Here are the next five cases in my ongoing coverage of lower courts applying Kisor v. Wilkie, 139 S. Ct. 2400 (2019). I'm proceeding in order from the earliest case to the most recent. Am I falling behind by the day? Yes. Yes, I am. Am I giving up? Nope. Not a chance. I've listed the cases that merited deeper treatment first.  The cases covered include a Ninth Circuit decision that I argue applied the Kisor framework well and a Third Circuit case that I argue did not. All comments welcome!

Wolfington v. Reconstructive Orthopaedic Associates II PC, 935 F.3d 187 (3d Cir. Aug. 20, 2019)*

Apologies for the long analysis that follows. As you'll see, I strongly disagree with the Third Circuit's approach to Kisor/Auer in this case. Out of respect for the courts, I'm always hesitant to criticize their work product. I realize they've spent way more time with these cases than I have, and Lord knows I make lots of mistakes. So when you see me question a court's reasoning/conclusions as I do here, just know that much hand-wringing and careful reading and re-reading preceded it. Finally, if you read this and think--or even suspect--I'm getting something wrong in any of my criticisms of the Court's opinion, please send me an e-mail or say something in the comments. If any of this is in error, I want to fix it as quickly as possible.


Wolfington alleged that a consumer-finance agreement between himself and a medical center (Rothman) made Rothman a "creditor" subject to disclosure requirements in the Truth in Lending Act and its implementing regulations. The Act (and Regulation Z) require creditors to make certain disclosures before the “consummation” of the credit transaction. See 15 U.S.C. § 1638(a); 46 Fed. Reg. 50,288, 50,323 (Oct. 9, 1981). To qualify as a "creditor" as Regulation Z defines the term, Rothman had to have extended credit to Wolfington in a "written agreement":

(17) Creditor means: (i) A person who regularly extends consumer credit that is subject to a finance charge or is payable by written agreement in more than four installments (not including a down payment), and to whom the obligation is initially payable, either on the face of the note or contract, or by agreement when there is no note or contract.

12 C.F.R. § 226.2(a)(17)(i) (emphasis added). The district court granted judgment on the pleadings, agreeing with Rothman that Wolfington's pleadings didn't allege sufficiently the existence of such a "written agreement." Wolfington appealed, arguing that he had. He pointed to allegations in his complaint regarding two e-mails he had received from Rothman confirming a prior agreement between the two. According to Wolfington, those e-mails either constituted a "written agreement for purposes of Regulation Z" or at least were ‘‘indicative of a separate written agreement between the parties.’’

The Third Circuit's Decision

The Third Circuit held that wasn't enough. "Although Regulation Z does not necessarily require the written agreement itself to meet all the formalities of a contractual agreement," Wolfington's allegations couldn't establish the existence of a "written agreement" under Regulation Z because Wolfington never alleged that he had signed or otherwise executed the agreement described in his Complaint. Of course, the definition of creditor quoted above doesn't mention a signature requirement. Nor is a signature necessary to create a binding written contract--much less a "written agreement." As the Third Circuit itself explained earlier in the same opinion:

It is black-letter law that, as a general matter, no signed document is required to create a contractual obligation. Instead, the exchange of promises to perform is sufficient to form a contract. Wolfington has pled such an exchange. This is sufficient, on a motion for judgment on the pleadings, to infer the existence of a contractual agreement.

(footnote omitted). So, you might understandably wonder, how could Wolfington have possibly predicted that his claims would get tossed at the pleadings stage simply because he didn't allege he had signed the agreement at issue? You guessed it! That "zombified," "paper tiger" of a judicial-deference doctrine we all know and love: Auer deference. See Kisor, 139 S. Ct. at 2425 (Gorsuch, J., concurring).

What Wolfington (and I) missed was a "Staff Interpretation" of Regulation Z issued by the Director of the Board of Governors of the Federal Reserve System, which, according to the Court, construed Regulation Z to require "a written agreement executed by the customer." For reasons I'll get into later, the Court deferred to this interpretation and, as a result, ruled against Wolfington.

Some Initial Thoughts

Before discussing the Third Circuit's application of Kisor/Auer, I'd like to point out a few things about the particular staff interpretation that really made my jaw drop:

  1. It was issued more than four decades ago. See Part 226—Truth in Lending Official Staff Interpretations, 42 Fed. Reg. 40,424, 40,425 (Aug. 10, 1977));
  2. It interpreted a different iteration of Regulation Z--one that didn't contain a "written agreement" requirement at all. The written-agreement language wasn't added to Regulation Z until 1981. See 46 Fed. Reg. 20,848, 20,851 (Apr. 7, 1981); and
  3. It interpreted a different provision of that very outdated version of Regulation Z. As discussed, Wolfington's case centered on 12 C.F.R. 226.2(a)(17), the definition of the term "creditor." The 1977 staff interpretation construed section 226.2(p) of the former Regulation Z, which defined "consumer credit."

In short, it's hard to blame Wolfington for overlooking the fatal flaw that doomed his case before it even started. As the Court candidly acknowledged, the staff interpretation that did him:

dates from 1977 and is buried in the annals of the Federal Register. Although those interpretations are entitled to deference, counsel’s failure to find them was not unreasonable. Instead, counsel raised a reasonable argument, interpreting the text of Regulation Z to require only a “writing ... to confirm what the oral agreement was,” an interpretation the District Court acknowledged was plausible. Thus, counsel’s reliance on the January 20 email as a “written agreement” was not unreasonable, despite ultimately being incorrect.

Poor Wolfington never saw it coming.

As you can probably tell from this introduction, I'm not a fan of the Court's decision granting Auer deference here. I'm going to summarize each step of the Court's analysis before circling back to explain some of the reasons I find its application of the Auer/Kisor framework unconvincing.

The Third Circuit's Reasons For Deferring

The Court began by explaining its view that Regulation Z's reference to a "written agreement" is ambiguous. The problem, as the Court saw it, is that the term's "plain meaning" is in tension with background principles of contract law:

On one hand, the plain text of the term suggests that the extension of credit must be reduced to a fully integrated written instrument. On the other hand, we assume that legislation and regulations are promulgated “against the background of the total corpus juris of the states,” including principles of contract such as the statute of frauds, which requires that a “writing” contain only the essential terms of an agreement. Neither the Act nor Regulation Z defines a “written agreement.” In light of those conflicting principles—the plain text of the regulation and the background of state law—the term “written agreement” is ambiguous.

Next, the Court argued that the 1977 Staff Interpretation is reasonable because "it resolves the ambiguity between the plain text of Regulation Z and state law closer to the former, requiring more than a 'memorandum ... indicating the terms of the oral agreement,' as would be required by the statute of frauds."

"The character and context of the staff interpretation" also weighed in favor of deference. After all, the Court emphasized, "t]he 1977 staff interpretation requiring a formal writing was published in the Federal Register, and the staff reaffirmed its interpretation after Regulation Z was amended to require a 'written agreement.'” Furthermore, the Consumer Financial Protection Bureau (which replaced the Federal Reserve Board as the agency in charge of administering the Act in 2010) reissued that same interpretation without alteration. Thus, concluded the Court, "the staff interpretation constitutes the agencies’ 'official position.'”

The Court was also convinced that the "staff interpretation implicates the agencies’ substantive expertise." Wolfington argued that the scope of a “written agreement” is an interpretive issue that “fall[s] more naturally into a judge’s bailiwick.” Kisor v. Wilkie, 139 S. Ct. 2400, 2419 (2019). The Court disagreed, concluding that Wolfington "ignore[d] the relationship between the scope of a 'written agreement' and the implementation of the Act and Regulation Z":

That implementation is uniquely within the Board’s province, as the scope of the “written agreement” requirement affects the efficient enforcement of the Act and the extent of creditors’ disclosure duties.

Finally, the Court held that the "staff interpretation reflects the agencies’ 'fair and considered judgment.'" (quoting Kisor, 139 S. Ct. at 2417). It emphasized that "[t]he requirement of a formal writing has been enforced by two different agencies for more than forty years and has been reaffirmed repeatedly both in staff interpretations and by the incorporation of the requirement in Regulation Z."

Accordingly, the Court deferred to the 1977 Staff Interpretation under Kisor.

My Objections To The Court's Kisor/Auer Analysis

Genuine Ambiguity. On ambiguity, the Court points to a supposed conflict between the plain meaning of "written agreement" and state contract law. Although the Court doesn't spell it out, its reasoning appears to go something like this:

  1. The plain meaning of "written agreement" would require a "fully integrated written instrument";
  2. A fully integrated written instrument would (the Court must be assuming) require a signature or some other signal of execution by the customer;
  3. State law principles (i.e., the statute of frauds) point the other direction. Even when contract law requires a writing, it only demands that writing include the essential elements of the deal. A signature isn't one of those essential elements, so state law principles wouldn't impose a signature requirement.
  4. Because these two bits of evidence point in different directions, the regulation is genuinely ambiguous.

I disagree with every step of that analysis except the third. Regarding steps one and two, the Court never explains why the ordinary meaning of "written agreement" must include a signature requirement. Instead, it says the ordinary meaning of the term "suggests that the extension of credit must be reduced to a fully integrated written instrument." The only support offered for that view is a cite to the definitions of "formal agreement" and "contract" from Black's Law Dictionary. Of course, the regulation says written agreement, not formal agreement or contract. The fact that Black's includes a signature requirement only in the definition of "formal agreement" (the definition of "contract" doesn't mention a signature requirement at all) thus undermines--and certainly doesn't support--the Court's plain-meaning argument.

As mentioned, I agree with step three of the Court's reasoning. Familiar principles of contract law counsel against reading a signature/execution requirement into Regulation Z's reference to a "written agreement." What the Court doesn't mention, however, is that Regulation Z's "rules of construction" make that state law controlling over the issue here. Section 226.2(b)(3), which appears just a few lines below the definition of "creditor" itself, says:

Rules of construction. For purposes of this regulation, the following rules of construction apply:
(3) Unless defined in this regulation, the words used have the meanings given to them by state law or contract.

12 C.F.R. § 226.2(b)(3).

As the Court acknowledges, neither the Truth in Lending Act nor Regulation Z defines "written agreement." It therefore follows that the term "ha[s] the meaning[] given to [it] by state law or contract." Id. In other words, even if the term "written agreement" were ambiguous in the sense the Court describes, the plain language of Regulation Z's rules of construction demanded the Court look to state law/contract principles--not a decades-old agency staff interpretation of a defunct regulatory definition of a different term that never even uses the words "written agreement."

Reasonableness. These same considerations also refute the Court's argument that the 1977 Staff Interpretation is reasonable because "it resolves the ambiguity between the plain text of Regulation Z and state law closer to the former, requiring more than a 'memorandum ... indicating the terms of the oral agreement,' as would be required by the statute of frauds." Even if they didn't, though, the Court's argument would still be unconvincing for other reasons. First, again, the 1977 Staff Interpretation wasn't interpreting the term "written agreement" because that term wasn't even part of the regulation in 1977. Second, even assuming the 1977 Staff Interpretation had construed the term written agreement, its demand for "more than a memorandum ... indicating the terms of the oral agreement" wouldn't support the Court's decision to impose a signature requirement. The Court never explains how it is so sure that "more than a memorandum ... indicating the terms of the oral agreement" means "more than a memorandum ... indicating the terms ... and also a signature."

Character And Context of the Regulation. The Court's repeated emphasis on the fact that the Board and the CFPB have repeatedly reaffirmed the 1977 staff interpretation doesn't help either. True, after the Board added the "written agreement" requirement to Regulation Z, it explained that a “letter that merely confirms an oral agreement does not constitute a written agreement.”46 Fed. Reg. at 50,293. But demanding more than a letter confirming an oral agreement obviously isn't the same thing as requiring a signature. Thus, the fact that the Board "reaffirmed" the 1977 Staff Interpretation in this limited sense is beside the point. For the same reasons, the fact that the CFPB "reissued the [1981] Federal Reserve Board staff interpretation [(the one just discussed, which never mentions a signature requirement)] verbatim," is also irrelevant. 12 C.F.R. pt. 1026, supp. I, cmt. 2(a)(17) (2019), available at

The Court's conclusion that the 1977 staff interpretation embodies the Board's considered judgment, which rests on the same regulatory-history argument, fails for the same reasons.

Agency Expertise. The Court also concluded that the staff interpretation implicated the Board's substantive expertise. Wasn't Wolfington right, though, that the meaning of a legal term like "written agreement" “fall[s] more naturally into a judge’s bailiwick”? Kisor, 139 S. Ct. at 2419. According to the Court, Wolfington "ignore[d] the relationship between the scope of a 'written agreement' and the implementation of the Act and Regulation Z":

That implementation is uniquely within the Board’s province, as the scope of the “written agreement” requirement affects the efficient enforcement of the Act and the extent of creditors’ disclosure duties.

Of course the implementation of the Act and Regulation Z is uniquely within the Board's province. That isn't the question, though. The question is whether the Board's subject-matter expertise gives it some sort of comparative advantage over courts when it comes to construing the term "written agreement." The Court never addresses that issue at all. If the agency expertise the Supreme Court referred to in Kisor extended automatically to anything and everything touching on "the implementation of the Act and Regulation," nothing would ever "fall more naturally into a judge's [as opposed to the agency's] bailiwick." Id.

I strongly disagree with the Court's decision to defer to the 1977 Staff Interpretation under Auer. Please let me know what you think, especially if you think my objections to the Third Circuit's analysis are misguided. I sure hope they are.

Amazon, Inc. v. Comm'r of Internal Revenue, 934 F.3d 976 (9th Cir. Aug. 15, 2019)*

The question presented was whether the definition of "intangible" in the 1994/1995 version of Treas. Reg. § 1.482-4(b) (which Congress rendered obsolete with a 2017 amendment to the relevant statutory provision) covered residual-business assets. After exhausting the traditional tools of construction, including an interesting and in-depth discussion of the regulation's drafting history, the Ninth Circuit concluded there was "not much room left for the Commissioner's interpretation" of the regulation. "But even if there were genuine ambiguity," the Court added, Auer deference would have been inappropriate anyway because regulated entities like Amazon didn't have fair notice of the agency's new interpretation of the regulation.

The Commissioner had announced the new interpretation for the first time in his briefs in the case, and the Court noted that the risk of unfair surprise to regulated entities is "most acute" where, as here, the agency announces an interpretation for the first time in an enforcement proceeding. Accordingly, even if the regulation had been ambiguous (and, presumably, even if the Commissioner's interpretation were reasonable) the Court would have refused to defer to it under Auer/Kisor.

In my view, the Court in this case did a great job of applying the Kisor framework. Before even addressing whether deference was appropriate, the Court worked through all the traditional tools of construction on its own and independently decided that Amazon's interpretation was far more persuasive than the Commissioner's. Only then, after concluding that no genuine ambiguity existed, did the Court explain that the unfair-surprise problem would have prohibited deference anyway.

I've read other commentators' analyses of this case, and many view it as an example of a case where the Court actually found the regulation ambiguous and refused to defer anyway. I think they're misreading the case. True the Court says the definition of "intangible" doesn't provide a clear answer to the question on its own. As the Supreme Court explained in Kisor, however, courts are not supposed to wave the ambiguity flag until they have exhausted the entire judicial toolkit, including a rigorous assessment of the entire regulatory scheme and the history of the regulation. That is exactly what the Ninth Circuit did here: When the plain text of the regulatory definition in isolation didn't resolve the issue, the Court proceeded to analyze the rest of the regulatory scheme and its drafting history. Once that analysis was complete, no ambiguity remained.

N. Carolina Div. of Servs. for Blind v. U.S. Dep't of Educ., 2019 WL 3997009, at *1 (M.D.N.C. Aug. 23, 2019) (report and recommendation later adopted by district court)

Here's the Magistrate Judge's introduction (cleaned up):

Moving pursuant to the Randolph-Sheppard Act, 20 U.S.C. §§ 107-107f (the “RSA”), and the Administrative Procedure Act, see5 U.S.C. §§ 701-706, the State of North Carolina Division of Services for The Blind (the “Plaintiff” or “NCDSB”) seeks judicial review of an “Opinion and Award” dated September 26, 2017 (the “Arbitration Award”). In response, Lloyd Chadwick Hooks (the “Defendant”), a blind vendor under the RSA, seeks confirmation of both the Arbitration Award and a supplemental arbitration award dated January 2, 2018 (the “Supplemental Award”). For the reasons that follow, the Court should affirm in part and vacate in part the Arbitration Award and affirm the Supplemental Award as specified herein.

Kisor came up in the Magistrate Judge's discussion of NCDSB's argument that as the state agency tasked with implementing the RSA, its longstanding interpretation of certain state regulations was entitled to Auer deference. The Magistrate Judge disagreed, concluding that the state regulation unambiguously foreclosed NCDSB's interpretation.

In a footnote, it added that "even assuming that the Regulation qualified as ambiguous, Plaintiff has not established that this purported interpretation reflects the 'authoritative, expertise-based, fair[, and] considered judgment'" of the agency. (quoting Kisor 139 S. Ct. at 2414). The Magistrate Judge explained that "[a]s support for its purported policy, [NCDSB] cites only to the conclusory statement in the Arbitration Award’s dissenting opinion that [NCDSB] interprets the Regulation as authorizing 'a single face-to-face event with two scoring components' ... and to [an NCSDB official's] testimony at the full evidentiary hearing ...."

The Magistrate Judge wasn't convinced that the "unsupported assertion of either the dissenting arbitrator or Clay Pope (chief of the BEP rather than of Plaintiff) represent[ed]" the agency's authoritative or official position rather than "a more ad hoc statement not reflecting the agency's views." (quoting Kisor, 139 S. Ct. at 2416). Absent proof that the alleged interpretation at the least emanate[s] from those actors, using those vehicles, understood to make authoritative policy in the relevant context,” the Magistrate Judge "could not defer to the purported interpretation even if the Regulation actually suffered from ambiguity." (quoting Kisor, 139 S. Ct. at 2416).

This case is interesting. First, I didn't know that a state agency's interpretation of a statute or regulation could be eligible for Chevron/Auer deference. The Court doesn't address that issue, so I've put it to the Twitter intelligentsia. We'll see if they can confirm that this is a thing.

Second, while the Magistrate Judge is of course correct that the lack of ambiguity thwarts NCDSB's bid for deference, its reasoning in support of the alternative conclusion that deference would be inappropriate even if the regulation were ambiguous appears questionable. It's true that the interpretation at issue must embody the agency's authoritative or official view. The Magistrate Judge concluded that the "unsupported assertion of either the dissenting arbitrator or [Mr. Pope--the NCDSB official]" wasn't sufficient "proof" that the interpretation NCDSB championed was, in fact, the agency's official position.

But the NCDSB was a party to the case, so it seems there is little reason to doubt its representation to the Court that the interpretation identified by the dissenting arbitrator and the NCDSB official was, in fact, the NCDSB's authoritative view of the law. As the Magistrate Judge points out earlier in the opinion, NCDSB's briefing on the issue had referred to the agency's "20-year 'policy' of interpreting the Regulation" in the same way.

In any event, all of that is ultimately beside the point given that the Magistrate Judge concluded that the regulation's unambiguous language foreclosed NDCSB's interpretation. I flag it only for thoroughness' sake. The whole point of this effort is, after all, to assess how well the lower courts are understanding and applying Kisor's framework.

Landis v. Washington State Major League Baseball Stadium Pub. Facilities Dist., 403 F. Supp. 3d 907 (W.D. Wash. Aug. 19, 2019)

This case involves a classic Auer problem. Indeed, it is one of the examples Justice Kagan offered in her Kisor opinion to introduce the doctrine:

In a rule issued to implement the Americans with Disabilities Act (ADA), the Department of Justice requires theaters and stadiums to provide people with disabilities “lines of sight comparable to those for members of the general public.” 28 C.F.R. pt. 36, App. A, p. 563 (1996). Must the Washington Wizards construct wheelchair seating to offer lines of sight over spectators when they rise to their feet? Or is it enough that the facility offers comparable views so long as everyone remains seated?

Kisor, 139 S. Ct. at 2410 (discussing Paralyzed Veterans of Am. v. D.C. Arena
, 117 F.3d 579, 581-82 (D.C. Cir. 1997)). Although the Court cites Kisor and calls the problem a "classic instance of an ambiguous regulation," it ultimately decides the summary-judgment issue before it without addressing the Kisor/Auer problem:

Thus, this Court is left with a deeply factual dispute to resolve. The parties present the Court with dueling standards presented by the same government agency. Further, this standard does not provide anthropometric dimensions with which to determine the comparability of sightlines. Plaintiffs and Defendants' experts used the same standards to review T-Mobile Field's sightlines and came to different conclusions about whether they comply with the ADA. See Dkt. No. 19 at 8; Dkt. No. 27 at 2. The need for further exploration by the parties is apparent, establishing that summary judgment is not appropriate.

Howell v. Advantage RN, 401 F. Supp. 3d 1078 (S. D. Cal. Aug. 15, 2019)

Doesn't actually apply the Kisor framework. Simply notes that the Department of Labor's interpretation of an FLSA regulation in a notice of proposed rulemaking might not be entitled to deference under Kisor because of unfair surprise resulting from the agency's change in position. I'm not sure I agree with the Court's Kisor-related aside, but because it doesn't appear to have impacted the outcome of the case, I won't dig any deeper here.

Lower Courts Applying Kisor VI
Share this