Reply Brief — James L. Kisor, Petitioner v. Robert Wilkie, Secretary of Veterans Affairs

No. 18-15

In the Supreme Court of the United States

JAMES L. KISOR,

Petitioner,

v.

ROBERT L. WILKIE,

Secretary of Veterans Affairs,

Respondent.

On Writ of Certiorari to the United States

Court of Appeals for the Federal Circuit

REPLY BRIEF FOR PETITIONER

KENNETH M. CARPENTER

PAUL W. HUGHES

Carpenter Chartered

1525 SW Topeka Blvd.,

Suite D

Topeka, KS 66601

(785) 357-5251

Counsel of Record

EUGENE R. FIDELL

Yale Law School

Supreme Court Clinic

127 Wall Street

New Haven, CT 06511

(203) 432-4992

MICHAEL B. KIMBERLY

ANDREW J. PINCUS

CHARLES A. ROTHFELD

E. BRANTLEY WEBB

ANDREW A. LYONS-BERG

Mayer Brown LLP

1999 K Street, NW

Washington, DC 20006

(202) 263-3000

phughes@mayerbrown.com

RACHEL R. SIEGEL

Mayer Brown LLP

1221 Ave. of the Americas

New York, NY 10020

(212) 506-2500

Counsel for Petitioner

i

TABLE OF CONTENTS

Table of Authorities.................................................... ii

Reply Brief for Petitioner ............................................1

I. No deference is warranted to the VA’s

construction. .........................................................1

A. Auer is incompatible with the APA. ..............1

B. Skidmore is most consistent with the

APA. ................................................................8

II. Stare decisis does not justify retention of

Auer deference. ................................................... 12

III. Petitioner offers the best reading of

Section 3.156(c). .................................................. 17

Conclusion ................................................................. 25

ii

TABLE OF AUTHORITIES

Page(s)

Cases

Accardi v. Shaughnessy,

347 U.S. 260 (1954) ................................................ 7

Alaska Dep’t of Envtl. Conservation v. EPA,

540 U.S. 461 (2004) .............................................. 11

Christopher v. SmithKline Beecham Corp.,

567 U.S. 142 (2012) .............................................. 10

City of Arlington v. FCC,

569 U.S. 290 (2013) ........................................ 1, 2, 5

Decker v. Northwest Envtl. Def. Ctr.,

568 U.S. 597 (2013) .............................................. 13

Exxon Shipping Co. v. Baker,

554 U.S. 471 (2008) .............................................. 13

Federal Express Corp. v. Holowecki,

552 U.S. 389 (2008) ........................................ 11, 12

Gonzales v. Oregon,

546 U.S. 243 (2006) .............................................. 11

Henry Schein, Inc. v. Archer & White Sales, Inc.,

139 S. Ct. 524 (2019) ............................................ 10

League of United Latin Am. Citizens v. Perry,

548 U.S. 399 (2006) .............................................. 19

Lennox v. Principi,

353 F.3d 941 (Fed. Cir. 2003) ................................ 5

Long Island Care v. Coke,

551 U.S. 158 (2007) .............................................. 16

Martin v. Occupational Safety & Health

Review Comm’n,

499 U.S. 144 (1991) ................................................ 4

iii

Cases—continued

Martinez v. Holder,

740 F.3d 902 (4th Cir. 2014) .................................. 7

Morton v. Ruiz,

415 U.S. 199 (1974) ................................................ 5

Nielson v. Shinseki,

607 F.3d 802 (Fed. Cir. 2010) ................................ 6

NLRB v. Bell Aerospace Co.,

416 U.S. 267 (1974) ................................................ 6

NLRB v. Curtin Matheson Sci., Inc.,

494 U.S. 775 (1990) ................................................ 6

Old Chief v. United States,

519 U.S. 172 (1997) .............................................. 19

Pearson v. Callahan,

555 U.S. 223 (2009) .............................................. 13

Perez v. Mortgage Bankers Ass’n,

135 S. Ct. 1199 (2015) ................................ 2, 3, 7, 9

SEC v. Chenery Corp.,

332 U.S. 194 (1947) ................................................ 6

Skidmore v. Swift & Co.,

323 U.S. 134 (1944) .............................................. 11

United States v. Mead Corp.,

533 U.S. 218 (2001) ...................................... passim

University of Tex. Sw. Med. Ctr. v. Nassar,

570 U.S. 338 (2013) .............................................. 11

Statutes, Rules, and Regulations

5 U.S.C.

§ 551(4) ............................................................... 2, 5

§ 551(5) ........................................................... 2, 5, 7

§ 551(6) ............................................................... 2, 5

iv

Statutes, Rules, and Regulations—continued

5 U.S.C.

§ 551(7) ................................................................... 2

§ 553 ....................................................................... 2

§ 706 ....................................................................... 4

29 U.S.C. § 160 ............................................................ 6

38 U.S.C.

§ 101(35) ............................................................... 19

§ 501(a) ................................................................... 5

§ 5103A(c) ............................................................. 22

§ 5107(a) ............................................................... 23

§ 20.1303................................................................. 7

49 Stat. 803 .................................................................. 6

38 C.F.R. § 3.156(a) ................................. 19, 20, 21, 22

38 C.F.R. § 3.156(c) ........................................... passim

38 C.F.R. § 3.156(c)(1) ......................................... 18, 20

38 C.F.R. § 3.156(c)(1)(i)............................................ 22

Fed. R. Evid. 401 ................................................. 19, 21

Fed. R. Evid. 403 ....................................................... 19

Other Authorities

71 Fed. Reg. 52,455 ............................................... 5, 18

Stephen G. Breyer et al., Administrative Law

and Regulatory Policy (8th ed. 2017) ................ 2, 7

U.S. Dep’t of Veterans Affairs, VA Achieves

Historic Goal by Delivering 81,000 Appeals Decisions (Sept. 18, 2018)............................. 6

Webster’s Third New International

Dictionary (1961)............................................ 20, 21

REPLY BRIEF FOR PETITIONER

Criticism of Auer deference now comes from all

corners. Leading immigration organizations highlight

its harmful effects. NIJC & AILA Br. 21-31. The AFLCIO agrees that “the Court should * * * discard the

strong form of deference represented by Auer” (AFLCIO Br. 16) and, moreover, that deference is not appropriate in this case (id. at 20-24). See also Nat’l

Right to Work Legal Def. Found. Br. 2. Several States

call for overruling Auer—and none defend it.

Even the United States agrees. Acknowledging

that Auer “raises significant concerns” (U.S. Br. 12),

the government now concurs with our principal argument: no deference to the agency is warranted here.

Auer deference circumvents the procedural protections that Congress established in the APA, and it

transfers to agencies substantial lawmaking authority absent any congressional delegation. The Court

should therefore overturn Seminole Rock and Auer.

The government instead asks the Court to adopt a

diminished form of Auer deference—one that looks

like Skidmore but carries Auer’s binding bite. Petitioner would prevail under this rule. That said, the

better course is to overrule Auer outright. What would

remain, Skidmore, properly respects agency expertise.

I.

No deference is warranted to the VA’s construction.

A. Auer is incompatible with the APA.

1. Binding deference (either Chevron or Auer) confers on an agency lawmaking authority, as its actions

have the force of law. See Pet’r Br. 25. In City of Ar-

2

lington v. FCC, 569 U.S. 290 (2013), the Court unanimously agreed that two conditions are necessary for

an agency to exercise this power: (1) “the agency must

have received congressional authority to determine

the particular matter at issue” and (2) the agency

must act in the “particular manner” Congress specified. Id. at 306. See also id. at 321-322 (Roberts, C.J.,

dissenting).

The APA specifies generally-applicable “manners”

by which an agency may act. An agency may promulgate a “rule,” which is defined as an “agency statement of general or particular applicability and future

effect designed to implement, interpret, or prescribe

law or policy.” 5 U.S.C. § 551(4). A “rule” is the result

of a “rule making” process. Id. § 551(5). See Perez v.

Mortgage Bankers Ass’n, 135 S. Ct. 1199, 1203 (2015).

The APA establishes both informal notice-and-comment rulemaking (5 U.S.C. § 553) and formal on-therecord rulemaking (id. §§ 556-557). See Stephen G.

Breyer et al., Administrative Law and Regulatory Policy 466 (8th ed. 2017).1

In two complementary lines of cases, the Court

has held that an agency’s entitlement to binding deference turns on whether it has used congressionallyprescribed procedures. Auer’s fundamental flaw is

Agencies may also act through formal and informal adjudications. See Breyer, supra, at 457. Agency adjudications result in

an “order” (5 U.S.C. § 551(7)), which excludes matter that is

properly the subject of a “rule making” (id. § 551(6)). “Strikingly,

the APA imposes no procedural requirements specific to informal

adjudications.” Breyer, supra, at 474.

1

3

that it provides agencies lawmaking authority in contravention of these settled principles. The government

now agrees. U.S. Br. 19-25.

Focusing on the APA’s protections in Section 553,

Perez recognized that “interpretive rules” are an exception to the general notice-and-comment requirement. “The absence of a notice-and-comment obligation makes the process of issuing interpretive rules

comparatively easier for agencies than issuing legislative rules.” Perez, 135 S. Ct. at 1204. “But that convenience comes at a price: Interpretive rules ‘do not

have the force and effect of law and are not accorded

that weight in the adjudicatory process.’” Ibid.

An agency’s construction of its own ambiguous

regulation is an interpretive rule, and Perez holds that

such rules do not command binding deference. See

Pet’r Br. 31-33; U.S. Chamber Br. 16-19; Prof. Merrill

Br. 7-9, 14-16. Auer thus circumvents the APA’s critical procedural protections. See Pet’r Br. 28-31. According to the government, it is “anomalous that interpretive rules about a statute’s meaning are generally accorded no Chevron deference,” while “interpretive

rules about a regulation’s meaning are accorded Seminole Rock deference.” U.S. Br. 24-25.

And in United States v. Mead Corp., 533 U.S. 218

(2001), the Court emphasized that binding deference

is limited to those agency actions that were “promulgated in the exercise” of the “authority” that Congress

delegated. Id. at 226-227. Mead held that, where there

is no “congressional intent” delegating lawmaking authority to a specific kind of agency procedure, Chevron

does not apply. Id. at 227.

4

Auer deference, however, applies regardless

whether there is any “congressional intent” delegating

lawmaking authority to an agency. See Pet’r Br. 3336. The doctrine therefore bestows power on agencies

in violation of Mead. See U.S. Br. 19-22.

This all accords with the text of the APA. Section

706 provides that “the reviewing court”—and not the

agency—“shall decide all relevant questions of law”

and shall “determine the meaning * * * of the terms of

an agency action.” 5 U.S.C. § 706. See also Pet’r Br.

27; Prof. Merrill Br. 14-16. There is no basis to conclude that Congress silently intended for agencies, not

courts, to decide these issues. See U.S. Br. 16-18.

2. The lone defender of Auer deference—a group

of six professors—offers two main legal contentions.

a. Amici (Admin. Law Scholars Br. 6-8) identify

Martin v. Occupational Safety & Health Review Commission, 499 U.S. 144, 151 (1991), which “presume[d]

that the power authoritatively to interpret its own

regulations is a component of the agency’s delegated

lawmaking powers.” See also id. at 152 (similar). But

this addresses only City of Arlington’s first requirement. The agency must also act in the “particular

manner” that Congress prescribed. Auer violates this

principle.

b. Amici alternatively suggest that one “manner”

for an agency to act with the force of law is via an informal adjudication. They imply that deference is warranted here because the construction of the regulation

stemmed from a single ALJ’s non-precedential opinion, rather than from a considered decision by VA

leadership with general application. See Admin. Law

5

Scholars Br. 21-24 & 28 n.11. That is incorrect. See

U.S. Br. 25 n.5.

First, Congress has never authorized the VA to engage in lawmaking through informal adjudication.

The APA expressly disclaims adjudication as a procedure for an agency “to implement, interpret, or prescribe law or policy.” 5 U.S.C. § 551(4)-(7). See also

Morton v. Ruiz, 415 U.S. 199, 232 (1974) (“No matter

how rational or consistent with congressional intent a

particular decision might be, the determination of eligibility cannot be made on an ad hoc basis.”).

Beyond the APA’s default rules, Congress specified the “particular manner” in which the VA may act.

City of Arlington, 569 U.S. at 306. It may “prescribe

all rules and regulations”—but not orders—“which

are necessary or appropriate to carry out the laws administered by the Department.” 38 U.S.C. § 501(a).

Most relevant here, the VA may promulgate “regulations with respect to the nature and extent of proof

and evidence” necessary “to establish the right to [veterans’] benefits.” Id. § 501(a)(1) (emphasis added).

The regulation at issue, 38 C.F.R. § 3.156(c), was

promulgated expressly pursuant to Section 501(a).

See 71 Fed. Reg. 52,455.

VA informal adjudications do not, accordingly,

qualify for any deference, including Chevron. See, e.g.,

Lennox v. Principi, 353 F.3d 941, 945 (Fed. Cir. 2003)

(“The [Board of Veterans’ Appeals’] explicit interpretation of these statutes and regulations * * * is a legal

6

ruling to be reviewed without deference.”); Nielson v.

Shinseki, 607 F.3d 802, 805 (Fed. Cir. 2010).2

Second, the Board’s decisions cannot warrant

binding deference, including Chevron, because they

are non-precedential.

In Mead, 533 U.S. at 232, the Court denied Chevron deference to an agency action that would not “naturally bind more than the parties to the ruling.” The

Court underscored that, when an “agency makes it

clear that [an action’s] binding character as a ruling

stops short of third parties,” and thus is “conclusive

only as between” the parties directly affected, the

agency cannot be understood to have “ever set out

with a lawmaking pretense in mind.” Id. at 233.

Mead observed that “[a]ny suggestion that rulings

intended to have the force of law are being churned

out at a rate of 10,000 a year * * * is simply self-refuting.” 533 U.S. at 233. So too here, where the Board of

Veterans’ Appeals issued 81,000 decisions in fiscal

year 2018. See U.S. Dep’t of Veterans Affairs, VA

Achieves Historic Goal by Delivering 81,000 Appeals

Decisions (Sept. 18, 2018), perma.cc/5AW5-3CLB. By

Congress may provide specific agencies different authority.

The governing statute in SEC v. Chenery Corp., 332 U.S. 194

(1947), authorized “such rules and regulations and such orders

as” the SEC “may deem necessary or appropriate.” 49 Stat. 803,

833 (emphasis added). The SEC’s flexibility “to act either by general rule or by individual order” (332 U.S. at 202) stemmed from

Congress’s express delegation of that power. Likewise, the National Labor Relations Act (see 29 U.S.C. § 160) at issue in NLRB

v. Bell Aerospace Co., 416 U.S. 267 (1974), authorized the NLRB

to act through adjudications. See NLRB v. Curtin Matheson Sci.,

Inc., 494 U.S. 775, 803 (1990) (Scalia, J., dissenting).

2

7

regulation, all 81,000 of these decisions are “[n]onprecedential.” 38 C.F.R. § 20.1303. Such mass-produced, non-precedential agency actions do not reflect

the “deliberation that should underlie a pronouncement” that has “the effect of law.” Mead, 533 U.S. at

230. See also Martinez v. Holder, 740 F.3d 902, 910

(4th Cir. 2014).

Third, to “amend[]” a regulation, the APA obligates agencies to engage in “rule making.” 5 U.S.C. §

551(5). The APA thus “mandate[s] that agencies use

the same procedures when they amend * * * a rule as

they used to issue the rule in the first instance.” Perez,

135 S. Ct. at 1206. Once an agency proceeds through

notice-and-comment rulemaking, it must return to

rulemaking for its actions to have the force of law. See

Pet’r Br. 36; Accardi v. Shaughnessy, 347 U.S. 260,

265-267 (1954); Breyer, supra, at 485.

*

*

*

Auer deference is irreconcilable with the Court’s

justifications (in Perez, City of Arlington, Mead, and

elsewhere) for affording agencies binding deference.3

And Auer impermissibly bestows on agencies lawmaking authority unbounded by the procedural limitations that Congress imposed in the APA to protect the

regulated public. For these reasons, the Court should

overrule Seminole Rock and Auer.

Amici (Admin. Law Scholars Br. 27) are wrong to suggest that

overturning Auer undermines Chevron. Perez, City of Arlington,

and Mead are all consistent with Chevron. See, e.g., Pet’r Br. 4547; Cato Br. 8-13; Prof. Merrill Br. 12-14; Admin. Law & Fed.

Reg. Profs. Br. 3-13.

3

8

B. Skidmore is most consistent with the

APA.

Acknowledging Auer’s severe flaws, the United

States urges a sharp retreat from Auer. In its view,

courts must give binding deference to an agency’s interpretation of its own ambiguous regulation when

(1) the regulation remains ambiguous after a “searching” analysis using “all the traditional tools of construction,” (2) the agency’s interpretation is “reasonable,” (3) the agency’s interpretation “was issued with

fair notice to regulated parties,” (4) the interpretation

“is not inconsistent with the agency’s prior views,” (5)

it “rests on the agency’s expertise,” and (6) it “represents the agency’s considered view, as distinct from

the views of mere field officials or other low-level employees.” U.S. Br. 12.

Under this Auer-light approach, deference is not

appropriate here. The ALJ’s non-precedential decision

was not “issued with fair notice to regulated parties,”

it does not rest on any agency “expertise,” and it does

not reflect “the agency’s considered view.” Not once

does the government suggest that deference would be

warranted here.

The government’s new position is preferable to existing Auer deference. But the most sound outcome,

both legally and practically, is to overturn Auer.

1. Dialing back Auer deference may reduce some

of its pragmatic harms, but that does not create a legal

foundation. The government’s reformulated Auer doctrine still conflicts with the APA’s procedural requirements (U.S. Br. 22-26), and it lacks congressional authorization (id. at 19-22). The government’s statement

that its new theory would “mitigate”—rather than

9

eliminate—“the tensions between Seminole Rock deference and the APA” (id. at 32) only confirms that the

legal defects would remain.

Additionally, while the government’s approach

may ameliorate some of Auer’s policy flaws, it would

introduce several problems of its own.

We agree that Auer is defective because binding

deference attaches notwithstanding agency “inconsistency.” U.S. Br. 31-33. As a remedy, the government would accord the agency’s first interpretation

binding deference, but not subsequent ones; changes

would require notice-and-comment. Ibid. The government thus seeks to resurrect the Paralyzed Veterans

doctrine that the Court rejected in Perez—that, to

change an initial interpretive rule, an agency must resort to notice-and-comment. Perez, 135 S. Ct. at 12071208. As Justice Scalia recognized, while Paralyzed

Veterans was “a courageous” “attempt to limit the mischief” of Auer, the proper remedy is to overturn Auer

altogether. Id. at 1212 (Scalia, J., concurring in the

judgment).

The government’s approach would also create a

race-to-interpret, where the first administration to

construe a regulation would claim privileged legal status. That would be a bizarre, if not destructive, rule.

Separately, the government asserts that the public must have “fair notice” of a regulation through “familiar, official agency channels.” U.S. Br. 31-33. But

the government fails to specify whether that requires

notice in the Federal Register, opportunity for public

comment, and an obligation of the agency to respond.

In the APA, Congress adopted specific procedures. See

Pet’r Br. 26-31. The proper result is to hold agencies

10

to the APA as written, not to permit agencies to invent

their own procedures that somewhat resemble it.

Tightening the standard for what qualifies as an

“ambiguous” regulation (U.S. Br. 28-29) is neither a

manageable nor sufficient solution. Because there is

no objective measure, parties and courts would be left

adrift as to how much ambiguity triggers deference.

See Henry Schein, Inc. v. Archer & White Sales, Inc.,

139 S. Ct. 524, 531 (2019) (“It is not unheard-of for one

fair-minded adjudicator to think a decision is obvious

in one direction but for another fair-minded adjudicator to decide the matter the other way.”).

Additional practical problems abound. Parties

would dispute whether an agency has actually

brought its “expertise” (U.S. Br. 33-34) to bear, including whether that is a factual or legal question. If factual, discovery disputes would ensue. The government, moreover, offers no standard to evaluate what

constitutes “agency acquiescence.” Id. at 31. And the

distinction between an “agency’s considered view” and

“the views of mere field officials” (id. at 12) would occupy appellate lawyers for years.

More broadly, the Court already imposed the sort

of limitations to Auer that the government envisions.

See Christopher v. SmithKline Beecham Corp., 567

U.S. 142, 155-157 (2012). But SmithKline did not

solve Auer’s legal or pragmatic flaws. See Nat’l Ass’n

of Home Builders (NAHB) Br. 14-16.

2. There is a better way to respect agency expertise (U.S. Br. 42-43) and to provide public guidance

(id. at 39-44), all without transgressing the APA—

Skidmore. See Prof. Merrill Br. 19-24; Cato Br. 15-17.

11

Skidmore addresses the same factors that the government now identifies. It considers whether “the

agency has applied its position with consistency” (Federal Express Corp. v. Holowecki, 552 U.S. 389, 399

(2008)) and the agency’s “formality” of procedure and

“relative expertness” (Mead, 533 U.S. at 228). See also

Alaska Dep’t of Envtl. Conservation v. EPA, 540 U.S.

461, 487 (2004) (respecting views of “an expert administrator”).

But Skidmore differs in one fundamental respect.

Auer (including the government’s diminished form) is

binding deference. Skidmore, however, reflects an

agency’s “power to persuade,” not its power to “control.” Skidmore v. Swift & Co., 323 U.S. 134, 140

(1944). This distinction is crucial. Because it does not

place a substantive thumb on the scale, Skidmore does

not provide agencies lawmaking authority. Rather,

Skidmore is a procedural mechanism that obligates

courts to respect a coordinate branch of government,

including the Executive’s expertise, by giving the

agency’s analysis due consideration. See Mead, 533

U.S. at 234-235. If the court arrives at a result different from the agency, it is incumbent on the court—as

a matter of inter-branch comity—to address the

agency’s reasoning and explain why the court disagrees.

Recent cases are illustrative. Applying Skidmore,

the Court in University of Texas Southwestern Medical Center v. Nassar, 570 U.S. 338, 360-361 (2013),

evaluated the agency’s reasons and, in coming to a

contrary result, explained why. See also Gonzales v.

Oregon, 546 U.S. 243, 269 (2006). Alternatively, in

Alaska Department of Environmental Conservation,

540 U.S. at 487-495, while recognizing that agencies’

12

views do not have “dispositive force,” the Court traced

the agency’s arguments, explained why they were persuasive, and so gave due consideration to the views of

a co-equal branch.4

We agree with the government (at 42-43) and the

Administrative Law Scholars (at 7-9) that agencies often have special expertise. Agencies can exercise that

expertise in a binding way via notice-and-comment

rulemaking. That provides the regulated public an opportunity to respond, negatively or favorably, to the

agency’s views. Outside of notice-and-comment rulemaking, an agency’s expertise deserves procedural respect. If a court disagrees, the court must explain why

with specificity.

The problem with Auer is that it permits an

agency to exert its expertise in binding fashion without any participation by the regulated public. Notwithstanding their expertise, agencies sometimes get

it wrong.

II. Stare decisis does not justify retention of

Auer deference.

Bound by Auer, the court of appeals deferred to

the VA. Now, the government appears to admit that

no deference is warranted. The government’s own approach thus requires departure from stare decisis—

which is no doubt the correct result.

Limited cases could suggest that Skidmore has substantive

weight. See, e.g., Holowecki, 552 U.S. at 403. But Skidmore itself,

and the heartland of its progeny, confirm that the doctrine does

not provide agencies “control.” Holowecki is best understood as

occupying the far end of the “spectrum of judicial responses”

(Mead, 533 U.S. at 228) warranted.

4

13

A. Multiple “special justifications” warrant overturning Seminole Rock and Auer. See Pet’r Br. 47-48.

The failure of Seminole Rock or its progeny to identify

a meaningful legal foundation is reason enough. Id. at

47. So too is the clear contradiction between Auer and

the law governing interpretive rules. Id. at 47-48. Additionally, Seminole Rock/Auer deference is a “direct

obstacle” to the realization of the policies embodied in

the APA. Id. at 48. And, as experience has proven,

Auer deference guts predictability. Ibid.5

B. Stare decisis carries less weight here because

Auer deference is a judge-made rule of construction.

See Pet’r Br. 48-50.

That “Congress remains free to alter or eliminate

the doctrine” (U.S. Br. 36) misses the mark. That was

equally true in Pearson v. Callahan, 555 U.S. 223,

233-234 (2009); as the Prison Litigation Reform Act

illustrates, Congress retains authority to alter rules of

procedure in Section 1983 litigation. The Court nonetheless held that, in the face of a “judge made” “rule,”

“change should come from this Court, not Congress.”

Ibid. See also Exxon Shipping Co. v. Baker, 554 U.S.

471, 507 (2008) (“[T]he judiciary [cannot] wash its

Auer deference also improperly vests agencies with simultaneous authority to both issue and interpret law. See Pet’r Br. 4345; Ctr. for Constitutional Jurisprudence Br. 3-13. The government responds by pointing to justifications for Chevron. U.S. Br.

45-46. But “Congress cannot enlarge its own power through

Chevron—whatever it leaves vague in the statute will be worked

out by someone else.” Decker v. Northwest Envtl. Def. Ctr., 568

U.S. 597, 620 (2013) (Scalia, J., concurring in part and dissenting

in part). By contrast, “when an agency interprets its own rules—

that is something else.” Ibid. It improperly “[c]ombin[es] the

power to prescribe with the power to interpret.” Ibid.

5

14

hands of a problem it created[] simply by calling [the

doctrine] legislative.”).

The government’s attempt to distinguish so-called

primary and secondary conduct rules fails. See U.S.

Br. 36-37. Evidentiary and procedural rules that do

not address “primary conduct” (ibid.) certainly may

bear on case outcomes. But these rules—like deference doctrines—do not warrant full stare decisis

weight when, as here, the public does not rely upon

them. See Pet’r Br. 49-50

C. Stare decisis has substantially less force because the public cannot legitimately rely on Auer deference; indeed, Auer undermines legal stability. See

Pet’r Br. 51-53.

1. The United States responds that “[p]rivate parties have ordered their affairs in reasonable reliance

on” Auer deference. U.S. Br. 37-44. But not a single

regulated party appearing in this Court agrees. The

lone brief defending Auer says nary a word about the

regulated public’s reliance interests, much less stare

decisis as a whole.

Instead, every regulated party to opine has explained that Auer is at war with reliance interests because agency interpretations are not durable. An

agency may shift its position via a press release, a policy memorandum posted to a website, or an amicus

brief.

Business associations explain that Auer deference

“nudges courts to acquiesce in agency actions that disrupt legitimate reliance interests” (NAHB Br. 12-16)

and “upsets the expectations of regulated parties and

deprives them of the notice provided through rulemaking” (U.S. Chamber Br. 3, 6-8). State and local

15

governments demonstrate that Auer “invites dramatic

shifts in federal policy with each new administration.”

State & Local Gov’t Ass’ns Br. 5. The American Immigration Lawyers Association states that “[t]he evils of

Auer are most notable in their effects on the regulated

public.” NIJC & AILA Br. 17-20. And veterans’ associations show that the “VA regularly advances interpretations that are intentionally vague, wildly unreasonable, or inconsistent with its past positions—all

the while demanding that courts defer under Auer.”

Nat’l Org. of Veterans’ Advocates (NOVA). Br. 10-21.

More plainly: “Application of Auer deference to VA

regulations harms veterans.” Nat’l Veterans Legal

Servs. Program (NVLSP) Br. 6-14.

These concerns are not theoretical. Amici have

identified numerous examples of administrative flipflops enabled by Auer deference. See, e.g., State & Local Gov’t Ass’ns Br. 13-19; Nat’l Right to Work Br. 611. They demonstrate several occasions where courts

have endorsed constructions of regulations that are

not the fairest reading. See, e.g., NIJC & AILA Br. 1720; U.S. Chamber Br. 8-14; NAHB Br. 12-16; Wash.

Legal Found. Br. 7-19. States illustrate how Auer undermines their unique, sovereign interests. Utah Br.

11-15. And amici show how agencies have acted in intentionally opaque ways; the VA has admitted as

much. See NOVA Br. 10-13; NVLSP Br. 20-27.

2. The government’s fear of “unsettling decisions

based on Seminole Rock deference” (U.S. Br. 37-41) is

baseless. To start, there is no reason to fear any mass

disruption in the lower courts or among the regulated

public. See Cato Br. 17-19; NAHB Br. 26-29.

In any event, the government’s argument—which

rests on the contention that the regulated public has

16

reliance interests in PLIVA, Chase Bank, Auer, and

Long Island Care (U.S. Br. 38)—defeats itself. Under

the government’s own test, Auer deference would not

attach in any of these cases.

In each of PLIVA, Chase Bank, and Auer, the

Court deferred to agency views expressed in briefs.

That is one of the very ills the government recognizes

with Auer. See U.S. Br. 19. (If the government’s rule

is meant to retain agency lawmaking-by-amicus-brief,

it is not a retraction from Auer at all.)

Long Island Care v. Coke, 551 U.S. 158, 170-171

(2007), hit a trifecta of Auer flaws: the Court deferred

to a non-public advisory memorandum, the memorandum was “written in response” to that litigation, and

the memorandum was inconsistent with the agency’s

past positions. See U.S. Br. 19. The government’s need

to use these cases to make its public reliance argument speaks volumes.

3. The government posits that Auer deference,

once narrowed to preclude agency flip-flops, would enhance predictability by insulating “the agency’s” interpretation from judicial review. U.S. Br. 40. As the

government sees it, courts will all fall in line, precluding review by this Court (id. at 40-42), and private

parties will not have a meaningful opportunity anyway to challenge agency action (id. at 43-44). To the

extent these arguments hold merit (earlier in its brief,

at 26, the government appears to take the opposite

view), the regulated public does not seek predictability in a vacuum. The public is entitled to participate

in, and seek judicial review of, agency actions that impact their interests. The APA provides both predictability and substantive protections. That framework—

and not any variation on Auer—should govern.

17

III. Petitioner offers the best reading of Section

3.156(c).

The government does not dispute that reconsideration under Section 3.156(c) has three elements: (1)

the VA originally erred by failing to consider a service

record in its possession; (2) the service record is “relevant”; and (3) the VA later grants an award, per the

usual standards, “based” at least “in part” on the record it initially overlooked. Pet’r Br. 55.

Nor does the government deny that petitioner has

satisfied the first and third elements: the VA originally erred by failing to consider his service records

(Pet’r Br. 55-56), and, when the VA granted petitioner

benefits, its award was “based” “in part” on those records (id. at 56). See also AFL-CIO Br. 23-24.

Instead, the government contends that service

records are “relevant” only if they demonstrate that

“the VA would have reached a different result at the

time of its prior decision had it considered the additional records” (U.S. Br. 49)—and not if they are “relevant” to the veteran’s claim generally. The government’s two arguments are not persuasive, and they

pale in comparison to our position, much of which the

government disregards.

A.1. The government begins by asserting that the

“matter at hand” in Section 3.156(c) is the “VA’s prior

decision denying the veteran’s claim.” U.S. Br. 49. But

the government has no textual support for that contention.

The regulatory text makes plain that the “official

service department records” must be “relevant” to the

veteran’s “claim.” It provides:

18

[A]t any time after VA issues a decision on a

claim, if VA receives or associates with the

claims file relevant official service department

records that existed and had not been associated with the claims file when VA first decided

the claim, VA will reconsider the claim, notwithstanding paragraph (a) of this section.

38 C.F.R. § 3.156(c)(1). The triggering event, accordingly, is the VA “associat[ing] with the claims file” material that is an “official service department record[]”

and that should have been “associated with the claims

file” the first time. What goes into the “claims file”?

Documents that are “relevant” to the veteran’s

“claim.”

In promulgating the current regulation, this is

precisely how the VA understood it would work: “If a

newly discovered service department record is one

that VA should have received at the time it obtained

the veteran’s service medical records, we believe it ordinarily would be within the scope of proposed [Section] 3.156(c)(1).” 71 Fed. Reg. 52,455 (emphasis

added). A “relevant” record is one that belonged in the

“claims file” at the time of the original adjudication.

Assuming, for the sake of argument, that the “official service department records” must be “relevant”

to the “VA’s prior decision denying the veteran’s

claim” (U.S. Br. 49.), these records still qualify. They

bear on whether an in-service event occurred and the

likelihood that petitioner suffered from PTSD—both

of which were elements of the original adjudication.

See Pet’r Br. 57-58 & n.16. Indeed, Dr. Davies criticized Dr. Henderson’s 1983 report (on which the government solely relies, see U.S. Br. 48) because Dr.

Henderson “misunderstood the impact of [petitioner’s]

19

war trauma upon him.” JA38. Petitioner’s war trauma

was undeniably a fact “of consequence in determining

the action.” Fed. R. Evid. 401. The records documenting it were thus “relevant” ones. Ibid.

The government argues that the service records

are not “relevant” because they reveal facts that “no

one disputed.” U.S. Br. 48. But it is long settled that

evidence is “relevant” even if “‘[t]he fact to which the

evidence is directed’” is not “‘in dispute.’” Old Chief v.

United States, 519 U.S. 172, 179 (1997) (quoting Advisory Committee’s Notes on Fed. R. Evid. 401). Federal Rule of Evidence 403 exists because—and expressly recognizes that—evidence may be both “relevant” and “cumulative.” (By contrast, the regulation

excludes “cumulative” evidence from the meaning of

the defined term “material.” 38 C.F.R. § 3.156(a). See

pages 20-21, infra.)

The government’s actual contention is that the

regulation’s use of “relevant” should be interpreted as

“dispositive.” While the government studiously avoids

the language now, the Board of Veterans’ Appeals focused on whether the service records are “outcome determinative.” Pet. App. 42a-43a. But “relevant” does

not mean “dispositive” or “outcome determinative.”

See Pet’r Br. 57; 38 U.S.C. § 101(35) (defining “relevant evidence” as “evidence that tends to prove or disprove a matter in issue”); League of United Latin Am.

Citizens v. Perry, 548 U.S. 399, 436 (2006) (distinguishing “relevant” and “dispositive”).

2. The government further asserts that the “hallmark of reconsidering * * * is that the VA concludes

that its prior decision on the claim was incorrect ab

initio.” U.S. Br. 49. But the government has no textual

20

support for this contention, either. In fact, the text explains that the “hallmark of reconsidering” (ibid.) is

an earlier VA procedural error—the VA’s failure to

“associate[]” records in the government’s possession

with the “claims file.” 38 C.F.R. § 3.156(c)(1).

Section 3.156(c)(3) strongly supports our argument. To trigger reconsideration, the VA’s award

must be “based” at least “in part” on the overlooked

records. See Pet’r Br. 55. Here, the VA did rely on the

overlooked records when it granted petitioner benefits. JA51-52. See also AFL-CIO Br. 23-24. The government admitted as much earlier. BIO 16 n.2. This

confirms that the evidence is “relevant.”6

B. Several other aspects of the regulatory text and

structure require our construction.

First, the government improperly renders “relevant” identical to the “new and material evidence”

standard that governs reopening under Section

3.156(a). See Pet’r Br. 57-59. “Material evidence” must

relate to “an unestablished fact necessary to substantiate the claim” and may not be “cumulative []or redundant” of existing evidence. 38 C.F.R. § 3.156(a).

That is the same as the government’s preferred test

here. See U.S. Br. 48-49. But, as we have shown, “relevant” is more expansive. See Fed. R. Evid. 401.7

Section 3.156(c)(4) also bolsters our conclusion. See AFL-CIO

Br. 22-23.

6

Webster’s Third New International Dictionary (1961) confirms

that, in ordinary usage, “relevant” and “material” differ in degree. “A thing is RELEVANT when it has a connection * * * with

a matter under consideration.” Id. at 1917. “A thing is MATERIAL when it has so close a relationship with a case in hand that

7

21

The government responds that Section 3.156(a)’s

“new and material evidence” standard is actually

“broader” than and subsumes Section 3.156(c). U.S.

Br. 51. The government alternatively contends that

the different language has the same meaning. Id. at

51-52.

The regulation’s text forecloses these arguments.

It provides that a veteran may satisfy the requirements of Section 3.156(c) “notwithstanding paragraph

(a) of this section.” 38 C.F.R. 3.156(c)(1) (emphasis

added). Section 3.156(c) must apply in circumstances

outside the scope of Section 3.156(a), or else this “notwithstanding” clause would be meaningless.

Additionally, the VA amended “material” out of

Section 3.156(c), replacing it with the broader term

“relevant.” See Pet’r Br. 58-59; AFL-CIO Br. 22. If the

VA had actually intended for reopening to hinge on

the defined concept of “material” evidence (see 38

C.F.R. § 3.156(a)), the VA would have retained that

term.

We made both arguments earlier (Pet’r Br. 58-59),

but the government responds to neither. For these two

reasons, “relevant” must be a broader standard than

“material”—a conclusion irreconcilable with the government’s construction.

Nor does the government’s construction meaningfully distinguish between “relevant” and “material.”

The government asserts that “material” evidence in

Section 3.156(a) “need not suggest that the VA’s prior

decision on the claim was incorrect at the time.” U.S.

it cannot be dispensed with without serious alteration of the

case.” Ibid.

22

Br. 51 (emphasis added). But that is just wordplay;

under the government’s view, “material” and “relevant” mean the same thing. And Section 3.156(c) is

limited to “official service department records” (ibid.)

because of the regulation’s use of those specific words,

not because of the term “relevant.”

Our construction—not the government’s—properly captures the different meaning of these words.

Second, the regulation states unequivocally that

“relevant official service department records” “include” “[s]ervice records that are related to a claimed

in-service event.” 38 C.F.R. § 3.156(c)(1)(i). That is

what these records are. See Pet’r Br. 56-57 & n.15.

Unable to muster a response, the government disregards our argument.

Third, current Section 3.156(c) was promulgated

against the backdrop of then-recently-enacted 38

U.S.C. § 5103A(c). That statute’s usage of “relevant

records” accords with our construction. See Pet’r Br.

59. These two provisions dovetail: Section 5103A(c)

obligates the VA to locate records to assist the veteran, and Section 3.156(c) provides the remedy when

the VA errs. The government fails to explain why “relevant” should have different meanings in these related provisions.

Fourth, we showed that the pro-veteran and antidrafter canons resolve any lingering ambiguity. See

Pet’r Br. 59-61; NOVA Br. 23-29; NVLSP Br. 6-7. Once

more, the government is silent.

C. The government rests heavily on its policy argument—that our construction could leave some veterans in a “better position” than if the VA had made

23

no underlying error. U.S. Br. 52. But the government

fails to address this regulation’s remedial nature.

The government’s construction requires a counterfactual inquiry about what the VA would have

done, often decades earlier, had it not made an error.

Because the veteran bears the burden of proof (see 38

U.S.C. § 5107(a)), that speculative analysis is a recipe

for mass under-compensation of veterans.

It is often impossible to determine with confidence

the precise effect that overlooked evidence would have

had on the original claim adjudication. Not only might

the evidence influence the ultimate decisionmaker,

but it would also be in the record for doctors and therapists to consider when they prepare their medical reports.

Moreover, when seeking reconsideration, a veteran will often present (as petitioner did here) both

the evidence that the VA overlooked and new medical

evidence confirming the injury. Under the government’s approach, VA adjudicators would have wide

discretion to reject claims for retroactive benefits by

heavily weighting the new medical evidence.

These concerns are magnified by the VA’s history

of irrational claims administration (see Sergeant Major Jeff Howard Br. 9-13) and the “VA’s penchant for

unreasonable litigating positions.” NOVA Br. 21-23

(summarizing alarming statistics).

As we construe it, the VA’s reconsideration mechanism operates far more objectively. The veteran

must show that the VA procedurally erred by failing

to evaluate records the government possessed, that

the records are relevant to the veteran’s claim, and

that, when the VA later awarded benefits (pursuant

24

to the traditional standards), it relied in part on the

records it previously overlooked. See Pet’r Br. 55.

The government’s main objection is that this construction could overcompensate some veterans, including, the government insinuates, petitioner. U.S.

Br. 52. But our construction is equitable, both here

and more generally.

In awarding petitioner benefits, the VA credited

Dr. Davies’ 2007 evaluation. JA52. In it, Dr. Davies

explained that petitioner has had PTSD since around

1980 and that Dr. Henderson bungled his analysis.

JA37-38. In sum, the evidence demonstrates that petitioner should have received benefits when he originally applied in 1982 and that the VA erred when it

first adjudicated his claim.

More generally, it was reasonable for the VA to

select a procedure that might overcompensate some

veterans rather than one that would undercompensate many.8 Not only is our construction true to the

regulation’s text, but it also evinces the VA’s interest

in providing a fair procedure for remedying its past

errors. Our veterans deserve no less.

If the VA has since changed its view, it may

amend the regulation through notice-and-comment

rulemaking.

Allocating the risk of loss this way also incentivizes the VA to

avoid errors. Under the government’s approach, VA errors

would, in the aggregate, result in financial advantage to the government at the expense of veterans.

8

25

CONCLUSION

The Court should reverse the judgment entered

below.

Respectfully submitted.

KENNETH M. CARPENTER

PAUL W. HUGHES

Carpenter Chartered

1525 SW Topeka Blvd.,

Suite D

Topeka, KS 66601

(785) 357-5251

Counsel of Record

EUGENE R. FIDELL

Yale Law School

Supreme Court Clinic

127 Wall Street

New Haven, CT 06511

(203) 432-4992

MICHAEL B. KIMBERLY

ANDREW J. PINCUS

CHARLES A. ROTHFELD

E. BRANTLEY WEBB

ANDREW A. LYONS-BERG

Mayer Brown LLP

1999 K Street, NW

Washington, DC 20006

(202) 263-3000

phughes@mayerbrown.com

RACHEL R. SIEGEL

Mayer Brown LLP

1221 Ave. of the Americas

New York, NY 10020

(212) 506-2500

Counsel for Petitioner

MARCH 2019