Braun v. Hhs ( 2021 )

Case: 19-1949 Document: 78 Page: 1 Filed: 06/04/2021
United States Court of Appeals
for the Federal Circuit
______________________
ALLEN R. BRAUN,
Petitioner
v.
DEPARTMENT OF HEALTH AND HUMAN
SERVICES,
Respondent
______________________
2019-1949
______________________
Petition for review of the Merit Systems Protection
Board in No. DC-0752-16-0743-I-2.
______________________
ON PETITION FOR REHEARING EN BANC
______________________
GEORGE CHUZI, Kalijarvi, Chuzi, Newman & Fitch, PC,
Washington, DC, filed a petition for rehearing en banc for
petitioner.
TANYA KOENIG, Commercial Litigation Branch, Civil
Division, United States Department of Justice, Washing-
ton, DC, filed a response to the petition for respondent.
Also represented by BRIAN M. BOYNTON, MARTIN F.
HOCKEY, JR., TARA K. HOGAN.
NED MILTENBERG, National Legal Scholars Law Firm,
Case: 19-1949 Document: 78 Page: 2 Filed: 06/04/2021
2 BRAUN v. HHS
P.C., Bethesda, MD, for amici curiae Peter A. Bandettini,
Peter J. Basser, Jack R. Bennink, Karen Berman, Bibiana
Bielekova, Carson C. Chow, G. Marius Clore, Leonardo Co-
hen, Cynthia E. Dunbar, Charles E. Egwuagu, R. Douglas
Fields, Joseph Frank, Charles Gerfen, Mark Hallett, Ken-
neth A. Jacobson, Elaine S. Jaffe, Stephanie J. London,
Alex Martin, Elisabeth A. Murray, David Lee Robinson,
Eric M. Wassermann, Howard Young, Joshua J. Zimmer-
berg.
______________________
Before MOORE, Chief Judge, NEWMAN, LOURIE, DYK,
PROST, O’MALLEY, REYNA, WALLACH*, TARANTO, CHEN,
HUGHES, and STOLL, Circuit Judges.
Opinion dissenting from the denial of the petition for re-
hearing en banc filed by Circuit Judge NEWMAN.
Opinion dissenting from the denial of the petition for re-
hearing en banc filed by Circuit Judge O’MALLEY.
PER CURIAM.
ORDER
Allen R. Braun filed a petition for rehearing en banc. A
response to the petition was invited by the court and filed
by the United States Department of Health and Human
Services. A committee of scientists concerned about tenure
requested leave to file a brief as amici curiae, which the
court granted. The petition was first referred as a petition
for rehearing to the panel that heard the appeal, and there-
after the petition for rehearing en banc was referred to the
circuit judges who are in regular active service. The court
conducted a poll on request, and the poll failed.
________________________
* Circuit Judge Wallach assumed senior status on
May 31, 2021.
Case: 19-1949 Document: 78 Page: 3 Filed: 06/04/2021
BRAUN v. HHS 3
Upon consideration thereof,
IT IS ORDERED THAT:
The petition for panel rehearing is denied.
The petition for rehearing en banc is denied.
The mandate of the court will issue on June 11, 2021.
FOR THE COURT
June 4, 2021 /s/ Peter R. Marksteiner
Date Peter R. Marksteiner
Clerk of Court
Case: 19-1949 Document: 78 Page: 4 Filed: 06/04/2021
United States Court of Appeals
for the Federal Circuit
______________________
ALLEN R. BRAUN,
Petitioner
v.
DEPARTMENT OF HEALTH AND HUMAN
SERVICES,
Respondent
______________________
2019-1949
______________________
Petition for review of the Merit Systems Protection
Board in No. DC-0752-16-0743-I-2.
______________________
NEWMAN, Circuit Judge, dissenting from denial of the peti-
tion for rehearing en banc.
The full court today denies the petition of Dr. Allen
Braun for en banc review of the panel's holding that the
National Institutes of Health (“NIH”) need not comply with
its own special procedures for tenured employees, and that
these procedures will not be enforced by the court. I write
to point out the concerns raised by this flawed holding, and
to dissent from the court's inaction.
“[T]he primary purpose of tenure [is] serving and
providing a benefit to society by the unimpeded search for
truth and its exposition.” Mark L. Adams, The Quest for
Tenure: Job Security and Academic Freedom, 56 Cath. U.
L. Rev. 67, 81 (2006) (citing Am. Ass’n of Univ. Professors,
Case: 19-1949 Document: 78 Page: 5 Filed: 06/04/2021
2 BRAUN v. HHS
Statement of Principles on Academic Freedom and Tenure
(1940), reprinted in AAUP Policy Documents and Reports
3 (9th ed. 2001)). The NIH adopted a tenure structure for
its scientists, tracking the tenure principles of academia:
“Rather than viewing tenure as a luxury or bonus provided
to faculty without a benefit to the employer, it is more cor-
rectly described as the foundational, legitimating corner-
stone of a university.” Id. at 80.
The NIH describes its system of tenure as designed to
attract the brightest and most gifted scientists to its em-
ploy:
to ensure the highest attainable quality in the sci-
entific staff engaged in intramural research and re-
lated medical care.
Nat. Insts. of Health, Tenure in the NIH Intramural Re-
search Program, https://oir.nih.gov/sourcebook/tenure-nih-
intramural-research-program (March 17, 2015). The NIH
system of tenure, and its contribution to stability and secu-
rity in employment, serve to encourage investigation into
complex and difficult problems; to support exploration of
contentious scientific issues; and to facilitate independence
of thought and action without fear of political or other re-
percussions.
Dr. Braun was a tenured scientist at NIH. 1 As relevant
to this appeal, NIH has designated procedures for review
1 Dr. Braun was employed by NIH in 1984 and re-
ceived tenure in 2003. The record describes him as a world-
recognized expert in the neural bases of language, sleep,
and motor functions, and states that his research at NIH is
reported in over 125 publications and has been cited about
14,000 times. At the time of the removal action here on
appeal, his position was Chief of the Voice, Speech, and
Language Branch; and Senior Investigator, Division of
Case: 19-1949 Document: 78 Page: 6 Filed: 06/04/2021
BRAUN v. HHS 3
of performance concerns of tenured scientists, including
participation of the Central Tenure Committee, as set forth
in the NIH Policy on Performance Management, Discipli-
nary Actions and Administrative Removals for Title 42
Employees (“NIH Policy”):
Tenured scientists must undergo the de-tenuring
process before a performance-based action may be
taken against them See section K-3 [The Tenure
Process].
NIH Policy § H.1 (Termination for Unacceptable Perfor-
mance); J.A. 67. The NIH refused to implement its desig-
nated procedures in terminating Dr. Braun, despite his
requests. After his termination he appealed to the Merit
Systems Protection Board, and the Board held that it
would not consider NIH's non-compliance with the tenure-
required procedures. This court affirmed, holding that only
the procedures set forth in Title 5 for all federal employees
are considered by the MSPB and the court.
I previously explained that the NIH action is contrary
to law and precedent, for “[a]n agency is required to act in
accordance with the procedures it adopts for itself, and the
Board will enforce employee rights derived from such
rules.” Campbell v. U.S. Postal Serv., 75 M.S.P.R. 273, 279
(1997); Stone v. FDIC, 179 F.3d 1368, 1378 (Fed. Cir. 1999)
(“Public employees are, of course, entitled to whatever
other procedural protections are afforded them by statute,
regulation, or agency procedure which is in addition to the
protections afforded by the Constitution.”). See Braun v.
Dep’t of Health & Hum. Servs., 983 F.3d 1295, 1306 (Fed.
Cir. 2020) (Newman, J., dissenting).
This judicial refusal to require compliance with tenure-
mandated protections has implications for the public
Intramural Research, National Institute of Deafness and
Other Communication Disorders.
Case: 19-1949 Document: 78 Page: 7 Filed: 06/04/2021
4 BRAUN v. HHS
interest in preserving NIH as a premier research institu-
tion. As summarized by the amici curiae, “undermining
the tenure system will have three noxious side-effects”
likely to manifest themselves immediately and in the long
run:
First, undercutting NIH’s tenure system will dis-
suade senior NIH scientists from remaining there;
Second, undermining NIH’s tenure system will im-
pede NIH’s ability to recruit qualified scientists to
replace ones who leave;
Finally[,] NIH’s impeded ability to retain and re-
cruit top-notch scientists will render NIH less able
to protect the Nation’s safety and health.
Amici curiae Committee of Scientists Concerned About
Tenure, Br. 8.
As federal judges we understand the power of our con-
stitutional tenure to protect independence of thought and
action, free of bias, pressure, and political influence. Jus-
tice Harlan in 1891 wrote:
Whoever is here clothed with a judicial office,
which empowers him to judge in any case affecting
the life, liberty, or property of the citizen, cannot be
restrained from the fearless exercise of its duties
by any apprehension of removal or suspension, in
case he should come athwart the will or pleasure of
the appointing power.
McAllister v. United States, 141 U.S. 174, 195 (1891). The
amici curiae summarize the corresponding purposes of ten-
ure at NIH:
[J]ust as judicial tenure (1) boosts intellectual “in-
dividualism” amongst judges, (2) inspires “public
confidence” in them, and (3) fosters the attraction
and retention of “well-qualified persons” to the
bench, academic tenure does exactly the same
Case: 19-1949 Document: 78 Page: 8 Filed: 06/04/2021
BRAUN v. HHS 5
three things for scholars and for institutions like
NIH.
Amici curiae Br. 14. The goal is to employ the highest lev-
els of talent, quality, and experience, to achieve at the NIH
the benefits of academic tenure, in ultimate service to the
nation:
[Without the] job security tenure pro-
vides . . . much experiment, scholarship and intel-
lectual risk would not be undertaken. Job security
not only allows the faculty member to pursue the
controversial, but also to investigate matters that
present a high probability of failure, including
those particular to the sciences, where failure can
occur after years and even decades of research.
James. J. Fishman, Tenure and its Discontents: The Worst
Form of Relationship Save All of the Others, 21 Pace L. Rev.
159, 182–83 (2000).
Today, as the nation increasingly relies on the NIH for
study of the most complex problems of humanity, our rul-
ing that NIH’s tenure protections will not be enforced by
the courts, warrants review en banc. From the court’s de-
nial of review, I respectfully dissent.
Case: 19-1949 Document: 78 Page: 9 Filed: 06/04/2021
United States Court of Appeals
for the Federal Circuit
______________________
ALLEN R. BRAUN,
Petitioner
v.
DEPARTMENT OF HEALTH AND HUMAN
SERVICES,
Respondent
______________________
2019-1949
______________________
Petition for review of the Merit Systems Protection
Board in No. DC-0752-16-0743-I-2.
______________________
O’MALLEY, Circuit Judge, dissenting from the denial of
petition for rehearing en banc.
Because the Department of Health and Human Ser-
vices (“HHS”) is bound by its own clear policy, I respectfully
dissent from the denial of rehearing en banc. The panel
majority in this case misinterpreted the HHS policy and
spilled substantial ink contemplating whether Dr. Braun
deserved termination. We need not have waded into those
murky waters. HHS granted Dr. Braun tenure and drafted
detailed procedures for the removal of tenure. It then chose
not to follow those procedures. We should rehear this case
en banc and require HHS to adhere to its own procedures.
Case: 19-1949 Document: 78 Page: 10 Filed: 06/04/2021
2 BRAUN v. HHS
BACKGROUND
In 2003, Dr. Allen Braun received tenure at the Na-
tional Institutes of Health (“NIH”), an agency within HHS,
pursuant to 42 U.S.C. § 209(f). Tenure at NIH grants cer-
tain benefits and job protections, including a detailed de-
tenuring procedure which applies in the “rare event” of re-
moval of tenure. NIH Policy on Performance Management,
Disciplinary Actions and Administrative Removals for Ti-
tle 42 Employees (NIH Policy), § K(3) (2007). Yet, when
NIH removed Dr. Braun from his position in 2016, it did
not follow that procedure.
THE NIH POLICY
An agency must follow its own procedures. See Fort
Stewart Schs. v. Fed. Lab. Rels. Auth., 495 U.S. 641, 654
(1990). Even where its procedures are more generous to
beneficiaries or more onerous on the agency than required
by statute, the agency is bound to follow them. See Vitarelli
v. Seaton, 359 U.S. 535, 539 (1959); Voge v. United States,
844 F.2d 776, 779 (Fed. Cir. 1988) (“It has long been estab-
lished that government officials must follow their own reg-
ulations, even if they were not compelled to have them at
all . . . .”).
NIH adopted a policy which applies to the removal of
NIH employees, tenured and non-tenured: the NIH Policy
on Performance Management, Disciplinary Actions and
Administrative Removals for Title 42 Employees. (the
“NIH Policy”). Four provisions of the NIH Policy are rele-
vant to this case: § B, § K, § H(1), and § L(1). Section B
makes clear that, while the NIH Policy covers all Title 42
employees, it provides additional benefits and protections
to tenured employees. NIH Policy § B. Section K(3) makes
clear that “[r]emoval of tenure . . . only occurs” through the
de-tenuring procedure established in that section. NIH
Policy § K(3) (emphasis added). Sections H(1) and L(1) ex-
plicitly state that the de-tenuring procedure must be fol-
lowed when a tenured employee is removed for
Case: 19-1949 Document: 78 Page: 11 Filed: 06/04/2021
BRAUN v. HHS 3
unacceptable performance or administrative reasons. NIH
Policy §§ H(1), K(1). I provide the relevant text of these
provisions below.
First, section B of the NIH Policy states that it applies
to all Title 42 employees, irrespective of tenure status. Cer-
tain provisions of the NIH Policy are applicable only to ten-
ured employees and provide additional benefits and job
protections beyond those provided for non-tenured employ-
ees. Section B states, in relevant part:
This policy applies to all Title 42 employees includ-
ing . . . [a]ll employees appointed under 42 U.S.C.
209(f) and 209(g) (tenured and tenure-track scien-
tists in the intramural program appointed under
Title 42 are also covered by these policies and pro-
cedures, but additional rules may apply to them as
noted) . . . .
NIH Policy § B.
Second, section K provides for benefits and job protec-
tions associated with tenure at NIH. It also states NIH’s
policies regarding tenure. For example, § K(1) states, inter
alia, that “it is the policy of the NIH that long, stable, pro-
ductive careers will continue to be the rule.” NIH Policy,
§ K(1). Section K(3) provides a detailed de-tenuring proce-
dure. It is not necessary to delve into the specifics of that
process here, as there is no dispute that the NIH did not
follow that procedure when it removed Dr. Braun’s tenure.
It suffices to note that § K(3) states that:
Removal of tenure is a rare event and only occurs
after thorough review by the IC and the Central
Tenure Committee (CTC), with final decision by
the Deputy Director for Intramural Research.
NIH Policy § K(3) (emphasis added).
Third, section H(1) of the NIH Policy provides for ter-
mination of employees for unacceptable performance. It
Case: 19-1949 Document: 78 Page: 12 Filed: 06/04/2021
4 BRAUN v. HHS
applies to all Title 42 employees but expressly provides ten-
ured employees with additional protections before they
may be removed for unacceptable performance. Section
H(1) states, in relevant part:
When an employee has demonstrated Unaccepta-
ble performance, based on the results of the oppor-
tunity given to improve performance and other
relevant information, the employee's supervisor or
other designated official will prepare a recommen-
dation for termination and a justification in sup-
port of that recommendation. Employees will be
notified in writing of the recommendation and the
reasons for it or thereof. Employees will have the
right to respond orally, in writing, or both, to the
IC Director or an individual designated by the IC
Director to receive the response. The employee
may be represented in this process. Normally a pe-
riod of not less than seven days and not more than
14 days will be given to the employee to provide the
response, with extensions granted, if warranted.
When the employee is a tenured or tenure-track sci-
entist, those specific policies and procedures also
apply. Tenured scientists must undergo the de-ten-
uring process before a performance-based action
may be taken against them. See section K.3. . . .
NIH Policy § H(1) (emphasis added). It is undisputed that
the conduct with which Dr. Braun was charged falls within
the scope of this provision.
Finally, section L(1) of the NIH Policy provides for ter-
minations “for cause” or for administrative reasons. This
provision applies to all Title 42 employees, as provided in
§ B. It explicitly references the de-tenuring process in ref-
erence to terminations for administrative reasons but is si-
lent as to the interplay between tenure and terminations
“for cause.” Section L(1) provides, in relevant part, that:
Case: 19-1949 Document: 78 Page: 13 Filed: 06/04/2021
BRAUN v. HHS 5
Appointments may be terminated before the expi-
ration date for cause, e.g., personal or scientific
misconduct. Under certain rare and extraordinary
circumstances appointments may be terminated []
for administrative reasons. Terminations for ad-
ministrative reasons may be made only for pro-
grammatic reasons, e.g., lack of funds, re-direction
of program resources. However, a tenured scientist
may not be terminated for administrative reasons
without going through the de-tenuring process. Ti-
tle 42 employees recommended for termination will
be notified in writing of the IC's recommendation
and the reasons thereof. IC Directors may delegate
the authority to issue such recommendations. Em-
ployees will have the right to respond orally, in
writing, or both, to the IC Director, and to be rep-
resented in this process. Normally, a period of not
less than seven days and not more than 14 days
will be provided for response, with extensions
granted, if warranted. The employee will receive a
written decision from the IC Director.
NIH Policy § L(1). As discussed below, the silence regard-
ing tenure in the context of “for cause” proceedings is un-
derstandable and does not carry the implications the
majority believes. “As one court has aptly put it, ‘[n]ot
every silence is pregnant.’” Burns v. United States, 501
U.S. 129, 136 (1991) (quoting Ill. Dep't of Pub. Aid v.
Schweiker, 707 F.2d 273, 277 (7th Cir.1983)), abrogated on
other grounds by United States v. Booker, 543 U.S. 220(2005).
THE PANEL MAJORITY’S OPINION
The panel majority found that, “[a]s a matter of unam-
biguous meaning, de-tenuring is not required under the
NIH Policy for a termination that comes within § L(1),” the
section that provides for “for cause” removal. Braun v.
Dep’t of HHS (Maj. Op.), 983 F.3d 1295, 1301 (Fed. Cir.
Case: 19-1949 Document: 78 Page: 14 Filed: 06/04/2021
6 BRAUN v. HHS
2020). It found that the section’s silence as to de-tenuring
meant that removals “for cause”—even where the only
“cause” relied upon is performance related—do not require
de-tenuring. Id. The majority found that its interpretation
was “reinforced by the NIH Policy’s express inclusion of de-
tenuring requirements for two other bases of removal,” re-
moval for unacceptable performance in § H(1) and removal
for administrative reasons in § L(1). Id. It found further
confirmation of its interpretation in “the statutory scheme
that the NIH Policy’s own structure clearly echoes,” chap-
ters 43 and 75 of title 5 of the United States Code. Id. at
1302. I disagree.
DISCUSSION
The normal rules of statutory interpretation apply to
interpretation of agency procedures. See Roberto v. Dep’t
of Navy, 440 F.3d 1341, 1350 (Fed. Cir. 2006). Applying
those rules of construction here reveals that the NIH Policy
unambiguously requires § K(3)’s de-tenuring procedure be-
fore removing an employee’s tenure in all cases, and par-
ticularly those where, as here, the grounds for the removal
are performance based.
The panel majority reads § L(1)’s silence as to the role
of the de-tenuring process in a “for cause” termination to
mean that the de-tenuring process does not need to be fol-
lowed in any circumstances that NIH classifies as based on
“cause.” The panel majority then goes further and con-
cludes that, where conduct falls within both the “unac-
ceptable performance” contemplated in § H(1) and the
concept of “for cause” in § L(1), NIH may choose to proceed
under § L(1) and ignore the requirements of § H(1) and
§ K(3). But that interpretation ignores the explicit state-
ment in § K(3) that “[r]emoval of tenure . . . only occurs”
through the de-tenuring process and it ignores the specific
provisions for tenured employees in § H(1). NIH Policy
§ K(3) (emphasis added); NIH Policy § H(1). One
Case: 19-1949 Document: 78 Page: 15 Filed: 06/04/2021
BRAUN v. HHS 7
provision’s silence should not eviscerate other provisions’
explicit instructions.
This reading is confirmed by application of the gen-
eral/specific canon of construction, generalia specialibus
non derogant. “[I]t is a basic principle of statutory con-
struction that a specific” provision “controls over a general
provision . . . particularly when the two are interrelated
and closely positioned[.]” HCSC-Laundry v. United States,
450 U.S. 1, 6 (1981). Section L(1) is a general provision
which applies to both tenured and non-tenured employees.
Section K(3) is a specific provision which provides addi-
tional procedural protection only to tenured employees.
Section H(1) similarly provides specific protection to ten-
ured employees. There is no conflict between the provi-
sions. There is an understandable silence in the general
provision because there are explicit instructions in the spe-
cific provisions. The specific explicit tenure provisions
should prevail over the general provision. This reading
also comports with the express policy of NIH “that long,
stable, productive careers will continue to be the rule” and
that removal of tenure will only occur rarely. NIH Policy
§§ K(1), K(3).
Although the negative-implication canon, expressio
unius est exclusio alterius, which states that expression of
one thing implies the exclusion of others, might apply to
§ L(1)’s failure to expressly provide for de-tenuring in a “for
cause” removal in a vacuum, that canon should not apply
here, where § K(3) expressly provides for de-tenuring in all
cases and § H(1) provides for de-tenuring whenever the re-
moval relates to the performance of ones duties as a ten-
ured scientist. See Orlando Food Corp. v. U.S., 423 F.3d
1318, 1325 (Fed. Cir. 2005) (“[T]he maxim expressio unius
est exclusio alterius is not useful when its application would
produce a result that is inconsistent with the plain lan-
guage of the statute.”); see also U.S. v. Polanco, 451 F.3d
308, 311 (3d Cir. 2006) (“Inclusio unius est exclusio alterius
is a key canon in our interpretive arsenal, but we do not
Case: 19-1949 Document: 78 Page: 16 Filed: 06/04/2021
8 BRAUN v. HHS
deploy it when it produces a patently absurd result or when
there is a direct statutory provision on point.”).
Because § L(1) is silent as to whether de-tenuring must
proceed a tenured scientist’s termination “for cause,” the
panel majority reasons that “the NIH Policy’s express in-
clusion of de-tenuring requirements for two other bases of
removal,” removal for unacceptable job performance under
§ H(1) and removal for administrative reasons under
§ L(1), reinforces its interpretation that de-tenuring is not
required for any “for cause” termination, regardless of the
claimed “cause.” Maj. Op. at 1301. But that reasoning
overextends the negative-implication canon to counteract
§ K(3)’s explicit statement that removal of tenure occurs
only through the de-tenuring procedure and ignores the
fact that a “performance-based action” under § H(1) may
only occur after the de-tenuring process. The NIH need not
have repeated itself in § L(1) to specify that de-tenuring
must proceed a performance-based “for cause” removal. It
had already made that abundantly clear in §§ K(3) and
H(1).
The panel majority’s reference to chapters 43 and 75 of
title 5 of the United States Code also cannot overcome a
plain reading of the NIH policy. None of the sections cited
by the panel majority reference tenure or provide for a de-
tenuring procedure. Even if the NIH drafted its own policy
with an understanding of those provisions, it is clear the
NIH adopted tenure protections that exceed those in the
statutes. Those tenure protections should not go by the
wayside where following them would be inconvenient or
cumbersome for NIH.
I express no opinion on the policy implications of tenure
protections, or the lack thereof, for NIH professionals. I
agree with NIH that challenges to the wisdom of its tenure
policy are beyond this court’s purview. It is for NIH to de-
cide its own tenure policy, within the bounds of the law. It
has done so. It should be held to that policy. The majority’s
Case: 19-1949 Document: 78 Page: 17 Filed: 06/04/2021
BRAUN v. HHS 9
conclusion that NIH may choose to ignore its tenure poli-
cies even when they clearly apply—as the majority con-
cedes they do here—renders NIH’s tenure provisions a
nullity in most instances. 1
I also pass no judgment on the severity of Dr. Braun’s
conduct. It is not for us to determine whether Dr. Braun
should have been terminated, and, thus, de-tenured. This
court is only called on to determine if NIH’s policies require
a de-tenuring process before removal of Dr. Braun’s tenure,
regardless of the reasons. They do.
CONCLUSION
I dissent from the denial of en banc in this matter. We
have not been called on to weigh the severity of Dr. Braun’s
conduct and pass judgment on him, or to invent an easy
way for NIH to do so. Rather, this case presents a much
simpler issue—whether NIH must be held to its own un-
ambiguous procedures. I believe it must.
1 The majority seems to understand that its holding
would eviscerate the protections in §§ K(3) and H(1). See
Maj. Op. at 1303. It attempts to ameliorate this fact by
saying that its holding that § L(1) encompasses the same
activities covered by § H(1) only applies to non-routine per-
formance failures. Id. But the majority cites nothing in
the language of the policy that justifies that distinction.