017, H.R. 5, 115th Cong. (2017) (as passed by the House of Representatives, Jan. 11, 2017).
8. See Regulatory Accountability Act of 2017, S. 951, 115th Cong. (2017) (as introduced in the Senate, Apr. 26, 2017).
9. Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 842–43 (1984). Chevron famously established a two-step framework to review statutory interpretations by administrative agencies. Id. At step one, if “the intent of Congress is clear,” both the court and the agency “must give effect to the unambiguously expressed intent of Congress.” Id. If, however, the statute is “silent or ambiguous,” courts proceed to step two and ask “whether the agency’s answer is based on a permissible construction of the statute.” Id. at 843.
10. Skidmore v. Swift & Co., 323 U.S. 134, 140 (1944). Though it was decided in 1944, the status of Skidmore deference was very much in doubt in the wake of Chevron, which was issued in 1984. See Kristin E. Hickman & Matthew D. Krueger, In Search of the Modern Skidmore Standard, 107 COLUM. L. REV. 1235, 1237 (2007) (“With its well-known 1984 decision in Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc. and its emphasis therein on mandatory deference toward reasonable agency interpretations of ambiguous statutes, the Supreme Court threw the viability of Skidmore into doubt.” (footnote omitted)). However, Skidmore was revived in Christensen v. Harris County, 529 U.S. 576, 587 (2000) and United States v. Mead Corp., 533 U.S. 218, 227–28, 234–35 (2001), both of which identified Skidmore as an alternative to controlling deference. Under Skidmore’s familiar formulation, the weight a court gives to an administrative interpretation depends on “the thoroughness evident in its consideration, the validity of its reasoning, its consistency with earlier and later pronouncements, and all those factors which give it power to persuade, if lacking power to control.” Skidmore, 323 U.S. at 140. The Skidmore principle, unlike Auer or Chevron deference, is not a reflection of inferred congressional intent; rather, it is a purely judicially made doctrine, whose purpose is to aid the process of textual interpretation. See Hickman & Krueger, supra, at 1239–40.
11. See Pauley v. Bethenergy Mines, Inc., 501 U.S. 680, 697–98 (1991) (rooting its discussion of Auer deference in the same grounds as Chevron by inferring congressional intent for federal courts to defer to an agency’s regulatory interpretation because of the agency’s relative advantage over courts in political accountability and expertise); Cass R. Sunstein & Adrian Vermeule, The Unbearable Rightness of Auer, 84 U. CHI. L. REV. 297, 307–08 (2017) (explaining “Auer is right for the same reason that Chevron is right: where Congress has not been clear, deference to the agency, in the face of genuine ambiguity, is the best instruction to attribute to it” due to the agency’s institutional advantages in expertise and political accountability).
12. See Perez v. Mortg. Bankers Ass’n, 135 S. Ct. 1199, 1211–12 (2015) (Scalia, J. concurring) (warning that the interaction between the Administrative Procedure Act’s exception for “interpretative rules” and Auer deference creates a paradox whereby avowedly non-binding interpretive rules can be accorded binding effect by courts of law); John F. Manning, Constitutional Structure and Judicial Deference to Agency Interpretations of Agency Rules, 96 COLUM. L. REV. 612, 647 (1996) (positing that the Auer doctrine encourages agencies to write vague rules so they can later interpret them through informal administrative procedures, yet still receive binding deference).
13. See Scott F. Angstreich, Shoring Up Chevron: A Defense of Seminole Rock Deference to Agency Regulatory Interpretations, 34 U.C. DAVIS L. REV. 49, 116–18 (2000) (arguing that reforming Auer could lead regulated entities to withhold their input during notice-and-comment rulemaking, thereby diminishing the rule’s quality); Conor Clarke, The Uneasy Case Against Auer and Seminole Rock, 33 YALE L. & POL’Y REV. 175, 193 (2015) (claiming that reforming Auer would cast doubt on “thousands” of long-standing interpretations); Nielson, supra note 4, at 948–49 (explaining that diminishing deference to an agency’s regulatory interpretations might lead agencies to exercise their discretion to promulgate policy by adjudication, rather than the more inclusive process of notice-and-comment rulemaking).
14. See Cynthia Barmore, Auer in Action: Deference After Talk America, 76 OHIO ST. L.J. 813, 825–27 (2015) (investigating circuit court application of Auer deference from 2011–2014 and concluding that Auer is no longer the extreme form of deference it was once thought to be); William N. Eskridge, Jr. & Lauren E. Baer, The Continuum of Deference: Supreme Court Treatment of Agency Statutory Interpretations from Chevron to Hamdan, 96 GEO. L.J. 1083, 1099–1100 (2008); Richard J. Pierce, Jr. & Joshua Weiss, An Empirical Study of Judicial Review of Agency Interpretations of Agency Rules, 63 ADMIN. L. REV. 515, 519–20 (2011) (investigating circuit court application of Auer deference from 1999–2001 and 2005–2007 and concluding that courts uphold government interpretations roughly the same under Auer, Skidmore, and Chevron deference).
15. See, e.g., Barmore, supra note 14, at 825–27 (reviewing circuit court application of Auer deference from 2011–2014); Kent Barnett & Christopher J. Walker, Chevron in the Circuit Courts, 116 MICH. L. REV. 1, 5–6, 28 (2017) (surveying 1558 agency statutory interpretations reviewed by circuit courts from 2003–2013 and finding that these courts upheld 71% of interpretations); Hickman & Krueger, supra note 10, at 1271–81 (analyzing circuit court application of Skidmore and finding that courts upheld 60.4% of interpretations); Orin S. Kerr, Shedding Light on Chevron: An Empirical Study of the Chevron Doctrine in the U.S. Courts of Appeals, 15 YALE J. ON REG. 1, 4–5, 30 (1998) (examining circuit court application of Chevron in 1995 and 1996, and finding that courts upheld 73% of interpretations); Pierce & Weiss, supra note 14, at 519–20 (investigating circuit court application of Auer deference from 1999–2001 and 2005–2007); Peter H. Schuck & E. Donald Elliott, To the Chevron Station: An Empirical Study of Federal Administrative Law, 1990 DUKE L.J. 984, 990 (1990) (accounting government win rates in administrative law cases before and after Chevron in four six-month periods from 1965–1985, and one two-month period in 1988).
16. None of the studies identified above in note 15 analyzed the application of multiple deference regimes at once. To this author’s knowledge, the only controlled study of multiple deference regimes considered Supreme Court decisions. See Eskridge & Baer, supra note 14, at 1097–1200.
17. See, e.g., Barmore, supra note 14, at 815–16, 827 (limiting period of analysis to 2011–2014); Pierce & Weiss, supra note 14, at 519 (investigating government’s win rate under Auer between 1999–2001 and 2005–2007); Schuck & Elliott, supra note 15, at 989–96, 989 n.13 (examining appeals court cases involving “judicial review of federal administrative action” in four six-month periods from 1965–1985, and one two-month period in 1988).
18. Some previous studies did not record the administrative process attendant to the interpretations. See, e.g., Hickman & Krueger, supra note 10; Pierce & Weiss, supra note 14, at 519–20 (omitting discussion of administrative procedure in study design). Others did. See, e.g., Barmore, supra note 14, at 826 (dividing administrative processes into six categories: “agency orders,” “public issuances,” “amicus briefs,” “administrative court decisions,” “party briefs,” and “private communications”); Eskridge & Baer, supra note 14, at 1206 (employing three categories of administrative process: “Legislative Rule or Executive Order,” “Formal Adjudication,” and “Informal Agency Interpretation”). By comparison, this study uses twelve categories to group administrative processes. See infra Section II.B.
19. See infra Table 1.
20. See Christopher v. SmithKline Beecham Corp., 132 S. Ct. 2156, 2168 (2012) (denying Auer deference to agency interpretation of regulation that failed to provide proper notice to regulated entities); Gonzales v. Oregon, 546 U.S. 243, 257 (2006) (holding that Auer deference is inappropriate for agency interpretation of regulation that merely repeats the statutory text).
21. See infra Table 1.
22. See infra Table 1.
23. Of the interpretations subject to Auer reviewed in this study, fifty-eight interpretations were contained in nonlegislative rules and seventy-two interpretations originated in precedential adjudication. See infra Table 2.
24. Of the cases that included interpretations subject to Auer deference where the government was a party reviewed in this study, 177 interpretations came in formats that carry the force and effect of law
and 169 interpretations did not. This accounting omits interpretations found in regulatory preambles, which are difficult to classify in the context of Auer deference. On the one hand, preambles do not carry
the force and effect of law. On the other hand, the preamble is arguably the optimal opportunity for agencies to interpret the rules they write. See generally Kevin M. Stack, Interpreting Regulations, 111 MICH. L. REV. 355 (2012) (arguing that regulatory preambles, which agencies are required to produce and which must include a detailed explanation of the grounds and purposes of the regulation, are among the most reliable sources for discerning regulatory purpose). This accounting also omits party briefs where the government was not a litigant. See infra Section II.B.8.
25. 332 U.S. 194 (1947).
26. See Nielson, supra note 4, at 948 & n.25.
27. There are two primary means—rules and adjudication—by which agencies issue policy with the
force and effect of law. Rules are prospective and of general applicability; adjudications, by contrast,
are retrospective and usually pertain to disputes between individuals and the government. Generally,
rulemaking is akin to the legislative process, while adjudication resembles the case-by-case policymaking
used in the common law. See GARY LAWSON, FEDERAL ADMINISTRATIVE LAW 48–49 (7th ed. 2016).
28. This second modeling possibility is meant to project the administrative burden of adopting a Step Zero for the Auer doctrine. See infra Section II.A.
29. This figure represents the difference between how often the government’s interpretation prevailed under Auer versus the number of times the government would have prevailed had the courts applied residual Skidmore deference to those cases instead. See infra Part III.
30. The equation that produces this result is: (51 interpretations ÷ 20 years) ÷ 13 U.S. Courts of Appeals = 1 interpretation per circuit court about every 5 years.
31. 5 U.S.C. §§ 500–596 (2012).