Lacey Stradford v. Secretary Pennsylvania Department of Corrections

PRECEDENTIAL
UNITED STATES COURT OF APPEALS
FOR THE THIRD CIRCUIT
____________
Nos. 21-2655 & 22-2027
____________
LACEY STRADFORD;
WILLIAM NETTLES;
JESSE STROUD;
WILLIAM SCOTT;
RICHARD RICHARDSON,
on behalf of THEMSELVES AND ALL OTHER
SIMILARLY SITUATED
v.
SECRETARY PENNSYLVANIA DEPARTMENT OF
CORRECTIONS,
Appellant
____________
On Appeal from the United States District Court
for the Eastern District of Pennsylvania
(D.C. No. 2-16-cv-02064)
District Judge: Honorable Juan R. Sanchez
____________
Argued: June 29, 2022

-- 1 of 21 --

2
Before: JORDAN, PORTER, and PHIPPS,
Circuit Judges.
(Filed: November 9, 2022 )
____________
Sean A. Kirkpatrick
Office of Attorney General of Pennsylvania
Strawberry Square, 15th Floor
Harrisburg, PA 17120
Daniel B. Mullen [Argued]
Office of Attorney General of Pennsylvania
1251 Waterfront Place
Mezzanine Level
Pittsburgh, PA 15222
Counsel for Appellant
Donald Driscoll [Argued]
Community Justice Project
100 Fifth Avenue
Suite 900
Pittsburgh, PA 15222
Alexandra Morgan-Kurtz
Pennsylvania Institutional Law Project
247 Fort Pitt Boulevard
4th Floor
Pittsburgh, PA 15222
Counsel for Appellees

-- 2 of 21 --

3
____________
OPINION OF THE COURT
____________
PORTER, Circuit Judge.
Class representatives Lacey Stradford, William Nettles,
Jesse Stroud, William Scott, and Richard Richardson
(“Appellees”), all convicted sex offenders, allege the
Pennsylvania Department of Corrections (“DOC”)1 enforces a
policy that unconstitutionally discriminates against sex
offenders. The policy requires DOC to consider, among other
things, “community sensitivity” when it evaluates parolees for
halfway house placement. The District Court entered summary
judgment for Appellees, holding that sex offenders and non-
sex offenders are similarly situated and consideration of
“community sensitivity” when making halfway house
assignments is irrational.
But not all crimes are alike. The differences among sex
crimes, and between sex crimes and non-sex crimes, preclude
the purported similarity between sex offenders and non-sex
offenders in this case. A discretionary grant of parole cannot
erase those differences. In any event, DOC’s halfway house
policy considering “community sensitivity,” among many
other factors, is rationally related to more than one legitimate
government interest. So we will reverse and remand for entry
of summary judgment for the DOC.
1 At the time of judgment, John Wetzel was the Secretary of
Corrections. George Little took his place. See Fed. R. App. P.
43(c)(2) (providing automatic substitution of officeholders).

-- 3 of 21 --

4
I
After completing a minimum sentence, inmates in
Pennsylvania are eligible to serve the rest of their sentence on
parole. See 61 Pa. Cons. Stat. § 6137(a)(3). Parole is “a matter
of grace and mercy shown to a prisoner who has demonstrated
to the Parole Board’s satisfaction his future ability to function
as a law-abiding member of society upon release before the
expiration of the prisoner’s maximum sentence.” Hudson v.
Pa. Bd. of Prob. & Parole, 204 A.3d 392, 396 (Pa. 2019)
(quoting Rogers v. Pa. Bd. of Prob. & Parole, 724 A.2d 319,
322–23 (Pa. 1999)).
The decision to grant parole is discretionary. 42 Pa.
Cons. Stat. § 2154.5(b). Before deciding, the Parole Board
must investigate, among other things, “[t]he general character
and background of the inmate,” “[t]he nature and
circumstances of the offense committed,” “[t]he written or
personal statement of the testimony of the victim or the
victim’s family,” the inmate’s “behavioral condition and
history,” and his “complete criminal record.” 61 Pa. Cons. Stat.
§ 6135(a).
When reviewing parole applications, the Parole Board
must determine whether “[t]he best interests of the offender
justify or require that the offender be paroled” and whether “the
interests of the Commonwealth will be injured by the
offender’s parole.” Id. § 6137(a)(1)(i)–(ii). In evaluating this
standard, the Parole Board considers its own guidelines, which
are designed to: (1) “[g]ive primary consideration to the
protection of the public and to victim safety,” (2) “[p]rovide
for due consideration of victim input,” (3) encourage proper
conduct of parolees, (4) “encourage inmates and parolees to
participate in programs that have been demonstrated to be

-- 4 of 21 --

5
effective in reducing recidivism,” (5) prioritize “incarceration,
rehabilitation and other criminal justice resources for offenders
posing the greatest risk to public safety,” (6) “[u]se validated
risk assessment tools,” and “take into account available
research relating to the risk of recidivism, minimizing the
threat posed to public safety and factors maximizing the
success of reentry.” 42 Pa. Cons. Stat. § 2154.5(a).
The Parole Board also has authority to delay parole
release until a satisfactory home plan is arranged and approved.
See 37 Pa. Code § 63.1(d) (“The date of parole may be
postponed until a satisfactory plan is arranged for the parolee
and approved by the Board.”); Barge v. Pa. Bd. of Prob. &
Parole, 39 A.3d 530, 548 (Pa. Commw. Ct. 2012). Because
finding housing is often difficult, most inmates first rely on
halfway houses. Those houses have limited capacity. Public
houses have only 700 spaces, and private contract facilities
have 2,100 spaces statewide. But each year, about 9,000
Pennsylvania inmates are released on parole.
Sex offenders face several collateral consequences due
to the nature of their criminal acts. They must participate in a
specialized treatment program to become eligible for parole.
Violent sex offenders must continue that specialized treatment
program even after release from prison. Sex offenders must
register with the Pennsylvania State Police. For violent sex
offenders, the Pennsylvania State Police must notify the victim
of their release. And relevant here, the Pennsylvania State
Police must notify each resident, school district, day-care
center, and college about nearby registered violent sex
offenders. 42 Pa. Cons. Stat. §§ 9718.1, 9799.13, 9799.26,
9799.27, 9799.70.

-- 5 of 21 --

6
According to DOC, that notification requirement makes
it difficult to place sex offenders into community halfway
houses because once neighbors are notified, some oppose sex
offenders’ presence. In at least one instance, community
backlash against high concentrations of sex offenders in
neighboring halfway houses caused a halfway house to close.
For the same reasons it is hard to place sex offenders
into halfway houses, sex offenders once placed tend to linger
in halfway houses longer than other parolees. Potential
landlords can use an applicant’s sex offender status to refuse
leasing to them, and many sex offenders cannot obtain
federally funded housing. See 42 U.S.C. § 13663(a); 24 C.F.R.
§ 5.856. As a result, paroled sex offenders often remain in
halfway houses until their maximum sentences expire, and
because sex offenders receive higher-than-average maximum
sentences, it can take years for their sentences to expire once
they are paroled. By contrast, according to DOC, other
parolees spend ninety days on average in a halfway house. The
cumulative effect of these phenomena is that sex offenders clog
the parole system.
As originally drafted, DOC Policy 8.1.1 Section 4
designated sex offenders as categorically “hard to place” and
rejected them for initial placement into halfway houses.
J.A. 76. When the putative sex offender class challenged that
policy in court, the District Court determined that, because
non-sex offenders have a greater likelihood of successfully
rejoining their communities after temporary placement in a
halfway house, the DOC’s policy served the legitimate interest
in avoiding clogging the system. Stradford v. Wetzel, No. CV
16-2064, 2017 WL 1196656, at *4 (E.D. Pa. Mar. 31, 2017).
The named plaintiffs appealed.

-- 6 of 21 --

7
While on appeal, DOC changed its policy. The new
policy lists thirteen factors DOC must consider before placing
a parolee in a halfway house:
a. community sensitivity to a criminal offense
or specific criminal incident;
b. board action stipulations;
c. program needs vs. program availability in a
particular area;
d. separations from other reentrants or staff;
e. multiple failures at one facility;
f. victim consideration;
g. medical or mental health needs;
h. final discharge of maximum sentence date;
i. gender status of the facility;
j. pilots or studies being conducted;
k. request by the reentrant for relocation;
l. available community resources/support; and
m. where the reentrant’s committing county;
requested release county; and home county are
in relation to an appropriate center.
J.A. 83.
After the policy change, we vacated the District Court’s
judgment and remanded for it to consider whether the lawsuit
was moot. Stradford v. Sec’y Pa. Dep’t of Corr., 783 F. App’x
150, 151 (3d Cir. 2019). Appellees filed an amended complaint
challenging the new policy and the District Court found the suit
not moot. See Stradford v. Wetzel, 519 F. Supp. 3d 214, 223
n.6 (E.D. Pa. 2021). The District Court said its former decision
was in error. Id. at 230. It held that paroled sex offenders are
similarly situated to other paroled offenders, and that there
could be no rational basis to delay their placement into halfway

-- 7 of 21 --

8
houses because of “community sensitivity.” Id. at 224–25,
230–31.2 This appeal followed.
II
Appellees filed this class action under 42 U.S.C. § 1983.
The District Court had subject matter jurisdiction under 28
U.S.C. §§ 1331 and 1343. We have jurisdiction under 28
U.S.C. § 1291.
We review de novo the District Court’s resolution of
cross-motions for summary judgment. Int’l Union, United
Mine Workers of Am. v. Racho Trucking Co., 897 F.2d 1248,
1252 (3d Cir. 1990). Summary judgment is appropriate when,
drawing all reasonable inferences in favor of the nonmoving
party, “the movant shows that there is no genuine dispute as to
any material fact,” and thus the movant “is entitled to judgment
as a matter of law.” Thomas v. Cumberland Cnty., 749 F.3d
217, 222 (3d Cir. 2014) (quoting Fed. R. Civ. P. 56(a)).
2 The DOC says that there is an unresolved dispute of material
fact over the degree of delay in halfway house placement
caused by the community-sensitivity factor. While this is a
factual dispute, it is not material because both parties agree that
the community sensitivity factor disfavors sex offenders. In
defense of the policy, DOC admits that community backlash
makes sex offenders difficult to place. And DOC recognizes
that “lack of community acceptance,” Appellant’s Opening Br.
12, prevents it from “[p]lacing too many sex offenders into
halfway houses at a given moment,” Appellant’s Opening Br.
13. Without a genuine dispute of material fact, we will evaluate
the equal-protection claim’s merit.

-- 8 of 21 --

9
III
The Fourteenth Amendment’s Equal Protection Clause
states that “[n]o State shall . . . deny to any person within its
jurisdiction the equal protection of the laws.” U.S. Const.
amend. XIV, § 1. The Supreme Court has said that “the equal
protection of the laws is a pledge of the protection of equal
laws.” Yick Wo v. Hopkins, 118 U.S. 356, 369 (1886). At
bottom, the Equal Protection Clause requires equal treatment
of “all persons similarly situated.” Ass’n of N.J. Rifle & Pistol
Clubs, Inc. v. Att’y Gen. N.J., 910 F.3d 106, 125 (3d Cir. 2018)
(quoting Shuman ex rel. Shertzer v. Penn Manor Sch. Dist., 422
F.3d 141, 151 (3d Cir. 2005)). “The Equal Protection Clause
does not forbid classifications.” Nordlinger v. Hahn, 505 U.S.
1, 10 (1992). But the distinctions between classes “must be
rationally related to a legitimate governmental purpose.” City
of Cleburne v. Cleburne Living Ctr., 473 U.S. 432, 446 (1985).
To establish an equal-protection claim, a plaintiff “must
show that the Government has treated it differently from a
similarly situated party and that the Government’s explanation
for the differing treatment does not satisfy the relevant level of
scrutiny.” Real Alternatives, Inc. v. Sec’y Dep’t of Health &
Hum. Servs., 867 F.3d 338, 348 (3d Cir. 2017) (emphasis
removed) (citing City of Cleburne, 473 U.S. at 439–40).3
3 The parties agree that the DOC’s policy is subject to rational
basis review because sex offenders do not belong to a suspect
or quasi-suspect class and the DOC’s policy implicates no
fundamental constitutional rights. See Artway v. Att’y Gen. of
N.J., 81 F.3d 1235, 1267 (3d Cir. 1996).

-- 9 of 21 --

10
A
Because “equality” is a rhetorically ambiguous concept,
it’s easy to “invoke any existing descriptive inequality as a
basis for asserting what is essentially a prescriptive grievance.”
Peter Westen, Speaking of Equality: An Analysis of the
Rhetorical Force of “Equality” in Moral and Legal Discourse
279 (1990). But the Fourteenth Amendment proscribes
unequal treatment only among persons similarly situated
according to a relevant standard of comparison. See
Nordlinger, 505 U.S. at 10 (Persons are similarly situated
under the Equal Protection Clause when they are alike “in all
relevant respects.”). So an equal-protection challenge must
allege more than “broad generalities” in identifying a
comparator. Griffin Indus., Inc. v. Irvin, 496 F.3d 1189, 1204
(11th Cir. 2007). Courts must “isolate the factor allegedly
subject to impermissible discrimination.” United States v.
Olvis, 97 F.3d 739, 744 (4th Cir. 1996) (quoting United States
v. Aguilar, 883 F.2d 662, 706 (9th Cir. 1989)); Cung Hnin v.
TOA (USA), LLC, 751 F.3d 499, 504–05 (7th Cir. 2014)
(same); Freeman v. City of Santa Ana, 68 F.3d 1180, 1187 (9th
Cir. 1995) (same). Other factors explaining disparate treatment
will usually preclude persons from being similarly situated. In
turn, the failure to identify similarly situated persons dooms an
equal-protection claim. See Hill v. Borough of Kutztown, 455
F.3d 225, 239 (3d Cir. 2006) (citing Levenstein v. Salafsky, 414
F.3d 767, 776 (7th Cir. 2005)) (stating equal-protection claim
“must fail because [plaintiff] does not allege the existence of
similarly situated individuals”).
Pennsylvania law creates three tiers of sex offenders
based on their offenses and further distinguishes sexually
violent predators. See 42 Pa. Cons. Stat. §§ 9799.14, 9799.24.

-- 10 of 21 --

11
DOC incorporates those distinctions into its decisions on when
and where to place sex offender parolees. Appellees must show
that these are irrational distinctions for those decisions.
When evaluating whether offenders are similarly
situated under the Equal Protection Clause, we must assess the
nature of their respective crimes. See Skinner v. Oklahoma ex
rel. Williamson, 316 U.S. 535, 541 (1942) (determining
whether embezzlement and grand larceny are “intrinsically the
same quality of offense”); Doe v. Settle, 24 F.4th 932, 940 (4th
Cir. 2022) (“When a law imposes collateral consequences
based on criminal convictions, two impacted offenders who are
treated differently can be similarly situated if their convictions
are similar enough.”). So Appellees need to show, for example,
that an offender convicted of rape and an offender convicted of
robbery, wire fraud, or a drug offense “are alike ‘in all relevant
respects.’ ” Harvard v. Cesnalis, 973 F.3d 190, 205 (3d Cir.
2020) (quoting Nordlinger v. Hahn, 505 U.S. 1, 10 (1992)).
Because Appellees assert that all sex offenders are
similarly situated to all other offenders eligible for parole, they
need to do this comparison for each crime of conviction. They
must also show that sex offenders are similarly situated among
themselves. Differences between crimes might reasonably
explain differences in treatment. So Appellees must evaluate
each crime to see if they are, in essence, the same quality of
offense.
Appellees have not made these comparisons. But courts
that have done so conclude, unsurprisingly, that sex crimes and
non-sex crimes—and even different types of sex crimes—are
dissimilar. See, e.g., Petitpas v. Martin, No. 20-3557, 2021 WL
6101469, at *2 (2d Cir. Dec. 22, 2021) (collecting cases);
Carney v. Okla. Dep’t of Pub. Safety, 875 F.3d 1347, 1353–54

-- 11 of 21 --

12
(10th Cir. 2017) (aggravated sex offenders are not similarly
situated to ordinary sex offenders); Litmon v. Harris, 768 F.3d
1237, 1243 (9th Cir. 2014) (mentally disordered offenders and
mentally disordered sex offenders are not similarly situated to
sexually violent predators).
These essential differences between crimes also explain
why laws imposing collateral burdens on sex offenders have
generally been upheld. See, e.g., United States v. Kebodeaux,
570 U.S. 387, 395–96 (2013) (explaining why differences
between sex offenders and non-sex offenders justify post-
release registration rules); Smith v. Doe, 538 U.S. 84, 103–04
(2003) (upholding reporting requirements for sex offenders);
Conn. Dept. of Pub. Safety v. Doe, 538 U.S. 1, 4 (2003)
(upholding publicly available sex offender registry).
“Sex offenders are a serious threat in this Nation.”
McKune v. Lile, 536 U.S. 24, 32 (2002) (plurality opinion).
“[T]he victims of sexual assault are most often juveniles,” id.,
and “[t]he sexual abuse of a child is a most serious crime and
an act repugnant to the moral instincts of a decent people.”
Ashcroft v. Free Speech Coal., 535 U.S. 234, 244 (2002).
Precisely because sex offenses are serious and different than
other types of crimes, Pennsylvania law imposes unique
collateral consequences on sex offenders. Sex offenders must
register with the state police, may not qualify for federally
funded public housing, and must participate in a sex offender
treatment program.4 And the state police must notify each
4 See 42 Pa. Cons. Stat. § 9799.13 (requiring registration with
the state police); 34 U.S.C. § 20920 (requiring states put
offender information on a publicly accessible website); 42
U.S.C. § 13663(a) (prohibiting those who are subject to a
lifetime registration requirement from public housing); 42 Pa.

-- 12 of 21 --

13
resident, school district, day-care center, and college about
nearby registered sex offenders. See 42 Pa. Cons. Stat.
§ 9799.27. Those collateral consequences are linked to
Pennsylvania’s finding that “[s]exual offenders pose a high
risk of committing additional sexual offenses and [that]
protection of the public from this type of offender is a
paramount governmental interest.” 42 Pa. Cons. Stat.
§ 9799.11(a)(4).
Because of these policies, sex offenders have a harder
time finding a job or a home. The notification policy also
means that residents are aware of the presence of sex offenders
(unlike other offenders) in their neighborhood. These are
inescapable facts for policymakers at DOC trying to make
rational halfway house policies.
Appellees argue that a favorable parole action
eliminates the differences between offenders because the
Parole Board considers them all safe to release into the public.
See Appellees’ Br. 17–18. The District Court adopted that
theory. Wetzel, 519 F. Supp. 3d at 224–25. We disagree. The
individualized assessment underlying a favorable parole action
doesn’t expunge one’s legal status as a sex offender, or change
him from a sex offender to a non-sex offender. And the parole
inquiry of “whether an individual poses a substantial danger of
physical harm to others is far broader than the inquiry into
whether one is likely [to] . . . engage in sexually violent
criminal behavior.” Litmon, 768 F.3d at 1243 (internal
quotation marks omitted). So a favorable parole action does not
Cons. Stat. § 9718.1 (requiring participation in treatment
program); 42 Pa. Cons. Stat. § 9799.70 (mandating a continued
treatment program for violent sex offenders).

-- 13 of 21 --

14
alleviate the differences between sex crimes and non-sex
crimes.
A parole board’s individualized assessment is merely a
reasoned “prediction[] of future behavior” about a particular
individual, so by itself it cannot make two people, let alone
entire groups of offenders, similarly situated. Conn. Bd. of
Pardons v. Dumschat, 452 U.S. 458, 464 (1981). As in other
states, the decision to parole in Pennsylvania is discretionary.
See 42 Pa. Cons. Stat. § 2154.5(b). The Parole Board uses a
general and flexible standard when deciding to grant parole.
See 61 Pa. Cons. Stat. § 6137(a)(1)(i)–(ii). When applying this
standard, the Parole Board considers numerous subjective
criteria. See 42 Pa. Cons. Stat. § 2154.5(a). And to aid in its
decision, it must investigate, among other things, the various
attributes of the inmate and the facts of his crime. See 61 Pa.
Cons. Stat. § 6135(a).
Despite its best efforts, the Parole Board can’t predict
any offender’s future conduct. The Parole Board’s
discretionary, predictive, and fallible determination is based on
individualized evaluation and imperfect knowledge. That is
legally relevant because state action that involves
“discretionary decisionmaking based on a vast array of
subjective, individualized assessments” necessarily results in
different treatment among those subject to the discretionary
action. Engquist v. Or. Dep’t of Agric., 553 U.S. 591, 603–04
(2008) (different but discretionary treatment of apparently
similarly situated employees does not raise equal protection
concerns).
The Equal Protection Clause allows state officials to
exercise their discretion to grant parole to “one class of
criminals and deny it to others,” to determine “the kind or

-- 14 of 21 --

15
amount of evidence upon which to base its determination,” or
attach whatever “conditions to the application for or to the
granting of [parole] as [the state] may deem proper.” Ughbanks
v. Armstrong, 208 U.S. 481, 488 (1908).5 The government “is
not bound to grant a parole in any case” so long as it treats
similarly situated persons equally. Id. at 487.
The District Court held that considering the collateral
consequences of sex offenses was irrelevant because they “are
part of the very classification on the basis of which Plaintiffs
argue they are suffering discrimination.” Wetzel, 519 F. Supp.
3d at 225. That is, the District Court agreed with Appellees that
DOC couched an equal protection violation within another,
antecedent equal protection violation. But these collateral
burdens have themselves survived Equal Protection Clause
scrutiny. See, e.g., Artway, 81 F.3d at 1267–68. And in any
event, Appellees’ counsel acknowledged in oral argument that
state and federal laws imposing collateral burdens on sex
offenders are reasonable attempts to protect the public and
disclaimed the suggestion that they, too, are unconstitutional.
Appellees are not similarly situated with non-sex
offender parolees, so the first prong of their equal protection
claim fails. Hill, 455 F.3d at 239. But even if Appellees could
show that they are similarly situated to non-sex offender
parolees, that would not save their claim. Because Appellees
5 Of course, Pennsylvania may not classify offenders based on
factors that are “foreign to the parole statute,” such as race,
religion, or political beliefs. Newman v. Beard, 617 F.3d 775,
784 (3d Cir. 2010). But none of these impermissible factors are
at issue here.

-- 15 of 21 --

16
don’t belong to a suspect class, they would have to show that
DOC’s halfway house policy is irrational.
B
Under the rational basis test, a law does not “run afoul
of the Equal Protection Clause if there is a rational relationship
between the disparity of treatment and some legitimate
governmental purpose.” Armour v. City of Indianapolis, 566
U.S. 673, 680 (2012) (quoting Heller v. Doe, 509 U.S. 312,
319–20 (1993)). The challenged “legislation enjoys a
presumption of validity, and [a] plaintiff must negate every
conceivable justification for the classification in order to prove
that the classification is wholly irrational.” Brian B. ex rel. Lois
B. v. Pa. Dep’t of Educ., 230 F.3d 582, 586 (3d Cir. 2000). “[I]f
there is any reasonably conceivable state of facts that could
provide a rational basis for the classification,” the policy
survives. FCC v. Beach Commc’ns, Inc., 508 U.S. 307, 313
(1993). And if DOC’s conceivable rationale seems tenuously
related to its governmental interest, the sex offenders still must
show that its criteria and actions are “wholly irrational.”
Cabrera v. Att’y Gen. U.S., 921 F.3d 401, 404 (3d Cir. 2019)
(citation omitted).
DOC’s halfway house policy is supported by rational
bases. Pennsylvania has a legitimate interest in considering
public safety and public concern over the concentration of
sexual offenders in a given area. Specifically, DOC has a
legitimate interest in considering a community’s rational
concern about sex offenders’ troubling crimes and risk of
recidivism.
Communities rationally fear that sex offenders pose a
serious “danger to the community” because they typically have

-- 16 of 21 --

17
a “higher risk of recidivism.” United States v. Juv. Male, 670
F.3d 999, 1010 (9th Cir. 2012) (quoting United States v.
Salerno, 481 U.S. 739, 747 (1987)); Litmon, 768 F.3d at 1244;
see also Doe v. Moore, 410 F.3d 1337, 1347 (11th Cir. 2005)
(“The increased reporting requirements based on evidence of
increased recidivism among [sex offenders] . . . [are] rationally
related to the state’s interest in protecting its citizens from
criminal activity.”); Cutshall v. Sundquist, 193 F.3d 466, 483
(6th Cir. 1999) (“Given the indications that sex offenders pose
a particular threat of reoffending, we cannot say that the Act is
irrational [under the Equal Protection Clause].”).
These fears are not based on mere animus or unfounded
prejudice. “Half of prisoners released after serving time for
rape or sexual assault had an arrest within 9 years that led to a
conviction.” U.S. Department of Justice, Bureau of Justice
Statistics, Recidivism of Sex Offenders Released from State
Prison: A 9-Year Follow-Up (2005-2014) 6 (2019)
(https://bjs.ojp.gov/library/publications/recidivism-sex-
offenders-released-state-prison-9-year-follow-2005-14, last
visited September 14, 2022). And the Justice Department
recently confirmed its prior findings that sex offenders released
from state prison are much more likely than other released
prisoners to be arrested for rape or sexual assault. Id. at 5. See
McKune, 536 U.S. at 32 (citing 1997 and 2000 studies).
But even if sex offenders recidivate at equal or lower
rates than other criminals, a community can rationally fear sex
offenders more than other criminals because sex offenders
target “vulnerable individuals.” Artway, 81 F.3d at 1267
(“Protecting vulnerable individuals from sexual offenses is
certainly a legitimate state interest.”); see also Doe v. Cuomo,
755 F.3d 105, 115 (2d Cir. 2014) (requiring a sex offender

-- 17 of 21 --

18
considered non-dangerous to remain on a sex offender registry
rationally relates to protecting the public).
Appellees contend that any consideration of community
sensitivity impermissibly opens the door to irrational prejudice
held unconstitutional by City of Cleburne, 473 U.S. 432. That
argument conflates “mere negative attitudes, or fear, [toward
the mentally retarded] unsubstantiated by factors which are
properly cognizable in a zoning proceeding” with one of
thirteen factors cumulatively considered by the Parole Board
before designating prisoners for discretionary placement in a
halfway house. Id. at 448. In City of Cleburne, the city council
denied a special use permit for a group home for the
intellectually disabled. To support its decision, the council
pointed to “the negative attitude of the majority of property
owners located within 200 feet” of the proposed location. Id.
But those concerns are not “properly cognizable in a zoning
proceeding.” Id. The council needed a different reason to
justify denying the permit.
The Court in Cleburne emphasized that only “irrational
prejudice” is unlawful under the Fourteenth Amendment. Id. at
450 (emphasis added); see also U.S. Dep’t of Agric. v. Moreno,
413 U.S. 528, 534–35 (1973) (holding that nothing more than
“a bare congressional desire to harm a politically unpopular
group” violates the Equal Protection Clause); Romer v. Evans,
517 U.S. 620, 632 (1996) (holding unlawful a state law that
precluded local ordinances from protecting homosexuals from
discrimination because it “seems inexplicable by anything but
animus toward the class it affects”). In other words, disfavor
based on nothing but mere animus toward a group violates the
Equal Protection Clause. Here, DOC relied on thirteen factors
to evaluate a parolee’s fitness for placement in a halfway

-- 18 of 21 --

19
house, and those factors are related to the success of the
halfway house system.
Appellees argue that by virtue of the Parole Board’s
favorable decision, sex offenders granted parole do not have
the same likelihood of recidivism as sex offenders denied
parole. This argument relies on two faulty premises. In its
strongest form, the argument presumes that a favorable parole
decision somehow eliminates or reduces the offender’s actual
likelihood of recidivism. In its weaker form, the argument
presumes that a decision to grant parole is a definitive finding
of offenders’ low risk to the public. But any parole decision is
an exercise of discretion considering, among many other
things, the Parole Board’s best assessment of an offender’s risk
of recidivism. The Parole Board doesn’t purport to predict the
future, and it certainly can’t determine the offender’s actual
likelihood of recidivism. The agency’s discretionary,
predictive decision doesn’t render irrational community
concerns about sex offender recidivism or the State’s
legitimate interest in protecting vulnerable people.
For administrative and efficiency reasons, DOC also
has a rational interest in considering community concern over
high concentrations of sex offenders. See Califano v. Jobst,
434 U.S. 47, 53 (1977) (administrative efficiency of Social
Security program is a legitimate government interest satisfying
rational basis test). According to DOC, space in halfway
houses is limited. There are more parolees than rooms
available in halfway houses. Thus, halfway houses cannot
accommodate every parolee, especially not for long periods.
The nettlesome issue with sex offenders, according to DOC, is
that they tend to stay longer in halfway houses and end up
clogging the halfway house system. DOC says this is

-- 19 of 21 --

20
attributable to the collateral burdens that sex offenders
uniquely face.
Because of those challenges, the population of sex
offenders in halfway houses tends to increase over time. Such
concentration, combined with sex offenders’ propensity to
recidivate, arguably creates an increased risk of sex crimes in
an area. Because of similar concerns, Pennsylvania law limits
the number of sexually violent predators in group homes,
including halfway houses. See 42 Pa. Cons. Stat.
§ 9799.55(d)(1). The community is rationally sensitive to sex
offender concentration, and DOC’s policy of considering
community sensitivity for halfway house placement rationally
relates to its interest in maximizing halfway house availability
for all offenders.
Appellees argue that most, but not all, offenders were
successfully placed in homes after their stays in halfway
houses, that most sex offenders left halfway houses before their
sentences expired, that at least one halfway house’s occupants
consist of forty percent sex offenders without controversy, and
that DOC officials did not produce evidence showing that sex
offenders commit sex offenses during their stays at halfway
houses. But this evidence doesn’t negate every conceivable
justification offered by DOC or show that its halfway house
policy is utterly irrational. “[T]he Constitution does not require
the [DOC] to draw the perfect line nor even to draw a line
superior to some other line it might have drawn. It requires only
that the line actually drawn be a rational line.” Armour, 566
U.S. at 685. Nor does rational basis review require specific
facts to justify the government’s legitimate purpose; all it asks
is whether a policy is rational based on “any reasonably
conceivable state of facts.” Beach Commc’ns, 508 U.S. at 313.

-- 20 of 21 --

21
Even “rational speculation unsupported by evidence or
empirical data” is enough. Id. at 315. DOC’s halfway house
policy satisfies that low bar.
Finally, we note that the entire criminal law system
reflects the community’s moral judgments. See United States
v. Bass, 404 U.S. 336, 348 (1971) (“[B]ecause of the
seriousness of criminal penalties, and because criminal
punishment usually represents the moral condemnation of the
community, legislatures and not courts should define criminal
activity.”). Criminal and penal laws graduate punishment to
account for the severity of the crime and the defendant’s moral
culpability. See, e.g., Wimberly v. Williams, 14 F.4th 1140,
1148–49 (10th Cir. 2021) (state had rational basis to treat sex
offenders differently because they “have been convicted of
crimes considered particularly heinous”). There is no reason to
depart from this principle for parole proceedings. The public’s
moral judgments about sex offenses are no less legitimate in
post-conviction matters, particularly where offenders are still
serving their term of punishment. See Commonwealth v.
Williams, 692 A.2d 1031, 1035 (Pa. 1997) (citing Griffin v.
Wisconsin, 483 U.S. 868, 873–75 (1987)) (“parole is a form of
criminal punishment imposed after a guilty verdict”).
* * *
Because the District Court erred in granting summary
judgment for Appellees, we will reverse and remand for entry
of summary judgment for the Department of Corrections. In
light of our disposition, the appeal of the District Court’s April
29, 2022 order will be dismissed as moot.

-- 21 of 21 --