Fraternal Order of Police, Metropolitan Police Department Labor Committee, D.c.… v. District of Columbia

United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued February 2, 2022 Decided August 19, 2022
No. 21-7059
FRATERNAL ORDER OF P OLICE, M ETROPOLITAN P OLICE
DEPARTMENT LABOR C OMMITTEE, D.C. P OLICE U NION,
APPELLANT
v.
DISTRICT OF C OLUMBIA AND M URIEL B OWSER , IN HER
OFFICIAL CAPACITY AS M AYOR OF THE DISTRICT OF
C OLUMBIA,
APPELLEES
Appeal from the United States District Court
for the District of Columbia
(No. 1:20-cv-02130)
Anthony M. Conti argued the cause for appellant. With
him on the briefs was Daniel J. McCartin.
Holly M. Johnson, Senior Assistant Attorney General,
Office of the Attorney General for the District of Columbia,
argued the cause for appellees. With her on the brief were Karl
A. Racine, Attorney General, Loren L. AliKhan, Solicitor
General, Caroline S. Van Zile, Principal Deputy Solicitor
General, and Carl J. Schifferle, Deputy Solicitor General.

-- 1 of 15 --

2
Before: R OGERS , M ILLETT, and KATSAS , Circuit Judges.
Opinion for the Court filed by Circuit Judge KATSAS .
KATSAS , Circuit Judge: This case involves federal
constitutional challenges to a District of Columbia statute
eliminating the right of D.C. police officers to bargain over
procedures for disciplining individual officers. The police
union contends that the statute violates equal protection
principles, the Bill of Attainder Clause, the Contract Clause,
and the Fifth Amendment Due Process Clause. We reject all
the challenges.
I
The Comprehensive Merit Personnel Act (CMPA)
governs collective bargaining by employees of the District of
Columbia government. It allows officers of the Metropolitan
Police Department, like other D.C. government employees, to
unionize and engage in collective bargaining. D.C. Code
§ 1-617.01(b). They have done so and are represented by the
plaintiff in this case, the Fraternal Order of Police,
Metropolitan Police Department Labor Committee, D.C.
Police Union (FOP).
The CMPA provides that “[a]ll matters shall be deemed
negotiable” except for a list of rights reserved to management.
D.C. Code § 1-617.08(b). Management rights include the right
to “hire, promote, transfer, assign, and retain employees” as
well as the right to “suspend, demote, discharge, or take other
disciplinary action against employees for cause.” Id.
§ 1-617.08(a). The parties have long understood the CMPA to
give management full discretion over whether or how to

-- 2 of 15 --

3
discipline officers who commit wrongdoing, while allowing
for negotiation over the procedures for adjudicating it.
Article 12 of the Metropolitan Police Department’s 2017
collective bargaining agreement contained detailed provisions
on disciplinary procedure. See Collective Bargaining
Agreement Between the District of Columbia Metropolitan
Police Department and the D.C. Police Union, art. 12 (J.A. 90–
95) (2017 Agreement). It also stated that these provisions
“shall be incorporated” into successor agreements unless
modified by a joint labor-management committee or, in the
event of an impasse, an arbitration panel. Id. § 2 (J.A. 91).
The 2017 Agreement expired on September 30, 2020.
Two months earlier, following the death of George Floyd while
in Minneapolis police custody, the D.C. Council passed
emergency legislation setting forth a wide range of police
reforms. See Comprehensive Policing and Justice Reform
Second Emergency Amendment Act of 2020 (Reform Act),
D.C. Act 23-336. At issue in this case is section 116 of the
Reform Act, which temporarily amends the CMPA to eliminate
the right of “sworn law enforcement personnel” to bargain over
disciplinary procedure. See D.C. Code § 1-617.08(c). The
amendment applies to “any collective bargaining agreement
entered into with the Fraternal Order of Police/Metropolitan
Police Department Labor Committee after September 30,
2020.” Id.1
1 As emergency legislation, the original Reform Act expired
after 90 days. Since then, the D.C. Council has re-enacted it seven
times, with the most recent enactment set to expire on September 26,
2022. See D.C. Act 23-336 (July 22, 2020); D.C. Act 23-437 (Oct.
28, 2020); D.C. Law 23-151 (Dec. 3, 2020); D.C. Act 24-76 (May 3,
2021); D.C. Act 24-128 (July 29, 2021); D.C. Law 24-23 (Sept. 3,

-- 3 of 15 --

4
Shortly after section 116 became law, the FOP sued to
enjoin its enforcement. The union raised federal constitutional
challenges based on equal protection principles, the Bill of
Attainder Clause, the Contract Clause, and the Fifth
Amendment Due Process Clause.
The district court rejected these claims and dismissed the
case without prejudice for failure to state a claim. Fraternal
Ord. of Police, Metro. Police Dep’t Lab. Comm., D.C. Police
Union v. District of Columbia, 502 F. Supp. 3d 45 (D.D.C.
Nov. 4, 2020). The FOP then moved to alter the judgment so
that it could amend its complaint. The district court denied the
motion as futile.
The FOP appealed both decisions. We have jurisdiction
under 28 U.S.C. § 1291.
II
We start with the dismissal order. We review the dismissal
of constitutional claims de novo. Patchak v. Jewell, 828 F.3d
995, 1001 (D.C. Cir. 2016).
A
The FOP first raises an equal-protection challenge. The
Equal Protection Clause provides that “[n]o State shall … deny
to any person within its jurisdiction the equal protection of the
laws.” U.S. Const. amend. XIV. The Supreme Court has held
that the Fifth Amendment Due Process Clause extends equal-
protection principles to actions by the D.C. government. See
Bolling v. Sharpe, 347 U.S. 497, 498–99 (1954). According to
2021); D.C. Act 24-370 (Apr. 7, 2022); D.C. Act 24-454 (June 28,
2022).

-- 4 of 15 --

5
the union, section 116 violates equal protection because it
irrationally discriminates between police officers and similarly
situated government employees. We disagree.
Legislation that covers some occupations but not others—
which neither burdens fundamental rights nor makes suspect
classifications—satisfies equal protection if the distinction at
issue is “rationally related to a legitimate state interest.”
Friedman v. Rogers, 440 U.S. 1, 17 (1979) (quoting New
Orleans v. Dukes, 427 U.S. 297, 303 (1976)); see Williamson
v. Lee Optical of Okla., Inc., 348 U.S. 483, 489 (1955)
(optometrists versus opticians). Under rational-basis review,
legislation carries “a strong presumption of validity.” Cent.
State Univ. v. Am. Ass’n of Univ. Professors, 526 U.S. 124, 126
(1999) (limitation on bargaining rights for college professors).
“Perfection in making the necessary classifications is neither
possible nor necessary.” Mass. Bd. of Ret. v. Murgia, 427 U.S.
307, 314 (1976) (police retirement age). Absent irrationality,
a law does not fail rational-basis review for being over- or
under-inclusive. Nordlinger v. Hahn, 505 U.S. 1, 16–17
(1992). And because legislative classifications may be
grounded in “rational speculation unsupported by evidence or
empirical data,” a challenger must negate “every conceivable
basis” that might support the distinction. FCC v. Beach
Commc’ns, 508 U.S. 307, 314–15 (1993) (cleaned up).
The FOP has failed to carry that considerable burden. The
D.C. Council could rationally have concluded that section 116
furthers a legitimate interest in improving police
accountability. By taking disciplinary procedures off the
bargaining table, it gave management more flexibility in
deciding how to consider allegations of police misconduct.
And even if new procedural rules would reduce the protections
for accused officers, “equal protection is not a license for courts

-- 5 of 15 --

6
to judge the wisdom, fairness, or logic of legislative choices.”
Beach Commc’ns, 508 U.S. at 313.
The FOP disputes that police accountability motivated the
Council. The union notes that the Reform Act included no
legislative findings or explanation supporting the choice to
curtail bargaining rights for the police while preserving those
rights for other public-sector workers. The union further
invokes language in a different provision of the legislation—
section 101, which recounts Floyd’s death from a neck restraint
and then states an intent to ban such restraints in the District of
Columbia. See D.C. Code § 5-125.01. According to the FOP,
this language shows that the Council unfairly sought to impute
to D.C. police concerns about misconduct elsewhere.
This argument misunderstands the basics of rational-basis
review. Under that level of scrutiny, the legislature’s actual
motive is “entirely irrelevant”; all that matters is whether there
are “plausible reasons” to conclude that the statutory
classification furthers a legitimate government interest. Beach
Commc’ns, 508 U.S. at 313–15 (cleaned up). Likewise,
because ordinary legislative choices are not subject to
“courtroom fact-finding,” the absence of findings, studies, or
statements of purpose has “no significance.” Id. at 315
(cleaned up). In the wake of Floyd’s death, the Council could
rationally have concluded that the use of neck restraints
“presents an unnecessary danger to the public.” D.C. Code § 5-
125.01. And regardless, it could rationally have concluded that
preserving management control over disciplinary procedures
would improve police accountability.
The FOP objects that section 116 does not apply to prison
guards or protective-services officers. But they differ from
police in key respects. Prison guards, for example, operate in
a highly regimented and supervised environment. See D.C.

-- 6 of 15 --

7
Code § 24-211.02. Protective-services officers safeguard
government agencies and property. See Cannon v. District of
Columbia, 717 F.3d 200, 203 (D.C. Cir. 2013). Given these
differences, the D.C. Council could rationally have concluded
that improving accountability for officers who directly police
the general public on a daily basis was a more pressing concern.
See Lee Optical, 348 U.S. at 489 (“Evils in the same field may
be of different dimensions and proportions, requiring different
remedies.”). Likewise, it could rationally have concluded that
targeting police discipline was an appropriate first step in
improving accountability for all law-enforcement personnel.
Beach Commc’ns, 508 U.S. at 316 (“the legislature must be
allowed leeway to approach a perceived problem
incrementally”). And because police officers make up the
lion’s share of workers that the union claims as similarly
situated, it could rationally have concluded that the amendment
would be at worst slightly under-inclusive. Under rational-
basis review, any of these rationales is good enough.
B
The FOP next contends that section 116 violates the Bill
of Attainder Clause, which provides that “[n]o Bill of Attainder
or ex post facto Law shall be passed.” U.S. Const. art. I, § 9,
cl. 3. This is so, according to the union, because the
amendment singles out “sworn law enforcement officers” for
negative treatment and mentions the FOP by name. See D.C.
Code § 1-617.08(c)(2). We disagree.
A bill of attainder is a law that “legislatively determines
guilt and inflicts punishment upon an identifiable individual
without provision of the protections of a judicial trial.” Nixon
v. Adm’r of Gen. Servs., 433 U.S. 425, 468 (1977). A law
counts as a bill of attainder if it “(1) applies with specificity,
and (2) imposes punishment.” Foretich v. United States, 351

-- 7 of 15 --

8
F.3d 1198, 1217 (D.C. Cir. 2003) (cleaned up). We focus on
the second element, which turns on this three-part inquiry:
(1) whether the challenged statute falls within the
historical meaning of legislative punishment; (2)
whether the statute, viewed in terms of the type and
severity of burdens imposed, reasonably can be said
to further nonpunitive legislative purposes; and (3)
whether the legislative record evinces a[n] … intent to
punish.
Selective Serv. Sys. v. Minnesota Pub. Int. Rsch. Grp., 468 U.S.
841, 851 (1984) (cleaned up). All three considerations cut
against the FOP.
For starters, section 116 lies far from the historical
meaning of legislative punishment. In the 1700s, the British
Parliament used bills of attainder to sentence specific
individuals to death, often as punishment for attempting to
overthrow the government. United States v. Brown, 381 U.S.
437, 441 (1965). Over time, American courts extended the Bill
of Attainder Clause to legislation imposing less severe
punishment, like “banishment, imprisonment, denial of the
right to vote, or confiscation of property.” Kaspersky Lab, Inc.
v. DHS, 909 F.3d 446, 454 (D.C. Cir. 2018). The change made
by section 116—giving management greater control over
procedure for disciplining employees—is not remotely
analogous to any of these historically grounded categories.
The second factor turns on whether the statute has
“punitive purposes” or “merely burdensome effects.”
Kaspersky, 909 F.3d at 455. It parallels the rational-basis
inquiry in some respects, by asking whether the law
“reasonably can be said to further nonpunitive legislative
purposes,” Nixon, 433 U.S. at 475–76, and whether it is
“overbroad” or “underinclusive,” Kaspersky, 909 F.3d at 455–

-- 8 of 15 --

9
56. But we have also described this test as “more exacting”
than rational-basis review. BellSouth Corp. v. FCC, 144 F.3d
58, 67 (D.C. Cir. 1998). Seeing opportunity in that distinction,
the FOP recycles its equal-protection arguments about means-
ends scrutiny.
To no avail. Any differences between police and other law
enforcement officers fall far short of showing the kind of
“grave imbalance” that might suggest a hidden punitive
purpose. Foretich, 351 F.3d at 1222; see also Kaspersky, 909
F.3d at 456 (“the Bill of Attainder Clause does not require
narrow tailoring”). Moreover, aspects of section 116
affirmatively undermine any such inference. For one thing,
that provision leaves in place significant “protective measures”
for police officers, Foretich, 351 F.3d at 1222, such as the right
not to be “fired, demoted, or suspended without cause,” Burton
v. Off. of Emp. Appeals, 30 A.3d 789, 792 (D.C. 2011); see
D.C. Code §§ 1-616.51–52. For another, section 116 “lasts
only temporarily.” See Kaspersky, 909 F.3d at 456. And while
it has now been reenacted on several occasions, each new
iteration has been limited to the 90-day period for emergency
legislation or the 225-day period for temporary legislation, thus
barring long-term change absent later legislative action. See
D.C. Code § 1-204.12(a); United States v. Alston, 580 A.2d
587, 590–91 (D.C. 1990). With full view of what section 116
has and has not changed, we cannot infer that the Council acted
with an illicit punitive purpose.
The third factor asks whether “the legislative record
evinces a[n] … intent to punish.” Selective Serv. Sys., 468 U.S.
at 852. But given the practical and theoretical concerns about
using legislative history to divine a legislature’s “collective
purpose,” this factor has weight only if the record shows
“unmistakable evidence of punitive intent.” Kaspersky, 909
F.3d at 463 (quoting Foretich, 351 F.3d at 1225).

-- 9 of 15 --

10
The FOP offers two arguments to satisfy this heavy
burden. First, it again points to section 101, which it says
betrays an intent to punish D.C. police officers for the
misconduct of officers in Minneapolis. As noted above,
section 101 simply expresses the Council’s intent to
prospectively “ban the use of neck restraints by law
enforcement” in the wake of Floyd’s death from such a
restraint. D.C. Code § 5-125.01. That hardly suggests an intent
to punish anyone. Second, the FOP asserts that the Reform Act
was passed as emergency legislation without any real
emergency. But the Council has significant leeway to pass
emergency legislation to protect the public safety or welfare,
see Alston, 580 A.2d at 590–91, and we deferentially review its
decision to do so, see Barnes v. District of Columbia, 102 A.3d
1152, 1154 (D.C. 2014). The union makes no serious effort to
show that the Council acted beyond its discretion. Again, we
can discern no express or hidden intent to punish.
C
The FOP next contends that section 116 violates the
Contract Clause, which provides that “[n]o state shall . . . pass
any . . . Law impairing the Obligation of Contracts.” U.S.
Const. art. I, § 10, cl. 1. The Home Rule Act extends the
Contract Clause to the District of Columbia. D.C. Code § 1-
203.02.
The Contract Clause “applies only to laws with
retrospective, not prospective, effect.” Loc. Div. 589,
Amalgamated Transit Union v. Mass., 666 F.2d 618, 637 (1st
Cir. 1981) (Breyer, J.); see Ogden v. Saunders, 25 U.S. (12
Wheat.) 213 (1827). The initial CMPA amendment had only
prospective effect: It became effective in June 2020, and it
applied only to collective bargaining agreements entered into
after the parties’ 2017 Agreement expired on September 30,

-- 10 of 15 --

11
2020. D.C. Act 23-336, § 116(c)(2). And the FOP points to
no successor agreement allegedly impaired by the later
iterations of section 116.
Instead, the union argues that all iterations of section 116
violate the Contract Clause because they impair rights under
the expired 2017 Agreement. The FOP points to article 12 of
that agreement, which provided that the existing disciplinary
procedure “shall be incorporated into any successor”
agreement unless changed through a prescribed process. 2017
Agreement, art. 12, § 2 (J.A. 91). According to the union,
article 12 made it impermissible for the Council to authorize
new rules governing future bargaining over successor
agreements.
We disagree. Retrospective laws violate the Contract
Clause only if they “substantially” impair existing contract
rights. Sveen v. Melin, 138 S. Ct. 1815, 1822 (2018). Whether
impairment is substantial turns in part on the parties’
reasonable expectations. Id. Here, the union could not have
reasonably expected to insulate itself from legal changes after
the 2017 Agreement had expired by its terms. For one thing,
the Contract Clause does not give parties the right to contract
out of generally applicable laws in perpetuity. See Home Bldg.
& Loan Ass’n v. Blaisdell, 290 U.S. 398, 435 (1934);
Amalgamated Transit, 666 F.2d at 638 (“It is difficult to
believe that the parties to the agreement thought they could
bind their successors forever.”). Moreover, the D.C.
government has heavily regulated collective bargaining for
decades, so the union was on notice that future statutory
changes were likely. See Energy Rsrvs. Grp., Inc. v. Kansas
Power & Light Co., 459 U.S. 400, 411 (1983); Veix v. Sixth
Ward Bldg. & Loan Ass’n, 310 U.S. 32, 38 (1940). When
signing the 2017 Agreement, the union could reasonably

-- 11 of 15 --

12
expect that its terms would last for the duration of the
agreement, but not longer.
In addition, we consider whether the law at issue serves a
“significant and legitimate public purpose.” Sveen, 138 S. Ct.
at 1822. In many respects, section 116 is like other state laws
that have survived past Contract Clause challenges. For one
thing, it deals with a “broad, generalized economic or social
problem,” and it operates in an area “already subject to state
regulation.” Allied Structural Steel Co. v. Spannaus, 438 U.S.
234, 249–50 (1978). For another, its changes were neither
“immediate” nor “retroactive.” Id. In sum, section 116’s
prospectivity, its modest effect on the 2017 Agreement, and its
legitimate purposes together doom the challenge here.
D
Finally, the FOP contends that section 116 violates the
Due Process Clause of the Fifth Amendment, which provides
that “[n]o person shall … be deprived of life, liberty, or
property without due process of law.” U.S. Const. amend. V.
The union invokes cases suggesting that laws causing “grave
unfairness” violate substantive due process. Tri Cnty. Indus. v.
District of Columbia, 104 F.3d 455, 459 (D.C. Cir. 1997).
The doctrine of substantive due process is narrow. See
Abigail Alliance for Better Access to Developmental Drugs v.
Von Eschenbach, 495 F.3d 695, 702 (D.C. Cir. 2007) (en banc).
We have found no cases invalidating state action under the
grave-unfairness test advocated by the union. And there are
several cases rejecting substantive due process claims resting
on assertions of grave unfairness.2 What we have said above
2 See Zevallos v. Obama, 793 F.3d 106, 118 (D.C. Cir. 2015);
Elkins v. District of Columbia, 690 F.3d 554, 561–62 (D.C. Cir.
2012); Am. Fed’n of Gov’t Emps., AFL-CIO, Loc. 446 v. Nicholson,

-- 12 of 15 --

13
is enough to show that section 116 is not gravely unfair: it
implicates no fundamental rights, it imposes no punishment,
and it has only modest prospective effect on past contractual
arrangements. In addition, the union makes no argument that
the right to bargain collectively over disciplinary procedures is
“deeply rooted in this Nation’s history and tradition.” Abigail
Alliance, 495 F.3d at 697 (quoting Washington v. Glucksberg,
521 U.S. 702, 720–21 (1997)).
III
After the district court rejected the union’s claims on the
merits, it dismissed the case without prejudice. The FOP
sought to amend its complaint, which would have required the
court first to amend its judgment under Federal Rule of Civil
Procedure 59(e). See Ciralsky v. CIA, 355 F.3d 661, 668 (D.C.
Cir. 2004). The court refused, concluding that the union’s
proposed amendments would not cure the flaws in its case. We
review de novo this determination of futility. See Osborn v.
Visa Inc., 797 F.3d 1057, 1062–63 (D.C. Cir. 2015).
Start with the equal protection claim. The proposed
amended complaint added allegations that police, prison
guards, and protective-services officers all carry firearms and
that prison guards are more likely to use force than police. But
this minor elaboration hardly negates “every conceivable
rationale” for treating police differently. See Beach Commc’ns,
508 U.S. at 315. For example, the Council could have
concluded that police discipline was more pressing because the
police exercise “all the common-law powers of constables”
475 F.3d 341, 353 (D.C. Cir. 2007); George Wash. Univ. v. District
of Columbia, 318 F.3d 203, 209 (D.C. Cir. 2003); Wash. Teachers’
Union Local No. 6 v. Bd. of Educ., 109 F.3d 774, 781 (D.C. Cir.
1997); Silverman v. Barry, 845 F.2d 1072, 1080 (D.C. Cir. 1988).

-- 13 of 15 --

14
against the population at large, D.C. Code § 5-127.04(a), while
prison guards and protective services interact with only a
narrow subset of the population.
For the attainder claim, the FOP offers a handful of
comments from members of the D.C. Council. The two most
noteworthy come from one member who stated that “there are
police in the District who are bad actors and who have been
going on without the proper penance.” J.A. 534. The same
member also expressed a desire, after hearing about an officer
who allegedly testified falsely to secure a conviction, “to have
some kind of retribution or some kind of justice in this criminal
justice system.” Id.
Neither of these statements suggests that section 116 was
punitive. For starters, the statements express no desire to
legislatively determine the guilt of individual officers or groups
of officers. For another, they express nothing about section
116, as opposed to the many other provisions of the Reform
Act. And in any event, “isolated statements” by a single
legislator, even if revealing a punitive intent, cannot turn an
otherwise valid law into an unconstitutional bill of attainder
absent evidence that other legislators shared the desire to
punish. See Kaspersky, 909 F.3d at 464. The union offers no
such evidence here.
The FOP’s attempt to rehabilitate its Contract Clause
claim also would fail. The proposed complaint alleged that the
disciplinary procedures negotiated in article 12 of the 2017
Agreement were important to the union. That may be so, but
the union could not reasonably have expected its agreement to
forever limit the legislature’s power to adjust the scope of
collective bargaining.

-- 14 of 15 --

15
Finally, the proposed complaint would not salvage the
substantive due process claim, for none of its new allegations
materially undercuts the analysis we have set forth above.
IV
The district court correctly concluded that the FOP’s
constitutional claims lack merit.
Affirmed.

-- 15 of 15 --