Real Alternatives, Inc. v. Secretary Department of Health & Human Services

PRECEDENTIAL

UNITED STATES COURT OF APPEALS
FOR THE THIRD CIRCUIT

_____________

No. 16-1275
_____________

REAL ALTERNATIVES, INC.; KEVIN I. BAGATTA,
ESQ.; THOMAS A. LANG, ESQ.; CLIFFORD W.
MCKEOWN,
Appellants

v.

SECRETARY DEPARTMENT OF HEALTH AND
HUMAN SERVICES; SECRETARY UNITED STATES
DEPARTMENT OF LABOR; SECRETARY UNITED
STATES DEPARTMENT OF THE TREASURY; UNITED
STATES DEPARTMENT OF HEALTH AND HUMAN
SERVICES; UNITED STATES DEPARTMENT OF
LABOR; UNITED STATES DEPARTMENT OF
TREASURY

_______________________________
 On Appeal from the United States District Court
for the Middle District of Pennsylvania
(District Court No.: 1:15-cv-00105)
District Judge: Honorable John E. Jones, III
_______________________________

Argued November 3, 2016

Before: JORDAN, GREENAWAY, JR., and RENDELL,
Circuit Judges

(Opinion Filed: August 4, 2017)

Matthew S. Bowman (Argued)
David A. Cortman
Alliance Defending Freedom
440 First Street, NW
Suite 600
Washington, DC 20001

Counsel for Appellants Real Alternatives, Inc.;
Kevin I. Bagatta, Esq.; Thomas A. Lang, Esq.;
Clifford W. McKeown

Kevin H. Theriot
Elissa M. Graves

2
Alliance Defending Freedom
15100 N. 90th Street
Scottsdale, AZ 85260

Randy Wenger
Independence Law Center
23 North Front Street
Harrisburg, PA 17101

Counsel for Appellants Real Alternatives, Inc.;
Kevin I. Bagatta, Esq.; Thomas A. Lang, Esq.;
Clifford W. McKeown

Benjamin C. Mizer
Peter J. Smith
Mark B. Stern
Alisa B. Klein
Patrick G. Nemeroff
Megan Barbero
Joshua M. Salzman (Argued)
United States Department of Justice
Civil Division
950 Pennsylvania Avenue, NW
Washington, DC 20530

3
 Counsel for Appellees Secretary Department of
Health and Human Services; Secretary United
States Department of Labor; Secretary United
States Department of the Treasury; United
States Department of Health and Human
Services; United States Department of Labor;
United States Department of Treasury

Richard B. Katskee
Natacha Y. Lam
Americans United for Separation of Church and State
1901 L Street, NW
Suite 400
Washington, DC 20036

Seth M. Marnin
David L. Barkey
Anti-Defamation League
605 Third Avenue
New York, NY 10158

Counsel for Amici Curiae Americans
United for Separation of Church and
State; Anti-Defamation League; Central
Conference of American Rabbis;

4
 Hadassah, The Women’s Zionist
Organization of America, Inc.; National
Council of Jewish Women; People for
the American Way Foundation; Union
for Reform Judaism; Women of Reform
Judaism

OPINION

RENDELL, Circuit Judge:

One of the many provisions of the Patient Protection
and Affordable Care Act, Pub. L. No. 111-148, 124 Stat. 119
(2010), requires employer-provided health insurance plans to
cover an array of preventative services, including FDA-
approved contraceptives, at no cost to participating
employees. Employees have the option of seeking out
covered medical providers and using their services, in which
case they are reimbursed, or they can choose not to use them.
The particular provision that includes contraceptive coverage
is commonly referred to as the “Contraceptive Mandate,” and

5
it includes a limited exemption for houses of worship and
their integrated auxiliaries. See 45 C.F.R. § 147.131(a); 77
Fed. Reg. 8,725, 8,726 (Feb. 15, 2012). A wider set of
religious non-profit and for-profit employers may receive an
accommodation whereby they opt out of providing
contraceptive coverage, with the Government then arranging
for their employees to receive the coverage through third
parties at no cost to, and with no participation of, the
objecting employers. See 45 C.F.R. § 147.131(b)–(c); 78
Fed. Reg. 39,870, 39,874–39,875 (July 2, 2013); Zubik v.
Burwell, 136 S. Ct. 1557, 1559 (2016).

Two years after we upheld this opt-out accommodation
in Geneva College v. Secretary United States Department of
Health and Human Services, 778 F.3d 422, 427 (3d Cir.
2015), vacated and remanded sub nom. Zubik, 136 S. Ct. at
1561, we now confront the house-of-worship exemption.
This appeal presents two primary questions that again derive
from the purported intersection of the Contraceptive Mandate
and religion: (1) whether the Contraceptive Mandate must
exempt a secular anti-abortion group with no religious
affiliation, and (2) whether an employee’s religious beliefs
are substantially burdened by the law’s requirement that his
or her employer’s insurance plan cover contraceptives. After
careful review, but without any hesitation, we answer both
questions in the negative.

Appellant Real Alternatives urges that, pursuant to the
Equal Protection Clause of the Fifth Amendment, if a
religious organization may be exempted from the
Contraceptive Mandate, then non-religious entities with an
identical stance on contraceptives must be exempted as well.
Real Alternatives additionally challenges the Contraceptive

6
Mandate and the criteria for the exemption as not only
arbitrary and capricious under the Administrative Procedures
Act but also contrary to federal law.

The other appellants, three employees of Real
Alternatives, bring individual challenges to the Contraceptive
Mandate. They argue that the Contraceptive Mandate
violates the Church Amendment, 42 U.S.C. § 300a–7(d).
They also argue that maintaining a health insurance plan that
covers contraceptives through their employer violates their
religious rights under the Religious Freedom Restoration Act,
42 U.S.C. §§ 2000bb to 2000bb-4 (“RFRA”).

The District Court denied Appellants’ motion for
summary judgment in its entirety and granted the
Government’s cross-motion for summary judgment in its
entirety. Because we agree with the District Court’s rulings
on all of the issues raised, we will affirm.

I. BACKGROUND

A. Statutory and Regulatory Framework

1. The Affordable Care Act and the Contraceptive
Mandate

In 2010, Congress passed the Patient Protection and
Affordable Care Act, Pub. L. No. 111-148, 124 Stat. 119
(2010), and the Health Care and Education Reconciliation Act
of 2010, Pub. L. No. 111-152, 124 Stat. 1029 (2010)
(collectively, the “ACA”). The ACA requires non-
grandfathered group health plans and insurance providers to
cover four categories of preventative health services, without

7
cost-sharing, as provided for in guidelines supported by the
Health Resources and Services Administration (“HRSA”), an
arm of the Department of Health and Human Services
(“HHS”). 1 One of these four categories is “preventative care
and screenings” for women.

HHS requested assistance from the Institute of
Medicine (“IOM”), a non-profit division of the National
Academy of Sciences, to develop guidelines on the specific
preventative services for women to be covered under the
ACA (none existed at the time the ACA was passed). The
IOM recommended that HRSA endorse a list of services that
included “[FDA]-approved contraceptive methods,
sterilization procedures, and patient education and counseling
for all women with reproductive capacity.” Institute of
Medicine, Clinical Preventative Services for Women:
Closing the Gaps 10 (2011). Examples of FDA-approved
contraceptive methods are diaphragms, oral contraceptives,
intrauterine devices, and emergency contraceptives. Id. at
105–06. HRSA adopted the IOM’s guidelines in full. Health
Resources & Service Administration, Women’s Preventative
Service Guidelines, available at
https://www.hrsa.gov/womensguidelines/ (last visited Jan. 27,
2017). In doing so, HRSA required every group health plan
and health insurance plan to include coverage for these
preventative care services to employees working at non-

1
Grandfathered health plan coverage is that which has
existed continually prior to March 23, 2010 and has not
undergone any of several specified changes since that time.
29 C.F.R. § 2590.715-1251 (2010).

8
exempt employers (the “Contraceptive Mandate”). It did not
require anything from the employee.

2. Exemption to the Contraceptive Mandate

At the same time as HRSA adopted IOM’s
recommended guidelines, an exemption from the
Contraceptive Mandate for certain religious employers was
proposed as an interim final regulation (the “Exemption”). 76
Fed. Reg. 46,621 (Aug. 3, 2011). Commenters to the
proposed guidelines had suggested that requiring religious
employers to sponsor group health plans that provide
contraceptive services could impinge on those employers’
religious freedom. Id. at 46,623. In light of those comments,
HHS and the Departments of Labor and Treasury
(collectively, the “Departments”), the agencies named in Real
Alternatives’s underlying lawsuit, authorized HRSA to
exempt certain religious employers from the Contraceptive
Mandate. The Departments specified that they sought “to
provide for a religious accommodation that respects the
unique relationship between a house of worship and its
employees in ministerial positions” and that “[s]uch an
accommodation would be consistent with the policies of
States that require contraceptive services coverage, the
majority of which simultaneously provide for a religious
accommodation.” 2 Id.
2
Though the language here refers to religious
accommodation, these statements refer to what would
ultimately become the exemption given to religious employers
under the ACA. The Departments established a separate
accommodation for certain employers, addressed supra, that
is not at issue in this litigation.

9
 The Departments originally defined a religious
employer as an employer that:

(1) has as its purpose the inculcation of
religious values;
(2) primarily employs persons who share its
religious tenets;
(3) primarily serves persons who share its
religious tenets; and
(4) is a non-profit organization under Section
6033(a)(1) and Section 6033(a)(3)(A)(i) or (iii)
of the [Internal Revenue] Code. 3

Id. The Departments also noted that HRSA’s “discretion to
establish an exemption applies only to group health plans
sponsored by certain religious employers and group health
insurance offered in connection with such plans,” and thus
“health insurance issuers in the individual health insurance
market would not be covered under any such exemption.” Id.
at 46,623–24 (emphasis added). The Departments formally
adopted the four-part definition for exempted employers in
2012. They also created a one-year safe harbor for non-
exempted, non-profit organizations with religious objections,
and announced that they would develop and propose changes
to the regulation that “would meet two goals—providing

3
Section 6033 of the Internal Revenue Code refers in
relevant part to “churches, their integrated auxiliaries, and
conventions or associations of churches,” and “the
exclusively religious activities of any religious order.” 26
U.S.C. § 6033(a)(3)(A)(i).

10
contraceptive coverage without cost-sharing to individuals
who want it and accommodating non-exempted, non-profit
organizations’ religious objections to covering contraceptive
services . . . .” 77 Fed. Reg. at 8,727.

The final rules regarding the Exemption went into
effect in 2012. The Departments replaced the multifactor
religious employer test with one definition, essentially the
fourth prong of the previous test: “[A]n employer that is
organized and operates as a nonprofit entity and is referred to
in section 6033(a)(3)(A)(i) or (iii) of the [Internal Revenue]
Code,” which addresses churches and their integrated
auxiliaries. 78 Fed. Reg. at 39,874. The Departments noted
that this new definition “continues to respect the religious
interests of houses of worship and their integrated auxiliaries
in a way that does not undermine the governmental interests
furthered by the contraceptive coverage requirement.” Id.
The Departments also stated:

Houses of worship and their integrated
auxiliaries that object to contraceptive coverage
on religious grounds are more likely than other
employers to employ people of the same faith
who share the same objection, and who would
therefore be less likely than other people to use
contraceptive services even if such services
were covered under their plan.

Id. The Departments added that their statement about a
religious employer’s likelihood to hire employees who share
religious beliefs opposing contraceptives was made in
response to commenters concerned that the Exemption would
“undermine the [G]overnment’s compelling interests in

11
promoting public health and ensuring that women have equal
access to health care . . . .” 80 Fed. Reg. 41,318, 41,325 (July
14, 2015); see also 78 Fed. Reg. at 39,887 (“Nor do the
exemption for religious employers and the accommodations
for eligible organizations undermine the [G]overnment’s
compelling interests.”).

In 2015, the Departments stated that the Exemption
was “provided against the backdrop of the longstanding
governmental recognition of a particular sphere of autonomy
for houses of worship, such as the special treatment given to
those organizations in the [Internal Revenue] Code.” 80 Fed.
Reg. at 41,325. They continued:

This exemption . . . is consistent with their
special status under longstanding tradition in
our society and under federal law, and is not a
mere product of the likelihood that these
institutions hire coreligionists. Hiring
coreligionists is not itself a determinative factor
as to whether an organization should be
accommodated or exempted from the
contraceptive requirements.

Id.

B. Factual Background and Procedural History

1. Appellant Real Alternatives

Appellant Real Alternatives is a non-profit, non-
religious, anti-abortion organization. It does not hold itself
out as a religious entity, is not incorporated as such, and has

12
not adopted any religious views or positions. Its primary
purpose is to provide “life-affirming alternatives to abortion
services,” and it offers pregnancy and parenting support
programs as well as abstinence education services to women
and families throughout Pennsylvania, Michigan, and Indiana.
J.A. 92–93.

Real Alternatives avers that its views on human life are
based on science, reason, and non-religious philosophical
principles. Id. at 93. In addition to opposing abortion, Real
Alternatives opposes the use of all contraceptives because it
considers these drugs to be “morally wrong.” Id. at 94.

Real Alternatives administers its programs through
networks of social service agencies, which Real Alternatives
hires as subcontractors. It requires all of its subcontracting
organizations to share its views and to agree not to provide or
recommend contraceptives or abortion. It only hires
employees who share the company’s stance on contraceptives
and abortion.

Since 2008, Real Alternatives has excluded
contraceptive care from the health insurance plan it offers to
its employees. Real Alternatives alleges that in 2014, because
of the ACA, its insurer stopped omitting contraceptive care
from coverage and, as a result, a new plan was offered to
employees. 4 According to Real Alternatives, were it not for
the ACA, its insurance provider would be willing to revert to
providing a plan that omits contraceptive coverage. Real

4
Because the original insurance plan was terminated, it
does not qualify for grandfathered status.

13
Alternatives avers that the Contraceptive Mandate violates the
Equal Protection Clause and the Administrative Procedure
Act (“APA”).

2. Appellants Real Alternatives Employees

Appellants Kevin I. Bagatta, Thomas A. Lang, and
Clifford W. McKeown work for Real Alternatives (the “Real
Alternatives Employees”). They are, respectively, the
President, Vice President of Operations, and Vice President
of Administration of Real Alternatives. They are the only
full-time employees of Real Alternatives, and they aver that
they share the company’s beliefs concerning contraceptive
drugs. Each employee receives health insurance coverage
through Real Alternatives, as do their wives and total of seven
minor children, three of whom are female.
The Real Alternatives Employees aver that the
Contraceptive Mandate violates the Church Amendment.
They also aver that the Contraceptive Mandate violates their
religious rights under RFRA. Specifically, they allege that
their “sincerely held religious beliefs prohibit them from
using, supporting, or otherwise advocating the use of
abortifacients, or participating in a health insurance plan that
covers such items for themselves or their families.” J.A. 123.

3. District Court Opinion

The District Court denied Real Alternatives’s motion
for summary judgment in its entirety and granted the
Government’s cross-motion for summary judgment in its

14
entirety. 5 We find the District Court’s analysis informative
and persuasive for the most part, and we review it here.

The District Court began by addressing Real
Alternatives’s equal protection claim, finding in the first
instance that Real Alternatives is not similarly situated to
religious employers with comparable objections to the
Contraceptive Mandate because, notwithstanding those
objections, they do not share the same bases for those
positions—namely, religion versus a single secular position.
As discussed infra, the District Court raised and distinguished
two relevant federal cases, Center for Inquiry, Inc., v. Marion
Circuit Court Clerk, 758 F.3d 869 (7th Cir. 2014), and March
for Life v. Burwell, 128 F. Supp. 3d 116 (D.D.C. 2015). The
District Court also focused on the “vast history of legislative
protections [that] exist[] to safeguard religious freedom,” and
contrasted “[m]oral philosophies,” which it found “have been
historically unable to enjoy the same privileged state.” J.A.
35. The District Court continued that even if Real
Alternatives were similarly situated to a house of worship,
respecting religious autonomy plainly constitutes a legitimate
purpose to allow the classification to stand under rational
basis review. The District Court examined the Government’s
statements in the ACA regulations and found that it had
sufficiently identified religious freedom as the purpose
furthered by the Exemption. The District Court concluded its
equal protection analysis by expressing concern that

5
The District Court had jurisdiction pursuant to 28 U.S.C.
§§ 1331, 1361, 2201, & 2202; 42 U.S.C. § 2000bb-1; and 5
U.S.C. § 702. This Court has appellate jurisdiction pursuant
to 28 U.S.C. § 1291.

15
“[a]llowing adherence to a single moral belief . . . to be
indistinguishable from religion or an entire moral creed . . .
leads us down a slippery slope.” Id. at 42–43. It reasoned
that “finding a singular moral objection to law on par with a
religious objection” could very well lead to a flood of similar
objections. Id. at 44.

Next, the District Court concluded that Real
Alternatives’s claim that the Contraceptive Mandate is
arbitrary and capricious “fail[s] for the same reasons that [its]
Fifth Amendment equal protection claim lacked merit.” Id. at
49; see also id. at 48 (noting that “[t]he standard for
determining whether an [APA] violation exists under the
arbitrary and capricious standard is markedly similar to
rational basis review”). The District Court also rejected Real
Alternatives’s argument that the Contraceptive Mandate
violates federal law—namely, the ACA and the Weldon
Amendment of the Consolidated Security, Disaster
Assistance, and Continuing Appropriations Act of 2009 (the
“Weldon Amendment”)—as well as the Real Alternatives
Employees’ claim that it violates the Church Amendment.

Finally, the District Court rejected the Real
Alternatives Employees’ RFRA claim. It found that the
burden at issue—maintaining an insurance plan that includes
coverage for preventative services—was not substantial
enough based on the Supreme Court’s approach in other
RFRA cases. See J.A. 62 (first citing Bowen v. Roy, 476 U.S.
693, 703 (1986) (holding that the Government could
condition public benefits on the religiously prohibited act of
providing a social security number without trampling on the
beneficiary’s free exercise rights); then citing Lyng v. Nw.
Indian Cemetery Protective Ass’n, 485 U.S. 439, 449 (1988)

16
(finding that building a road through sacred land did not
violate the free exercise rights of those who believed in the
land’s religious significance)). The District Court concluded
in the alternative that, even if the Contraceptive Mandate did
impose a substantial burden, it would still satisfy RFRA
because it was the least restrictive means of furthering the
Government’s compelling interest in a broadly applicable
system of health care that advances public health and gender
equality.

II. DISCUSSION

A. Standard of Review

We exercise plenary review over a district court’s
grant of summary judgment, applying the same standard that
the district court should have applied. 6 Abramson v. William
Paterson Coll. of N.J., 260 F.3d 265, 276 (3d Cir. 2001). A
court grants summary judgment if “there is no genuine
dispute as to any material fact and the movant is entitled to
judgment as a matter of law.” Fed. R. Civ. P. 56(a). There
are no material facts in dispute; the questions raised by the
parties are matters of law, which we review de novo. Shuman
ex. rel. Shertzer v. Penn Manor Sch. Dist., 422 F.3d 141, 146
(3d Cir. 2005).

6
When the parties were before the District Court, Real
Alternatives moved for summary judgment and the
Departments moved to dismiss or, in the alternative, for
summary judgment. The District Court treated the
Departments’ motion as one for summary judgment, and we
will review accordingly.

17
B. Equal Protection Claim 7

Real Alternatives challenges the constitutionality of
the Exemption’s scope, arguing that it violates the
organization’s right to equal protection under the Fifth
Amendment by exempting only religious employers and not
other secular entities, such as itself, that oppose the
requirements set forth in the Contraceptive Mandate. Real
Alternatives urges that “[t]here is no rational purpose to
impose the Mandate on those who do not want the items and
will not use them,” and contends that it is excluded from the
Exemption “simply because [it] is a ‘non-religious ethical
group[]’ instead of a church.” Appellants’ Br. at 28 (final
alteration in original). If churches receive a religious
exemption, the argument goes, then so too must non-religious
entities.

1. Legal Standard

To prevail on its equal protection claim, Real
Alternatives 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. City of Cleburne v.
Cleburne Living Ctr., 473 U.S. 432, 439–40 (1985). The

7
The equal protection and APA claims are brought both
by Real Alternatives the entity and by Bagatta, Lang, and
McKeown, the organization’s three full-time employees. For
the sake of concision, we will refer to this group of appellants
as “Real Alternatives.”

18
parties agree, as they must, that rational basis review is the
applicable standard. Thus, there must be “a rational
relationship between the disparity of treatment and some
legitimate governmental purpose.” United States v. Pollard,
326 F.3d 397, 407 (3d Cir. 2003) (quoting Heller v. Doe, 509
U.S. 312, 320 (1993)). Rational basis review confers a
“presumption of validity” on legislation, and “the 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) (citing FCC v. Beach
Commc’ns, 508 U.S. 307, 314–15 (1993)).

2. Analysis

We must first determine whether Real Alternatives is
similarly situated to a religious employer, such that the
Exemption must be available to the group absent a legitimate
rationale. There is no question it is not.

Real Alternatives leans on its company-mandated
eschewal of contraceptives in an attempt to situate itself in
lockstep with religious employers who can avail themselves
of the Exemption, contending that it is in fact “more
favorably” or “identically” situated to houses of worship
because all of its employees by definition oppose
contraceptive coverage. Appellants’ Br. at 28, 30. In making
this claim, Real Alternatives invokes Center for Inquiry, in
which the Seventh Circuit struck down an Indiana statute that
permitted religious officials to solemnize marriages but
prohibited their counterparts from secular groups from doing
the same. 758 F.3d at 875. There, the court reasoned that
“[a]n accommodation cannot treat religions favorably when

19
secular groups are identical with respect to the attribute
selected for that accommodation.” Id. at 872.

But Real Alternatives ignores key distinctions between
that case and this one. Most notably, Real Alternatives
disregards the stark contrast between itself and the appellant
in Center for Inquiry, a humanist group that resembles a
“religion[] in everything except belief in a deity.” Id. at 871.
Real Alternatives is a completely different type of entity,
particularly because of its structure, aim, purpose, and
function in its members’ lives. Indeed, Real Alternatives’s
credo is limited to a one-sentence mission statement that says
it “exists to provide life-affirming alternatives to abortion
services throughout the nation.” J.A. 92. In Center for
Inquiry, the humanist organization explicitly argued that “its
methods and values play the same role in its members’ lives
as religious methods and values play in the lives of
adherents.” 758 F.3d at 871 (emphasis added). Real
Alternatives makes no such claim, as it is solely concerned
with administering programs that reflect its moral opposition
to contraceptives and abortion. Thus, Center for Inquiry does
not help Real Alternatives demonstrate that it is similarly
situated to a religious entity.

However, Real Alternatives does bear some
resemblance to the plaintiffs in March for Life, the district
court decision upon which it heavily relies. There, the
District Court for the District of Columbia granted summary
judgment to a non-profit, secular anti-abortion group on its
equal protection challenge to the Contraceptive Mandate. We
cannot accept the district court’s reasoning in that case.
Relying almost exclusively on Center for Inquiry, the district
court found that the secular group at issue was “similarly

20
situated with regard to the precise attribute selected for
accommodation”—specifically, a shared view that abortion is
wrong. March for Life, 128 F. Supp. 3d at 126 (emphasis
omitted). But that court—and, by extension, Real
Alternatives—ignored a crucial point: Unlike the corporation
in Center for Inquiry, which involved a comprehensive belief
system that happened not to be deity-centric, a secular anti-
abortion group mirrors a single-issue interest group and not a
religious organization that takes advantage of the Exemption.
We agree with Judge Jones’s observation regarding the
disparities between the two groups: “In every other respect,
they are different. Real Alternatives is an employer, a
company, and not a belief system . . . and its single mission
statement cannot guide believers comprehensively throughout
life as a religion can.” J.A. 42; cf. United States v. Seeger,
380 U.S. 163, 187 (1965) (accommodating a secular pacifist’s
objections to the draft because his beliefs “occup[y] the same
place in his life as the belief in a traditional deity holds in the
lives of” adherents to religion).

Real Alternatives is in no way like a religious
denomination or one of its nontheistic counterparts—not in
structure, not in aim, not in purpose, and not in function. We
do not doubt that Real Alternatives’s stance on contraceptives
is grounded in sincerely-held moral values, but “religion is
not generally confined to one question or one moral teaching;
it has a broader scope.” Malnak v. Yogi, 592 F.2d 197, 209
(3d Cir. 1979) (Adams, J., concurring). We have accordingly
noted three “guideposts” courts ought to use when identifying
a religion:

First, a religion addresses fundamental and
ultimate questions having to do with deep and

21
 imponderable matters. Second, a religion is
comprehensive in nature; it consists of a belief-
system as opposed to an isolated teaching.
Third, a religion often can be recognized by the
presence of certain formal and external signs.

Africa v. Pennsylvania, 662 F.2d 1025, 1032 (3d Cir. 1981).
We thus agree with Amici Curiae that “Real Alternatives is
functionally similar not to a church, but to the countless
nonreligious 501(c)(3) nonprofit organizations that take
morally informed positions on some discrete set of issues,”
such as the NAACP and the National Organization for
Marriage. 8 Amici Curiae Br. at 16.

8
We further agree with Amici Curiae that while
commitment to an anti-abortion platform may be important to
the people who hold them, that commitment is “not a religion
in any legally or theologically accepted sense; and
organizations do not become quasi-churches for equal-
protection purposes merely by espousing a commitment of
that sort.” Amici Curiae Br. at 15; see also Malnak, 592 F.2d
at 208–10 (Adams, J., concurring) (defining nontheistic belief
system as “religion” if it (1) deals with questions of “ultimate
concern”; (2) provides answers that speak to comprehensive
and ultimate truth; and (3) has formal characteristics
analogous to those of traditional religions); Wash. Ethical
Soc’y v. District of Columbia, 249 F.2d 127, 128 (D.C. Cir.
1957) (finding nontheistic ethical society that had regular
Sunday meetings, “leaders” who preached to members and
provided spiritual guidance, and ceremonies for naming,
marrying, and burying members qualified for tax exemption
as church); Fallon v. Mercy Catholic Med. Ctr. of Se. Pa.,
200 F. Supp. 3d 553, 562 n.4 (E.D. Pa. 2016) (distinguishing

22
 Real Alternatives overemphasizes its shared opposition
to contraceptive coverage while inexplicably dismissing the
Government’s repeated statements that the Exemption “was
provided against the backdrop of the longstanding
governmental recognition of a particular sphere of autonomy
for houses of worship . . . .” 80 Fed. Reg. at 41,325. But
framing the Exemption—or any religious exemption for that
matter—so broadly as to encompass any employer who
disagrees with any aspect of an underlying law lies in direct
contradiction to the Supreme Court’s refusal to broaden
religion-based exemptions in similar contexts. See United
States v. Lee, 455 U.S. 252, 260–61 (1982) (in a Social
Security-related matter, rejecting a claim to extend a limited
exemption because “[c]onfining the . . . exemption . . .
provided for a narrow category which was readily
identifiable,” and noting that “every person cannot be
shielded from all the burdens incident to exercising every
aspect of the right to practice religious beliefs”). Permitting
Real Alternatives to qualify for the Exemption would
similarly run afoul of this country’s vast history of legislative
protections that single out and safeguard religious freedom

between plaintiff’s beliefs, which “consist[ed] solely of his
‘conscience’ and personal moral code,” and the “structural
characteristics” of secular moral systems that are equivalent
to religion except for non-belief in God); Fellowship of
Humanity v. Cty. of Almaeda, 315 P.2d 394, 409–10 (Cal.
Dist. Ct. App. 1957) (finding that nontheistic fellowship
qualified for tax exemption as church because “it is conceded
that in all respects the Fellowship’s activities are similar to
those of the theistic groups, except for their belief or lack of
belief in a Supreme Being”).

23
but not moral philosophy. See Corp. of the Presiding Bishop
of the Church of Jesus Christ of Latter-Day Saints v. Amos,
483 U.S. 327, 338 (1987) (“Where, as here, [G]overnment
acts with the proper purpose of lifting a regulation that
burdens the exercise of religion, we see no reason to require
that the exemption comes packaged with benefits to secular
entities.”); 9 Religious Land Use and Institutionalized Persons
Act, 42 U.S.C. § 2000cc (“RLUIPA”) (requiring religious
accommodation for zoning and land use regulations);
Employee Retirement Income Security Act of 1974, 29
U.S.C. § 1003(b)(2) (exempting “church plan[s]” from
retirement-plan regulations); Internal Revenue Code, 26
U.S.C. §§ 6033(a)(3)(A)(i), (iii) (carving out “churches, their
integrated auxiliaries, . . . conventions or associations of
churches,” and “the exclusively religious activities of any
religious order” from a tax-filing requirement); Title VII of
the Civil Rights Act of 1964, 42 U.S.C. § 2000e-1(a)
(requiring that employers not discriminate on the basis of
religion). If mere disagreement, however vehemently felt,
were sufficient to bring an equal protection claim, virtually
any law implicating religion would be rendered moot because
the exemption would be too easy to invoke. 10 Cf. Cutter v.

9
We do not find persuasive Real Alternatives’s belabored
efforts to distinguish Amos, and we agree with the District
Court that the Supreme Court’s holding in that case is
applicable here.
10
We share the concerns of Amici Curiae that if such
disagreement were enough to substantiate an equal protection
claim, there would also be strong disincentives from granting
any religious exemption because of how easy it would be to
utilize or to extend it, thereby seriously undermining

24
Wilkinson, 544 U.S. 709, 724 (2005) (acknowledging that “all
manner of religious accommodations would fall” if the Court
struck down one law that implicated religion because of the
similarities among religious accommodation laws).

Finally, even if Real Alternatives were deemed
similarly situated to a religion, the group’s challenge would
still fail because of the historic principle of respect for the
autonomy of genuine religions. This principle provides the
legitimate purpose for the preferential treatment of religious
organizations. The Exemption “provide[s] for a religious
accommodation that respects the unique relationship between
a house of worship and its employees in ministerial
positions.” 76 Fed. Reg. at 46,623. It “was provided against
the backdrop of the longstanding governmental recognition of
a particular sphere of autonomy for houses of worship,” is
“consistent with their special status under longstanding
tradition in our society and under federal law, and is not a
mere product of the likelihood that these institutions hire
coreligionists.” 80 Fed. Reg. at 41,325. Real Alternatives
brazenly dismisses these statements as disingenuous. 11 In

countless legislative and regulatory programs. Relatedly,
there would be immense pressure to repeal the thousands of
religious accommodations that have been enacted at the
federal, state, and local levels for fear that they would become
vehicles to avoid compliance by anyone who dislikes the
underlying laws.
11
Without any supporting evidence, Real Alternatives
repeatedly contends that the Government is asking the Court
“to ignore the actual explanation in its regulations,” i.e., the
likelihood of religious employees using contraceptives, “and

25
doing so, it misses a crucial point about rational basis review:
It is “constitutionally irrelevant whether this [legitimate]
reasoning in fact underlay the legislative decision” because
this Court has never insisted that a legislative body articulate
its reasons for enacting a statute. U.S. R.R. Ret. Bd. v. Fritz,
449 U.S. 166, 179 (1980); see also Beach Commc’ns, 508
U.S. at 318 (applying Fritz to an administrative action). In
any event, the attribute Congress selected for classification is
not opposition to contraceptives; it is status as a house of
worship and based on the long-established governmental
desire to respect the autonomy of houses of worship
regardless of their particular stance on contraceptives.

It is beyond dispute that respecting church autonomy is
a legitimate purpose—one that not only satisfies rational basis
review but also is enshrined in the constitutional fabric of this
country. Principles of noninterference trace back to “the text

instead to suppose that the exemption was offered solely
because of the ‘church character’ of some religious
employers.” Appellants’ Reply Br. at 7. This theory hinges
on the Government’s acknowledgment that “employees of
employers availing themselves of the exemption would be
less likely to use contraceptives even if contraceptives were
covered under their health plans.” 77 Fed. Reg. at 8,728.
While we agree that likelihood of use would not alone satisfy
rational basis review, that statement was part of the
Government’s explanation that the Exemption “does not
undermine the overall benefits” of the Contraceptive
Mandate. Id. It does not negate or in any way undermine the
actual and legitimate purpose of the historic respect for
religion put forth by the Government.

26
of the First Amendment itself, which gives special solicitude
to the rights of religious organizations,” and recognizes their
“independence from secular control or manipulation—in
short, [their] power to decide for themselves, free from state
interference, matters of church government as well as those of
faith and doctrine.” Hosanna-Tabor Evangelical Lutheran
Church & Sch. v. EEOC, 132 S. Ct. 694, 704, 706 (2012)
(quoting Kedroff v. Saint Nicholas Cathedral of Russian
Orthodox Church in N. Am., 344 U.S. 94, 116 (1952))
(internal quotation marks omitted). Even when
noninterference is not strictly required, the Government has
discretion to grant certain religious accommodations subject
to constitutional limitations. 12 See Cutter, 544 U.S. at 720–
22. These accommodations may be extended to houses of
worship and religious denominations without applying to all
nonprofit entities in order to “alleviate significant
governmental interference with the ability of religious

12
The First Amendment prohibits the Government from
inserting itself in theological disputes, appointments of
ministers, or questions of distribution of church property; the
Government may not dictate to houses of worship what to
believe or how to structure their relations with clergy to
implement and teach those beliefs. See, e.g., Hosanna-Tabor,
132 S. Ct. at 706 (employment decisions for ministers);
Serbian E. Orthodox Diocese for the U.S. & Can. v.
Milivojevich, 426 U.S. 696, 713–14 (1976) (internal
theological disputes and religious tribunals); Presbyterian
Church in the U.S. v. Mary Elizabeth Blue Hull Mem’l
Presbyterian Church, 393 U.S. 440, 449 (1969) (church
property); Gonzalez v. Roman Catholic Archbishop of
Manila, 280 U.S. 1, 16 (1929) (appointment of clergy).

27
organizations to define and carry out their religious
missions.” 13 Amos, 483 U.S. at 335; see also Walz v. Tax
Comm’n, 397 U.S. 664, 676 (1970) (upholding a property tax
exemption for houses of worship); Hosanna-Tabor, 132 S. Ct.
at 706 (finding a “special rule for ministers grounded in the
Religion Clauses themselves”).

We reiterate, however, that Real Alternatives cannot
satisfy the first prong of a successful equal protection claim.
Finding all single-issue non-profit organizations to be
similarly situated to houses of worship based on their
adherence to a shared position on one issue would expand
religious exemptions beyond what is constitutionally
required. That a legitimate purpose of the highest order—
respect for religious autonomy—justifies the Exemption only
underscores the inevitability of our conclusion. We therefore
find that Real Alternatives’s equal protection claim fails as a
matter of law.

C. APA Claim

Real Alternatives asserts two claims under the APA:
(1) the Contraceptive Mandate is arbitrary and capricious
because it does not serve a rational governmental purpose as

13
In this way, Center for Inquiry may be distinguished as
an outlier example of organized secular belief systems
gaining protected treatment. The District Court correctly
noted that “the majority of precedent continues to support
preferential treatment for religion under the law, without
explicitly extending that treatment to include secular beliefs.”
J.A. 36.

28
applied to Real Alternatives, an organization that employs
only people who oppose contraceptive coverage; and (2) it
violates the Constitution and federal law. Both claims lack
merit.

1. Legal Standard

A reviewing court may “hold unlawful and set aside
agency action” that is “(A) arbitrary, capricious, an abuse of
discretion, or otherwise not in accordance with law” or “(B)
contrary to constitutional right, power, privilege, or
immunity.” 5 U.S.C. § 706(2)(A)–(B).

We have held that the standard for determining
whether an APA violation exists under the arbitrary and
capricious standard is substantially similar to rational basis
review:
Agency action is arbitrary and capricious if the
agency offers insufficient reasons for treating
similar situations differently. If [an] agency
makes an exception in one case, then it must
either make an exception in a similar case or
point to a relevant distinction between the two
cases. Review of an equal protection claim in
the context of agency action is similar to that
under the APA. That is, an agency’s decision
must be upheld if under the Equal Protection
Clause, it can show a rational basis for its
decision. As such, the equal protection
argument can be folded into the APA argument,
since no suspect class is involved and the only
question is whether the . . . treatment of

29
 [appellees] was rational (i.e., not arbitrary and
capricious).

Nazareth Hosp. v. Sec’y U.S. Dep’t of Health & Human
Servs., 747 F.3d 172, 179–80 (3d Cir. 2014) (alteration in
original) (citations and internal quotation marks omitted)
(emphasis added).

2. Analysis

Because we find that Real Alternatives’s equal
protection claim fails, we need not reexamine its arbitrary and
capricious claim, which is subject to the same standard of
review. Id.

Real Alternatives argues that the Contraceptive
Mandate also violates the APA because it infringes on two
other federal laws: the ACA and the Weldon Amendment.
The Real Alternatives Employees argue that the
Contraceptive Mandate also violates the Church Amendment
and, therefore, the APA. We disagree with these contentions
and find no violations. We address each law in turn.

a. ACA

The ACA states that none of its provisions “shall be
construed to require a qualified health plan to provide
coverage of [abortion] services as part of its essential health
benefits for any plan year.” 42 U.S.C. § 18023(b)(1)(A)(i).
Real Alternatives argues that the Contraceptive Mandate
violates this provision by “requiring coverage of certain
‘FDA-approved contraceptives’ which act as abortifacients, in
that they cause the demise of human embryos after

30
conception and before and/or after implantation in the
uterus.” Appellants’ Br. at 57. Real Alternatives does not
cite any statutory or regulatory definition of abortion that
includes contraceptives. 14

However, longstanding FDA regulations treat
pregnancy as “the period of time from implantation until
delivery,” 45 C.F.R. § 46.202(f), and categorize drugs that
may prevent implantation as contraceptives rather than as
abortifacients. 62 Fed. Reg. 8,610, 8,611 (Feb. 25, 1997)
(“Emergency contraceptive pills are not effective if the
woman is pregnant; they act by delaying or inhibiting
ovulation, and/or altering tubal transport of sperm and/or ova
(thereby inhibiting fertilization), and/or altering the
endometrium (thereby inhibiting implantation).”). Further,
we defer to the Government’s definition because “this Court
will normally accord particular deference to an agency
interpretation of longstanding duration.” Barnhart v. Walton,
535 U.S. 212, 220 (2002) (internal quotation marks omitted).
Thus, we conclude that the Contraceptive Mandate does not
require coverage for abortion services and that Real
Alternatives’s claim to the contrary fails.

b. Weldon Amendment

14
In its brief, Real Alternatives relies on a number of
dictionary definitions to suggest, contrary to statutory and
regulatory definitions, that the modes of contraceptives
covered by the Contraceptive Mandate are capable of
inducing abortion.

31
 Real Alternatives raises a similar claim based on the
Weldon Amendment, which requires that no funds provided
by the ACA’s underlying appropriations bill be made
available to a federal agency or program that “subjects any
institutional or individual health care entity to discrimination
on the basis that the health care entity does not provide, pay
for, provide coverage of, or refer for abortions.” Pub. L. No.
112-74, §§ 506, 507, 125 Stat. 786, 1111–12 (Dec. 23, 2011).
This claim fails for the reasons stated in the preceding
section. 15
c. Church Amendment

The final APA claim asserts a violation of the Church
Amendment, which prohibits an individual from being
required to “perform or assist in the performance of any part
of a health service program or research activity funded . . . by
the Secretary of [HHS] if his performance or assistance . . .
would be contrary to his religious beliefs or moral

15
See also J.A. 51 (noting Representative Weldon’s
statement when proposing the eponymous amendment: “The
provision of contraceptive services has never been defined as
abortion in Federal statute, nor has emergency contraception,
what has commonly been interpreted as the morning-after
pill. Now, some religious groups may interpret that as
abortion, but we make no reference in this statute to religious
groups or their definitions; and under the current FDA policy
that is considered contraception, and it is not affected at all by
this statute.”).

32
convictions.” 16 42 U.S.C. § 300a-7(d). This claim fails for
lack of standing. The Real Alternatives Employees purchase
their health insurance from a company in the health insurance
market, not from HHS or an HHS-administered health
insurance program that falls under the purview of the Church
Amendment. See Geneva Coll. v. Sebelius, 929 F. Supp. 2d
402, 449–50 (W.D. Pa. 2013) (where individuals obtain
health insurance through their employer, who in turn
purchases coverage from the private health insurance market
(and not HHS), the Church Amendment is not implicated)
(citing Ass’n of Data Processing Serv. Orgs., Inc. v. Camp,
397 U.S. 150, 153 (1970)).

D. RFRA

We now turn to the RFRA claim, which presents a
question of first impression for this Court: whether
employees, who oppose contraceptives on religious grounds
but work for secular employers, experience a substantial
burden on their religious exercise when the Government
regulates group health care plans and health care insurance
providers by requiring them to offer health insurance
coverage that includes coverage for services the employees
find incompatible with their religious beliefs. This claim is
distinct from an employer’s RFRA claim objecting to the
mandated provision of contraceptive services that was found

16
The Church Amendment claim was brought only by the
Real Alternatives Employees because the Church
Amendment applies only to individuals.

33
to be meritorious in Burwell v. Hobby Lobby Stores, Inc., 134
S. Ct. 2751, 2775 (2014). 17

Under RFRA, the “Government may substantially
burden a person’s exercise of religion only if it demonstrates
that application of the burden to the person—(1) is in
furtherance of a compelling governmental interest; and (2) is
the least restrictive means of furthering that compelling
governmental interest.” 42 U.S.C. § 2000bb-1(b) (emphasis
added). The “exercise of religion” constitutes “any exercise
of religion, whether or not compelled by, or central to, a
system of religious belief.” 42 U.S.C. § 2000cc-5(7)(A).

Congress enacted RFRA in 1993 in response to the
Supreme Court’s decision in Employment Division v. Smith,
494 U.S. 872 (1990), which overruled the earlier method of
analyzing free exercise claims used in Sherbert v. Verner, 374
U.S. 398 (1963), and Wisconsin v. Yoder, 406 U.S. 205
(1972). See Geneva, 778 F.3d at 430. The earlier decisions
used a balancing test that took into account whether the

17
Echoing the District Court, we state what we consider to
be obvious: Hobby Lobby did not answer the RFRA question
we confront today. In Hobby Lobby, the Supreme Court
found that an employer’s provision, not an individual’s
maintenance, of coverage may violate RFRA. 134 S. Ct. at
2778. As they did before the District Court, the Real
Alternatives Employees ignore this important distinction and
attempt to stretch the holding of Hobby Lobby well beyond its
factual confines. The Dissent similarly misstates the
applicability of Hobby Lobby, characterizing the issue there
as “very like the one at issue here.” Dissent Op. at 14.

34
challenged action imposed a substantial burden on the
practice of religion, and if it did, whether it was needed to
serve a compelling governmental interest. Id. Smith rejected
this test because applying it whenever a person objected on
religious grounds to the enforcement of a generally applicable
law “would open the prospect of constitutionally required
religious exemptions from civic obligations of almost every
conceivable kind . . . .” 494 U.S. at 888.

Courts look to pre-Smith free exercise jurisprudence
when assessing RFRA claims. See Hobby Lobby, 134 S. Ct.
at 2772. The Supreme Court has characterized RFRA as
“adopt[ing] a statutory rule comparable to the constitutional
rule rejected in Smith.” Gonzales v. O Centro Espirita
Beneficente Uniao do Vegetal, 546 U.S. 418, 424 (2006).
RFRA may be applied to situations not previously addressed
under pre-Smith jurisprudence. See Hobby Lobby, 134 S. Ct.
at 2772 (“It is simply not possible to read these provisions as
restricting the concept of the ‘exercise of religion’ to those
practices specifically addressed in our pre-Smith decisions.”).

1. Legal Standard

Religious exercise is impermissibly burdened when
government action compels individuals “to perform acts
undeniably at odds with fundamental tenets of their religious
beliefs.” Yoder, 406 U.S. at 218 (emphasis added).
Accordingly:

Where the state conditions receipt of an
important benefit upon conduct proscribed by a
religious faith, or where it denies such a benefit
because of conduct mandated by religious

35
 belief, thereby putting substantial pressure on
an adherent to modify his behavior and to
violate his beliefs, a burden upon religion
exists.

Thomas v. Review Bd., 450 U.S. 707, 717–18 (1981)
(emphasis added). This Court has found “a substantial
burden exists where: 1) a follower is forced to choose
between following the precepts of his religion and forfeiting
benefits otherwise generally available . . . versus abandoning
one of the precepts of his religion in order to receive a
benefit; OR 2) the [G]overnment puts substantial pressure on
an adherent to substantially modify his behavior and to
violate his beliefs.” Mack v. Warden Loretto FCI, 839 F.3d
286, 304 (3d Cir. 2016).

Important principles circumscribe the RFRA inquiry,
and it is for the reviewing court to determine whether a
burden is “substantial.” We thus reiterate a self-evident
principle that we set forth in Geneva: “While the Supreme
Court reinforced in Hobby Lobby that we should defer to the
reasonableness of the [RFRA claimant’s] religious beliefs,
this does not bar our objective evaluation of the nature of the
claimed burden and the substantiality of that burden on the
[claimant’s] religious exercise.” 18 778 F.3d at 436 (emphasis

18
Although our judgment in Geneva was vacated by the
Supreme Court, it nonetheless sets forth the view of our
Court, which was based on Supreme Court precedent, that we
continue to believe to be correct regarding our duty to assess
substantiality as well as our conclusion that the regulation at
issue there did not impose a substantial burden. Cf. Zubik,
136 S. Ct. at 1560 (specifying that vacatur and remand do not

36
added). As such, “[w]hether a burden is ‘substantial’ under
RFRA is a question of law, not a question of fact.” Id. at 442
(citing Mahoney v. Doe, 642 F.3d 1112, 1121 (D.C. Cir.
2011)); see also Kaemmerling v. Lappin, 553 F.3d 669, 679
(D.C. Cir. 2008) (“Accepting as true the factual allegations
that [plaintiff’s] beliefs are sincere and of a religious nature—
but not the legal conclusion, cast as a factual allegation, that
his religious exercise is substantially burdened . . . .”); cf.
Hobby Lobby, 134 S. Ct. at 2775 (“Because RFRA applies in
these cases, we must next ask whether the HHS contraceptive
mandate ‘substantially burden[s]’ the exercise of religion.”)
(citing 42 U.S.C. § 2000bb-1(a)) (alteration in original)).

RFRA’s legislative history underscores the
requirement that the burden be substantial. The version of
RFRA initially introduced in the House of Representatives

express the Supreme Court’s “view on the merits” of
Geneva). That judgment, and others cited here that addressed
similar claims, was vacated because the Supreme Court
wanted the parties to attempt—after the parties signaled they
might be able—to develop a way for existing or modified
ACA regulations to provide continued contraceptive coverage
to petitioners’ employees and through petitioners’ insurers
without any notice from petitioners. Id. Thus, Zubik vacated
our judgment in Geneva but did not attack our reasoning.
The Dissent mischaracterizes our holding today to be saying
that Geneva is “controlling” for purposes of this case.
Dissent Op. at 18. That is not our position. While Geneva is
no longer controlling, there is nothing that would require us—
or anyone else—to conclude that our reasoning in that
opinion was incorrect.

37
provided only that “Government shall not burden a person’s
exercise of religion” unless the burden satisfied strict
scrutiny. H.R. 1308, 103d Cong. § 3 (1993). It was only later
in the enactment process that it was modified to include the
word “substantially” immediately before “burden.” 42 U.S.C.
§ 2000bb-1(a)–(b); Little Sisters of the Poor Home for the
Aged v. Burwell, 794 F.3d 1151, 1176 (10th Cir. 2015)
(“Congress added the word ‘substantially’ before passage to
clarify that only some burdens would violate the act. 139
Cong. Rec. S14352 (daily ed. Oct. 26, 1993) (statements of
Sen. Kennedy and Sen. Hatch). . . . If plaintiffs could assert
and establish that a burden is ‘substantial’ without any
possibility of judicial scrutiny, the word ‘substantial’ would
become wholly devoid of independent meaning.”), vacated
and remanded sub nom. Zubik, 136 S. Ct. at 1561. This
important change made it explicit that RFRA would provide
relief only from “substantial” government burdens on
religious exercise, not from all government burdens. 19 The

19
See also Matthew A. Melone, Corporations and
Religious Freedom: Hobby Lobby Stores—A Missed
Opportunity to Reconcile a Flawed Law with a Flawed
Health Care System, 48 Ind. L. Rev. 461, 502–03 (2015)
(“[T]he lack of any principled limitation on the meaning of
religious exercise should prompt the courts to examine
whether any burden on such exercise is substantial.
Otherwise, RFRA becomes anarchical. . . . The notion that the
judiciary has no business questioning the substantiality of a
burden in this context is illogical. The law imposes objective
standards on beliefs in other contexts and appears to do so
without inordinate difficulty. . . . Every person has the right to
attach whatever religious meaning to act to an act their
conscience demands. The law, however, should not be

38
Dissent would have us read “substantial” out of the statute,
revert to a never-enacted version of RFRA, and supplant our
charge to conduct judicial review of a RFRA claim with
automatic deference to the claimant. 20 We will not. 21

hostage to the vagaries of the hypersensitive.”) (footnotes
omitted).
20
For a persuasive discussion of the untenable
consequences of the Real Alternatives Employees’ and
Dissent’s theory of absolute deference to an allegation that a
burden is substantial, see Brief of Baptist Joint Committee for
Religious Liberty as Amicus Curiae in Support of
Respondents 14–16, Zubik v. Burwell, 136 S. Ct. 1557 (2016)
(Nos. 14-1418, 14-1453, 14-1505, 15-35, 15-105, 15-119, 15-
191).
21
The Dissent grounds its aversion to judicial review of
substantiality in a Tenth Circuit dissent from the denial of en
banc review in the Little Sisters case, and points to James
Madison’s critique of the “notion that a civil judge can be a
competent Judge of Religious Truth” for support. Dissent
Op. at 42 (internal quotation marks omitted). Madison’s
writings are indeed instructive, as our refusal today to permit
a claimant’s bare allegations to automatically render a burden
“substantial” is embedded in Madison’s Federalist No. 10:
“No man is allowed to be a judge in his own cause, because
his interest would certainly bias his judgment, and, not
improbably, corrupt his integrity. With equal, nay with
greater reason, a body of men are unfit to be both judges and
parties at the same time.” The Federalist No. 10, 59 (James
Madison).

39
 There is no substantial burden if the governmental
action does not coerce the individuals to violate their religious
beliefs or deny them the “rights, benefits, and privileges
enjoyed by other citizens”—even if “the challenged
Government action would interfere significantly with private
persons’ ability to pursue spiritual fulfillment according to
their own religious beliefs.” Lyng, 485 U.S. at 449. Nor can
a party use RFRA to “require the Government to conduct its
own internal affairs in ways that comport with the religious
beliefs of particular citizens.” Bowen, 476 U.S. at 699.
“Congress has required qualitative assessment of the merits of
. . . RFRA claims.” Geneva, 778 F.3d at 435. At the same
time, we must be careful to conduct only a review into the
substantiality of the religious burden and not to question the
reasonableness of the religious belief itself. See Hobby
Lobby, 134 S. Ct. at 2778 (RFRA does not permit courts to
address “whether the religious belief asserted in a RFRA case
is reasonable”). Courts are not to accept every allegation of
substantial burden. To the contrary, RFRA’s demand for
judicial review has been recognized by the Supreme Court, 22
by this Court in Geneva, and by virtually all of our sister
circuits, which have not hesitated to examine whether an
alleged burden is sufficiently “substantial” under RFRA. 23

22
See Bowen, 476 U.S. at 700–01 n.6 (“[F]or the
adjudication of a constitutional claim, the Constitution, rather
than an individual’s religion, must supply the frame of
reference.”).
23
See Little Sisters, 794 F.3d at 1176–77 (“RFRA’s
statutory text and religious liberty case law demonstrate that
courts—not plaintiffs—must determine if a law or policy
substantially burdens religious exercise” and finding alleged

40
burden not “substantial.”) (emphasis added); Priests for Life
v. U.S. Dep’t of Health & Human Servs., 772 F.3d 229, 247
(D.C. Cir. 2014) (“Accepting the sincerity of plaintiffs’
beliefs, however, does not relieve this Court of its
responsibility to evaluate the substantiality of any burden on
plaintiffs’ religious exercise . . . .”), vacated and remanded
sub nom. Zubik, 136 S. Ct. at 1561; E. Tex. Baptist Univ. v.
Burwell, 793 F.3d 449, 456 (5th Cir. 2015) (“We begin and
end our analysis with the substantial-burden prong. The
plaintiffs must show that the challenged regulations
substantially burden their religious exercise . . . .”), vacated
and remanded sub nom. Zubik, 136 S. Ct. at 1561; Korte v.
Sebelius, 735 F.3d 654, 683 (7th Cir. 2013) (characterizing
impermissible questions about the “centrality of the religious
practice to the adherent’s faith” as distinct from the
substantial burden inquiry, which “evaluates the coercive
effect of the governmental pressure on the adherent’s
religious practice and steers well clear of deciding religious
questions”); Navajo Nation v. U.S. Forest Serv., 535 F.3d
1058, 1068 (9th Cir. 2008) (en banc) (characterizing the
“crux” of the RFRA case as determining whether the
Government “impose[d] a ‘substantial burden’ on the exercise
of the Plaintiffs’ religion” and finding no substantial burden
where Government sought to use artificial snow for skiing on
a mountain sacred to Indian tribe claimant); Branch
Ministries v. Rossotti, 211 F.3d 137, 142–43 (D.C. Cir. 2000)
(directly assessing whether claimant’s alleged religious
burden was sufficiently substantial under RFRA and finding
that it was not); see generally infra note 37. But see Sharpe
Holdings, Inc. v. U.S. Dep’t of Health & Humans Servs., 801
F.3d 927, 939 (8th Cir. 2015) (“[O]ur narrow function . . . in
[the RFRA] context . . . is to determine whether the line

41
Rather than confront this precedent, our dissenting colleague
would prefer to ignore the import, even the existence, of the
“substantial” qualifier in the RFRA test. The Dissent reduces
our position to say that “[r]eligious beliefs are not being
burdened in any meaningful sense, so people should just stop
complaining.” Dissent Op. at 2. But whether the alleged
burden is “meaningful”—or, more accurately, “substantial”—
is not a question that can be so easily dismissed with a
reductionist turn of phrase. To the contrary, it is the very
essence of a RFRA claim, the threshold inquiry posed to any
individual attempting to bring a successful RFRA claim, and
it is undoubtedly for the court to answer whether it has been
satisfied. 24 Turning to the burden alleged by the Real
Alternatives Employees, we will now do just that.

drawn reflects an honest conviction.” (internal quotation
marks omitted) (alteration and omission in original)), vacated
and remanded on other grounds sub nom. U.S. Dep’t of
Health & Human Servs. v. CNS Int’l Ministries, 136 S. Ct.
2006 (2016).
24
Urging that we are wrongly questioning the “validity,”
Dissent Op. at 10, and “legitimacy,” id. at 16, of the Real
Alternatives Employees’ religious beliefs, the Dissent
conflates our dual responsibilities in adjudicating a RFRA
claim: accepting the sincerity of a RFRA claimant’s religious
belief and deciding whether the alleged burden is
“substantial.” See Hobby Lobby, 134 S. Ct. at 2778 (noting
that the RFRA presents the question of “whether the
[Contraceptive Mandate] imposes a substantial burden” on
the claimant, and not “whether the religious belief asserted . .
. is reasonable”); Little Sisters, 794 F.3d at 1176
(“[A]ccepting any burden alleged by Plaintiffs as ‘substantial’

42
 2. Analysis

RFRA centers on the intersection between the specific
conduct in which the objector is forced to engage and his or
her religious exercise, and that is where we begin our
analysis. The Real Alternatives Employees characterize their
purchase of insurance as somehow enabling the provision of
contraceptives, thereby substantially burdening their religious
exercise. They allege that their “sincerely held religious
beliefs prohibit them from [(1)] using, [(2)] supporting, or
otherwise [(3)] advocating the use of abortifacients, or [(4)]
participating in a health insurance plan that covers such items
for themselves or their families.” J.A. 123 (emphasis added).
We address each enumerated allegation in turn, and we
conclude that the Real Alternatives Employees have failed to
demonstrate that the Contraceptive Mandate forces them to
violate their religious beliefs.

The act complained of is the signing on to coverage
followed by the request for reimbursement of services chosen.
That basic scheme is the same for any individual whose
employer provides him or her with insurance: The plan
deems the employee eligible to be reimbursed for hundreds of
different services, and that employee can take advantage of
that eligibility as he or she sees fit. Should the employee opt
to use a particular service, he or she fills out a form and asks
to be paid back for costs incurred. In the end, the employee
uses a covered service, or not; either way, there is no

would improperly conflate the determination that a religious
belief is sincerely held with the determination that a law or
policy substantially burdens religious exercise.”).

43
requirement to support or advocate for whatever service he or
she, or others, selects. Checking off a box to be eligible for
reimbursement of services—services of the employee’s
choosing—in no way indicates, let alone suggests, support or
advocacy for that service. The disconnect between the use of
any one service and the use of contraceptives is arguably even
greater—and it calls into question the “substantiality” of the
purported burden. After all, a substantial burden on the
exercise of religion exists only where the Government
“demands that [an individual] engage in conduct that
seriously violates [his or her] religious beliefs,” Hobby
Lobby, 134 S. Ct. at 2775 (emphasis added), and such
engagement, as discussed in the following sections, is clearly
lacking here.

We are then left with the fourth proscribed conduct
that is central to the RFRA claim: participation. As with
their equal protection claim, the Real Alternatives Employees
rely primarily on March for Life for key support, as the
district court there reasoned that “participating in” a health
insurance plan, by its very nature, effects a substantial change
in behavior because “health insurance does not exist
independently of the people who purchase it.” 128 F. Supp.
3d at 129. There, the district court found that, “[g]iven the
nature of health insurance, [employees] do play a role in the
health care plans that provide contraceptive coverage.” Id.
While characterizing what employees do by subscribing to a
plan as “playing a role,” March for Life would have us
position this fact pattern in lockstep with Hobby Lobby. But
do employees really “play a role?” The Real Alternatives
Employees, along with the Dissent, assume the affirmative,
relying on March for Life’s treatment of the concepts “buy

44
into” and “participate in” as interchangeable. But they are
not.

Subscribing to an insurance plan involves no real
“participation,” just as there is no active “participation” when
subscribing to a magazine or joining AARP or enrolling in a
credit card that has membership benefits. These are all
packages that involve a one-time enrollment, followed by
essentially passive eligibility for certain services that the
member opts in or out of. By declaring that an insurance plan
does not exist without participants, the district court in March
for Life somehow equates the plan with the employees as if
they actively engage in a way that—were it factual—might be
objectionable. Let us be clear: There is no “participation” in
the real sense of the word. The employee pays for coverage
and thereby avails him or herself of the ability to be
reimbursed for services. But payment for the ability to have
coverage does not give an employee an active “role” in the
underlying plan. The insurance company offers a package of
health benefits, including certain benefits mandated by the
Government. The plan does not assure the availability of
specific services. Those services are for the employee to seek
out and use—or not. And the employee, by merely
subscribing to that plan in the first instance, is even less
directly related to whatever specific services he or she, or
anyone else, might or might not use later on. 25 The

25
One could analogize that a bank does not “exist
independently” of its individual accountholders, whose
money the bank lends at interest in order to earn profit. But
the accountholders have no say in lending decisions (what
rates to charge, which borrowers to lend to) and no direct
control over the bank. They, like a subscriber to an insurance

45
employees’ actions under the ACA are mediated by the
insurance company, and any link between the decision to sign
up for insurance on the one hand and the provision of
contraceptives to a particular individual on the other is “far
too attenuated to rank as substantial.” Hobby Lobby, 134 S.
Ct. at 2798–99 (Ginsburg, J. dissenting).
This attenuation is fatal to the RFRA claim. Cases
finding a substantial burden under RFRA have consistently
done so where, unlike here, there is a burden that interfered
with the claimants’ exercise and religion is directly
implicated by federal action. See Hobby Lobby, 134 S. Ct. at
2751 (provision required employer-plaintiffs to provide

plan, are offered a panoply of services that are
predeterminately attached to whichever account (or plan) they
choose. Some are desirable to the accountholder and some
are not. Assume that the individual’s bank account is
mandated by the Government under a privatized Social
Security regime, for example. If an accountholder had a
religious objection to the bank’s practices—lending money at
interest—we do not see how that accountholder could
successfully vindicate his or her religious beliefs through
RFRA. So too in the context of health insurance, every
participant pays a premium so that the health insurer will
provide coverage, and every participant also receives (some
of) the benefits of that coverage as they so choose. But
paying a premium simply is not equivalent to active
participation—at a minimum, the insured employee has no
say in what benefits the insurance company will offer or to
whom—and “playing a role,” however important to the plan’s
existence, does not automatically translate into experiencing a
burden, let alone a substantial one.

46
contraceptive coverage in any group plan that they provided
to their employees); Church of the Lukumi Babalu Aye, Inc. v.
City of Hialeah, 508 U.S. 520, 527–28 (1993) (ordinance
prohibited plaintiffs from sacrificing animals); Lee, 455 U.S.
at 254 (statute required plaintiffs to pay Social Security
taxes); Thomas, 450 U.S. at 712 (law denied plaintiff
unemployment benefits); Sherbert, 374 U.S. at 399–400
(same); Yoder, 406 U.S. at 207 (law required plaintiffs to
send their children to school); see also Civil Liberties for
Urban Believers v. City of Chi., 342 F.3d 752, 761 (7th Cir.
2003), (holding, in the context of RLUIPA, that “a substantial
burden on religious exercise is one that necessarily bears
direct, primary, and fundamental responsibility for rendering
religious exercise . . . effectively impracticable”), reh’g en
banc denied; cf. Fernandez v. Mukasey, 520 F.3d 965, 966
(9th Cir. 2008) (per curiam) (“Petitioners have failed to
establish that the [statutory provision at issue] places a
substantial burden on their religious exercise under RFRA. . .
. [T]he connection between [the statutory requirement and
their religious exercise] is too attenuated to create a
substantial burden on petitioners’ religious exercise.”)
(footnote omitted).

These cases underscore that the connection between
the conduct and the religious belief matters, 26 for “the law

26
While the Dissent urges that whether a burden is direct
or indirect is no matter, even March for Life intimated
otherwise. Adopting Priests for Life, the District Court in
March for Life stated that “it is true that [a]n asserted burden
is also not an actionable substantial burden when it falls on a
third party, not the religious adherent.” 128 F. Supp. 3d at
129 (alteration in original). It then reasoned: “Even though

47
distinguishes between direct participation and remote
facilitation, treating the former as compelling and the latter as
negligible.” 27 Amy J. Sepinwall, Conscience and Complicity:
Assessing Pleas for Religious Exemptions in Hobby Lobby’s
Wake, 82 U. CHI. L. REV. 1897, 1938 (2015). The
Government is not mandating an endorsement, or preventing
someone from sacrificing an animal as part of a religious
ritual, or anything of that nature. The Contraceptive Mandate
increases the number of choices an employee has when he or
she purchases health insurance—in this case, broadening the
availability of services that an employee might or might not
access. But that is all it is: a choice. It is still up to the
employee to decide what to do with those options, to seek out
relevant providers, to submit claims for reimbursement for the
service he or she selects, and so on. The act complained of—
the filling out of a form that triggers eligibility for
reimbursement for services the employee chooses to use (or
not)—has not changed, and it in no way amounts to the sort
of “substantial” burden consistently found contrary to

the plaintiffs are not the direct objects of the Mandate, they
are [] very much burdened by it.” Id. By its own logic,
March for Life acknowledged that directness matters in
assessing whether there is an “actionable substantial burden,”
but then found a different means (by erroneously focusing on
participation) of concluding that plaintiffs were nonetheless
“very much burdened.”
27
See also Elizabeth Sepper, Substantiating the Burdens
of Compliance, 2016 U. ILL. L. REV. ONLINE 53, 68 (noting
that courts in multiple areas of law, including criminal law
and torts, “evaluate[] burdens along a scale between
directness and attenuation”).

48
RFRA. 28 And the possibility that others might avail
themselves of services that the employees find objectionable
is no more burdensome than filling out the form in Geneva. 29

28
This point is particularly relevant in light of the Real
Alternatives Employees’ allegation that the Contraceptive
Mandate “fundamentally chang[es] the compensation package
that can be offered to the individual employees.” J.A. 123.
That change, fundamental or not, still does not alter the nature
of the conduct that the employees engage in when signing up
for, or submitting a claim for reimbursement under, an
insurance plan.
29
The Dissent criticizes our consideration of how directly
the burden affects the religious exercise and highlights the
Supreme Court’s statement in Lyng that “indirect coercion or
penalties on the free exercise of religion, not just outright
prohibitions, are subject to scrutiny under the First
Amendment.” Dissent Op. at 26 (citing 485 U.S. at 450).
But the Dissent ignores the remainder of that paragraph,
which specifically warns against implying from that
observation that “incidental effects of government programs,
which may make it more difficult to practice certain religions
but which have no tendency to coerce individuals into acting
contrary to their religious beliefs, require [G]overnment to
bring forward a compelling justification for its otherwise
lawful actions.” 485 U.S. at 450–51. Subsequent appellate
courts applying Lyng have heeded that advice. Cf. Klem, 497
F.3d at 279 (Lyng did not “hold that its conclusion must be
read to mean that any incidental effect of a government
program which may have some tendency to coerce
individuals into acting contrary to their religious beliefs
satisfies the substantial burden standard.”).

49
Unlike in Hobby Lobby, which literally required the objecting
employers to “arrange for” contraceptive coverage in a way
that effectively amounted to sponsorship, 134 S. Ct. at 2775,
the Contraceptive Mandate requires nothing of the employees
that implicates their religious beliefs as stated. There is a
material difference between employers arranging or providing

The Dissent further aims to supplement its mistaken view
of “substantial burden” by couching it in the context of the
recent Supreme Court case Trinity Lutheran Church of
Columbia, Inc. v. Comer, 137 S. Ct. 2012 (June 26, 2017),
pointing to that decision as demonstrative of the idea that
“laws that coerce religious claimants to disavow their religion
in order to receive a government benefit[] are inconsistent
with our constitutional traditions.” Dissent Op. at 34. But
Trinity Lutheran has no real bearing on the specific question
before us today. As our dissenting colleague implicitly
recognizes, Trinity Lutheran is not a RFRA case. It dealt
with a church’s constitutional challenge to a state program
that automatically denied grants to any applicant owned or
controlled by a religious entity. 137 S. Ct. at 2017. “[T]he
[state program’s] policy put[] Trinity Lutheran to a choice: It
may participate in an otherwise available benefit program or
remain a religious institution.” Id. at 2021–22. The question
before the Supreme Court addressed only the treatment of an
institution based on its religious status, not the effect of a
federal program on individual religious beliefs. Signaling its
intent to confine its holding to the particular facts and issue
before it, the opinion noted: “This case involves express
discrimination based on religious identity with respect to
playground resurfacing. We do not address religious uses of
funding or other forms of discrimination.” Id. at 2024 n.3.

50
an insurance plan that includes contraception coverage—so
that employees can avail themselves of that benefit—and
becoming eligible to apply for reimbursement for a service of
one’s choosing. 30

30
By contrast, “[t]he religious costs at issue in Hobby
Lobby were generated by the owners’ direct participation in
the purportedly wrong act—arranging and paying for the
coverage of emergency contraception that they knew would
be used by at least some employees and beneficiaries of their
health plan. While one might have argued, as Justice
Ginsburg did, that the independent decisions of employees
and beneficiaries to use contraception were something like
‘intervening causes’ which cut off the owners’ responsibility,
it is also reasonable to conclude that those third-party
decisions are insufficient to terminate responsibility when
owners’ themselves are required to arrange and (partially) pay
for coverage of the objectionable contraceptives.” Frederick
Mark Gedicks, “Substantial” Burdens: How Courts May
(and Why They Must) Judge Burdens on Religion Under
RFRA, 85 GEO. WASH. L. REV. 94, 147 (2017) (footnotes
omitted) (first emphasis added); see also Geneva, 778 F.3d at
436–37 (“The issue of whether there is an actual burden was
easily resolved in Hobby Lobby, since there was little doubt
that the actual provision of services did not render the
plaintiffs ‘complicit.’ And in Hobby Lobby, the Court came
to its conclusion that, without any accommodation, the
contraceptive coverage requirement imposed a substantial
burden on the religious exercise of the for-profit corporations,
because those plaintiffs were required to either provide health
insurance that included contraceptive coverage, in violation
of their religious beliefs, or pay substantial fines.”) (final
emphasis added). The contrast with this case, which the

51
 The Real Alternatives Employees ultimately fail to
grasp that one size does not fit all: The fact that the
Government may require insurers to offer coverage for
expenditures for certain services that some might find
objectionable on religious grounds cannot form the basis of
requiring the Government to adjust its programs on behalf of
all employees. The categories of services that could offend
religious beliefs is wide-ranging and, as discussed infra,
denying reimbursement for such services to all on the basis of
the religious objections of some would be neither desirable
nor administrable. It is certainly not mandated under RFRA,
which has long protected against substantial, usually direct,
burdens on the individual bringing the claim, not those utterly
disconnected from the claimants themselves.

In fact, the only behavior that the Contraceptive
Mandate governs is the behavior of a third party, the insurer.
And as Amici Curiae rightly note, RFRA does not afford the
Real Alternatives Employees a “religious veto over
governmental action that affects them only incidentally and
does not coerce them to violate their faith.” Amici Curiae Br.
at 24. This principle, that a RFRA claimant show that a
penalty or benefit be more than incidental in order to amount
to a substantial injury, is well-rooted in RFRA jurisprudence.
In Lyng, the Supreme Court rejected the RFRA claimants’

Dissent fails to reconcile, is abundantly clear: Whereas an
employer fashioning a plan for employees and offering it to
them might arguably signal approval of that plan and its
contents, the employee’s act of signing up for a pre-defined
health insurance plan that provides reimbursement for
services that include contraceptive services does not.

52
free exercise claim because the injury only amounted to an
incidental effect. 485 U.S. at 453. The Court held that the
indirect burden cases “cannot imply that incidental effects of
government programs, which may make it more difficult to
practice certain religions but which have no tendency to
coerce individuals into acting contrary to their religious
beliefs, require [G]overnment to bring forward a compelling
justification for its otherwise lawful actions.” Id. at 450–51.
As discussed at length supra, in passing RFRA, Congress
bolstered Lyng’s reading of the Free Exercise Clause with
RFRA’s text 31 and legislative history. 32 We incorporated this
logic in Geneva, finding that “free exercise jurisprudence
instructs that we are to examine the act the [claimants] must
perform—not the effect of that act—to see if it burdens
substantially the [claimants] religious exercise,” 778 F.3d at
440, and we reinforce that conclusion today. 33

31
42 U.S.C. § 2000bb-1 (“Government shall not
substantially burden a person’s exercise of religion even if the
burden results from a rule of general applicability, except as
provided in subsection (b) of this section.”) (emphasis added).
32
S. Rep. 103-111, 9 (1993) (“The act thus would not
require such a justification for every government action that
may have some incidental effect on religious institutions.”).
33
Other courts have come to similar conclusions in
various contexts. Newdow v. Peterson, 753 F.3d 105, 109–10
(2d Cir. 2014) (holding that a currency’s slogan did not
substantially burden the plaintiff’s free exercise rights);
United States v. Friday, 525 F.3d 938, 947 (10th Cir. 2008)
(holding that requiring some to receive a permit before
engaging in a religiously mandated activity did not
substantially burden their free exercise rights); Henderson v.

53
Kennedy, 253 F.3d 12, 17 (D.C. Cir. 2001) (holding that a law
did not substantially burden people’s free exercise rights by
preventing them from distributing religious shirts on the
National Mall because they “can still distribute t-shirts for
free on the Mall, or sell them on streets surrounding the
Mall”); Goodall by Goodall v. Stafford Cty. Sch. Bd., 60 F.3d
168, 172–73 (4th Cir. 1995) (“We find that the financial
burden which the [RFRA claimants] must bear in order to
provide [their son] with a cued speech interpreter at his
private sectarian school does not constitute a substantial
burden under RFRA.”); Smith by Smith v. Bd. of Educ., 844
F.2d 90, 94 (2d Cir. 1988) (holding that a school did not
substantially burden a student’s free exercise rights by
holding graduation on the Sabbath); Azeez v. Fairman, 795
F.2d 1296, 1300 (7th Cir. 1986) (holding that an
administrative name change procedure did not substantially
burden a prisoner’s free exercise rights); Friedman v. Bd. of
Cty. Comm’rs, 781 F.2d 777, 791 (10th Cir. 1985) (holding
that a county seal did not substantially burden the plaintiff’s
free exercise of religion); Lakewood, Ohio Congregation of
Jehovah’s Witnesses, Inc., v. City of Lakewood, 699 F.2d 303,
307–08 (6th Cir. 1983) (finding that a local ordinance did not
substantially burden a church’s free exercise rights by
preventing the church from constructing a new church in just
ten percent of a city); Walsh v. Louisiana High Sch. Athletic
Ass’n, 616 F.2d 152, 158 (5th Cir. 1980) (finding that an
interscholastic athletic rule did not substantially burden the
plaintiff’s free exercise rights by preventing him from
competing in interscholastic high school sports for a year
after a transfer); Berman v. Bd. of Elections, 420 F.2d 684,
686 (2d Cir. 1969) (holding, in the alternative, that the
Government did not substantially burden a voter’s free

54
 RFRA precedent further instructs that the Real
Alternatives Employees’ requested remedy, lifting a penalty
imposed on a third party—the insurer—would run afoul of
this Court’s and others’ findings that individuals cannot use
RFRA to compel the Government to structure its relations
with a third party in a certain way. “The Supreme Court has
consistently rejected the argument that an independent
obligation on a third party can impose a substantial burden on
the exercise of religion in violation of RFRA . . . .” Geneva,
778 F.3d at 440–41 (outlining cases); see also Estate of
Thornton v. Caldor, Inc., 472 U.S. 703, 710 (1985)
(recognizing as “a fundamental principle of the Religious
Clauses” that “[t]he First Amendment . . . gives no one the
right to insist that in pursuit of their own interests others must
conform their conduct to his own religious necessities”)
(omission in original) (internal quotation marks omitted); E.
Texas Baptist Univ. v. Burwell, 793 F.3d 449, 459 (5th Cir.
2015) (“RFRA confers no right to challenge the independent
conduct of third parties . . . .”), vacated and remanded sub
nom. Zubik, 136 S. Ct. at 1561; Priests for Life, 772 F.3d at
246 (“[N]o RFRA right to be free from the unease, or even
anguish, of knowing that third parties are legally privileged or
obligated to act in ways their religion abhors.”); Ave Maria
Found. v. Sebelius, 991 F. Supp. 2d. 957, 965–66 (E.D. Mich.
2014) (“[A] great number of religious objections based on
third-party actions are dismissed simply because the plaintiff
is not pressured to act in any way.”) (citing cases).

exercise rights when it accommodated his religious
opposition to voting in a church by allowing him to change
voting districts and vote by absentee ballot).

55
 Before we end our discussion of the “substantial
burden” inquiry under RFRA, we note that while the Dissent
would downplay the workability concerns exposed by the
District Court regarding the ramifications of finding a
substantial burden here, we believe they are real. As one
Seventh Circuit Court of Appeals jurist observed,
“contraceptive care is by no means the sole form of health
care that implicates religious concerns.” J.A. 66 (citing Grote
v. Sebelius, 708 F.3d 850, 866 (7th Cir. 2013) (Rovner, J.,
dissenting)). Medical treatments that some might view as
objectionable are as varied as they are numerous. Examples
that are by no means exhaustive include artificial
insemination and other reproductive technologies; genetic
screening and counseling; preventative and remedial
treatment for sexually transmitted diseases; sex reassignment;
vaccination; organ transplant from deceased donors; blood
transfusions; euthanasia or physician-assisted suicide; and so
on. See id. (noting that “in some religions, virtually all
conventional medical treatments[] are objectionable”). By
extension, “[a] finding that coverage for one set of
objectionable services constitutes a substantial burden would
imply that coverage for all such services imposes a substantial
burden”—an implication that would “render the health care
system totally unworkable.” Id.; see also Navajo Nation, 535
F.3d at 1072 (“[G]overnment simply could not operate if it
were required to satisfy every citizen’s religious needs and
desires.”).

The Dissent parrots March for Life’s dismissal of these
workability concerns, pointing to the incentives of insurance
companies as safeguards against “a world in which the
[G]overnment would require third-party insurance companies
to provide coverage in every possible form requested by an

56
individual on religious grounds.” 34 March for Life, 128 F.
Supp. 3d at 132; see also Dissent Op. at 35–39. But the
incentives argument is off-point and not curative of our
concerns. The Dissent transplants March for Life’s
discussion of insurance companies’ incentives—reviewed
there in the context of deciding whether the Government
satisfied the third “least restrictive” prong of the RFRA
analysis—into its analysis of the first “substantial burden”
prong. And even if insurance companies’ incentives were
relevant, they would still not satisfy our concerns. The
district court’s presumption in March for Life, the backbone
of the Dissent’s rebuttal here, is that “[i]insurance companies
have every incentive to maintain a sustainable and
functioning market . . . .” Dissent Op. at 37–38 (alteration in
original) (quoting 128 F. Supp. 3d at 132). This is a false
premise: Insurance companies have an interest in a
sustainable and functioning insurance market only to the
extent that it is profitable for them. 35 Nor is the identification

34
Gonzales, the only other case that the Dissent cites to
address workability, said nothing about our concerns
regarding the end-run on legislation that a ruling in favor of
the Real Alternatives Employees would unleash. See Dissent
Op. at 35–36 (citing Gonzales, 546 U.S. at 436).
35
The Dissent would prefer that “we leave to the
insurance companies themselves the decision of what
coverage options they can profitably provide.” Dissent Op. at
39. By the Dissent’s logic, any regulation of any market is
unnecessary because sellers in any market presumably have
some interest in keeping that market functioning. Why
require car manufacturers to provide seatbelts if market
demands will necessitate them anyway? It is entirely within

57
of an insurance company that is allegedly willing to provide a
satisfactory plan relevant to our analysis. The RFRA test
does not ask whether a claimant is able to offset a purported
burden with an alternative scheme of his or her choosing, and
neither the Real Alternatives Employees nor the Dissent have
pointed to any case indicating otherwise. 36

Our inquiry today urges an examination of the claimed
substantial nature of an alleged burden. This approach
contrasts sharply with that of the district court in March for
Life, which assumed—without any analysis whatsoever—the
“substantial” nature of the so-called burden of “participating”
in an insurance plan. Id. at 129–30. Yet we arguably need
not even address the issue of whether the employee’s choice
is coercive when the so-called burden of signing up for
coverage that might enable themselves or others to be
reimbursed for various services is clearly not substantial. No
matter how sincerely held their beliefs may be, we cannot
accept at face value that subscribing to the plan imposes a
“substantial burden.” Surely the word “substantial” is a
matter of subjectivity, not as to genuineness of belief but as to
the nature and extent to which religious exercise is hampered
or restrained by the conduct in question. It is, after all, an
imperative safeguard, else religious beliefs would invariably
trump government action.

the legislature’s prerogative to regulate an industry regardless
of whether that industry may otherwise and on its own
impose similar regulations.
36
The existence of an alternative plan is only relevant to
standing and questions of redressability.

58
 In characterizing the facts in an inaccurate manner,
sidestepping the statutory text, legislative history, and
controlling case law entirely, the Real Alternatives
Employees urge us to put an active gloss on what is
essentially a passive commercial monetary decision:
enrolling in a plan so as to be reimbursed for services of
which one later chooses to avail him or herself. Viewing the
situation for what it is compels us to conclude that whatever
burden there might be, it is certainly not substantial.

Because we conclude that the Real Alternatives
Employees have not—and cannot—show that the
Contraceptive Mandate imposes a substantial burden on their
religious beliefs, we need not reach the question of whether
the Contraceptive Mandate is the least restrictive means of
furthering a compelling governmental interest. 37
III. CONCLUSION

For the foregoing reasons, we will affirm the District
Court’s order denying Appellants’ motion for summary

37
We note that the Dissent’s assertion that “[t]ime and
again courts have rejected the regulation because it is not the
least restrictive means of achieving its objective,” Dissent Op.
at 4, is simply wrong, for only one case in addition to March
for Life has addressed the precise question before us today.
That case, Wieland v. United States Department of Health
and Human Services, 196 F. Supp. 3d 1010 (E.D. Mo. 2016),
essentially adopted the reasoning of March for Life in finding
for the RFRA claimants and did not perform a meaningfully
distinct analysis.

59
judgment in its entirety and granting the Government’s cross-
motion for summary judgment in its entirety.

60
Real Alternatives, et. al. v. Burwell, et al., No. 16-1275
JORDAN, Circuit Judge, concurring in part in the judgment
and dissenting in part.

Not so long ago, the idea of making nuns sign
government documents they believe would involve them in
grievous sins relating to life and death, or forcing devout
Mennonites to pay for health insurance coverage for drugs
and devices they view as abortifacients, would probably have
been unthinkable in this country. Then came the Patient
Protection and Affordable Care Act, known variously as
Obamacare, or the Affordable Care Act, or the ACA. It has
trailed in its wake a number of highly contentious lawsuits
but none more intensely fought than the ones in which the
government has sought to sweep aside the religious
objections of individuals and organizations opposed to the
portion of the ACA called the “Contraceptive Mandate.”
That feature of the statute, which requires non-grandfathered
group health care plans to include coverage for certain
controversial contraceptive items, was at the center of the
aforementioned disputes involving the nuns and the
Mennonites. See Burwell v. Hobby Lobby Stores, Inc., 134 S.
Ct. 2751 (2014); Little Sisters of the Poor Home for the Aged
v. Burwell, 794 F.3d 1151, 1168 (10th Cir.), vacated and
remanded sub nom. Zubik v. Burwell, 136 S. Ct. 1557 (2016).
And it is here for a return engagement in this case.

Having been beaten back in earlier efforts to force the
Contraceptive Mandate on the populace, the government has
changed its tune a bit – it has come up with a new rationale
for its erratically aggressive enforcement of that feature of the
ACA – but the song it sings is essentially the same:
individuals whose faith prompts sincere opposition to paying

1
for or facilitating the purchase of contraceptives cannot be
heard to object; the only thing legitimately at issue is the
regulation of insurance markets. According to the
government, the Mandate has nothing to do with deep
questions about the beginning of life, or the boundaries of
moral culpability, or about faith and one’s obligations to God.
Religious beliefs are not being burdened in any meaningful
sense, so people should just stop complaining. That is the
line pressed by the United States Department of Justice, and it
is the line accepted by my colleagues in the Majority, but I
reject it.

Even if this case could properly be characterized as
nothing more than an examination of insurance markets,
though, I could not agree with my friends in the Majority on
the central point of the dispute. They believe that citizens
who buy health insurance are ciphers, that they do not have
any “‘participation’ in the real sense of the word” when it
comes to the coverage they sign up and pay for, and therefore
the answer to the question “do employees really ‘play a role’”
in the market for health care services is, according to my
colleagues, a resounding no. (Maj. Op. at 45.) I disagree.
After the federal government gave itself a vastly greater role
in the health insurance market, there has no doubt been less
room for decision making by individual purchasers. But that
does not mean that people were not meaningfully
participating in the market before. There were plans available
that employers were free to sponsor, and employees were free
to seek, that did not require payment for contraceptive
coverage. And there are still, as this record demonstrates,
insurers who are ready, willing, and able to provide such
plans again, if the government did not forbid it. So, while it
is true that individual choice has been drastically reduced by

2
the federal government, that subtraction of freedom cannot be
a reason to say that government coercion of payment for
unwanted contraceptive products – indeed, to some people,
morally abhorrent products – is no burden on individuals.
The circularity of the government’s and the Majority’s
reasoning is stark.

I do not disagree with every aspect of my colleagues’
decision. The portion of the judgment that deals with the
Equal Protection and Administrative Procedures Act claims
of Real Alternatives, Inc. is sound. 1 I write separately,
however, to specify my disagreements with the Majority’s
treatment of the Religious Freedom Restoration Act
(“RFRA”) claim brought by Real Alternatives’ employees
Kevin I. Bagatta, Thomas A. Lang, and Clifford W.
McKeown (the “Individual Plaintiffs”). In my view, the
Individual Plaintiffs have adequately pled and provided
sufficient evidence to demonstrate that the Contraceptive
Mandate is a substantial burden on their free exercise of
religion. 2

1
I do not agree in full with the reasoning the Majority
employs for the APA claim, see infra footnote 4, but I do
agree that precedent requires the result.
2
In the District Court, the government asked either for
dismissal or summary judgment. The District Court accepted
and perpetuated that ambivalence, granting the government’s
motion to dismiss or in the alternative for summary judgment.
The Majority Opinion treats the District Court opinion as
solely granting summary judgment to the government. As I
see it, the government does not win either way; it loses either
way. See Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) (“To

3
 Having reached that conclusion, I confront the
question that the Majority avoids: whether the Contraceptive
Mandate is narrowly tailored to support a compelling
government interest. The answer is no. Time and again
courts have rejected the regulation because it is not the least
restrictive means of achieving its objective. There are several
other options the government could have chosen to enforce its
regulation without impinging on the rights of religiously
devout individuals. For that reason, I respectfully dissent.

I. Background

The Individual Plaintiffs are full-time employees of
Real Alternatives, 3 a non-profit organization devoted solely
“to promoting alternatives to abortion.” (Opening Br. at 2.)
All three men, their wives, and collectively seven minor
children, are covered by Real Alternatives’ health insurance
plan.

survive a motion to dismiss, a complaint must contain
sufficient factual matter, accepted as true, to ‘state a claim to
relief that is plausible on its face.’”) (quoting Bell Atlantic
Corp. v. Twombly, 550 U.S. 544, 570 (2007)); Fed. R. Civ. P.
56 (“The court shall grant summary judgment if the movant
shows that there is no genuine dispute as to any material fact
and the movant is entitled to judgment as a matter of law.”).
The Individual Plaintiffs’ RFRA claim should survive.

3
Bagatta serves as the President of Real Alternatives,
Lang is the Vice President of Operations, and McKeown is
the Vice President of Administration.

4
 In addition to dedicating their professional lives to
preventing abortion, the Individual Plaintiffs hold religious
beliefs that honor life from conception. It is undisputed that
all three men are devout in their respective religious faiths –
Bagatta and Lang are Catholics, and McKeown is an
Evangelical Christian. Among their sincerely held
convictions, “[e]ach of the employees and their families
believe that all human lives have full human dignity from the
moment of conception/fertilization.” (JA 99.) That is the
baseline, undisputed factual background upon which we are
obligated to proceed. The Individual Plaintiffs’ belief that
life begins at conception entails the further belief “that they
are prohibited from using, supporting, or otherwise
advocating abortifacient drugs and devices, including IUD
and any hormonal birth control method … .” (JA 99.)

The Contraceptive Mandate, promulgated under the
ACA, requires non-grandfathered group health care plans to
include coverage for the full range of FDA-approved
contraceptive methods, which encompasses diaphragms, oral
contraceptives, intrauterine devices, and drugs such as “Plan
B” and “Ella.” 4 (JA 6.) The latter two are sometimes called,

4
Grandfathered plans are defined as those that existed
prior to March 23, 2010. See Burwell v. Hobby Lobby Stores,
Inc., 134 S. Ct. 2751, 2764 (2014) (citing 42 U.S.C.
§§ 18011(a), (e)). They “need not comply with many of the
[ACA’s] requirements, including the [C]ontraceptive
[M]andate.” Id. As I indicated in dissent in Conestoga Wood
Specialties Corp. v. Secretary of United States Department of
Health & Human Services, 724 F.3d 377 (3d Cir. 2013)
(Jordan, J., dissenting), rev’d and remanded sub nom. Burwell
v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751 (2014), the

5
Mandate and related regulations were “not the product of any
legislative debate” or “even the result of work within an
administrative agency.” Id. at 391 n.2. They were drafted by
the Institute of Medicine, a private entity that, as a result of
the ACA’s complicated scheme “has ended up dictating
regulations that the government insists override[] the
[Individual Plaintiffs’] rights to religious liberty.” Id.
The Majority takes issue with whether the products
and services covered by the Contraceptive Mandate include
abortifacients. See (Maj. Op. at 31-32). While we may be
bound to accept the Department of Health and Human
Services’ definition of “abortifacients” for purposes of APA
review, see Auer v. Robbins, 519 U.S. 452, 461 (1997)
(holding – not without controversy – that courts must defer to
an agency’s interpretation of its own ambiguous regulation so
long as it is not “plainly erroneous or inconsistent with the
regulation” (citation omitted)), that is not true when we are
considering the burden imposed on the Individual Plaintiffs’
exercise of their religious beliefs. The Individual Plaintiffs
are persuaded that life begins at conception and that, by
definition, a drug or device that prevents implantation of a
fertilized ovum is an abortifacient. In other litigation, the
government has admitted that some items covered by the
Contraceptive Mandate can indeed prevent implantation.
Brief of Respondent United States in Opposition to Cert.,
Conestoga Wood Specialties Corp. v. Sebelius, 2013 WL
5740267 at *10 n.5 (filed October 21, 2013) (“Plan B, an
emergency contraceptive, is a pill that works mainly by
stopping the release of an egg from the ovary but may also
work by preventing fertilization of an egg or by preventing
attachment (implantation) to the womb (uterus). … Ella,
another emergency contraceptive, is a pill that works mainly

6
respectively, the “morning-after pill” and the “week-after
pill.” (Id.)

The Individual Plaintiffs currently elect to obtain their
health insurance through their employer, Real Alternatives.
Before the Mandate went into effect, Real Alternatives
bought an insurance plan for its employees that did not
contain contraceptive coverage. Because of the
Contraceptive Mandate, that plan is no longer available. If
the Individual Plaintiffs decide to decline coverage through
their employer, the government requires them to obtain it in
the open market, either independently or through “insurance
exchanges,” which are organizations created pursuant to the
ACA to facilitate the purchase of health insurance. All the
plans available on the open market – again because of the
Contraceptive Mandate – contain coverage for the
contraceptives. 5 In other words, the government has declared
that the Individual Plaintiffs must buy health insurance and,
simultaneously, has made it impossible for them to purchase
any coverage that conforms to their religious beliefs.

Enforcement of the Contraceptive Mandate, however,
is far from uniform. The government has granted a great

by stopping or delaying the ovaries from releasing an egg but
may also work by changing the lining of the womb (uterus)
that may prevent attachment (implantation).” (quotations and
citation omitted)).
5
According to the Verified Complaint, all available
plans “will include all contraceptives, including
abortifacients, and might also include surgical abortion.” (JA
114.)

7
many exceptions. See Conestoga Wood Specialties Corp. v.
Sec’y of U.S. Dep’t of Health & Human Servs., 724 F.3d 377,
413 (3d Cir. 2013) (Jordan, J., dissenting), rev’d and
remanded sub nom. Burwell v. Hobby Lobby Stores, Inc., 134
S. Ct. 2751 (2014) (“By its own choice, the government has
exempted an enormous number of employers from the
Mandate, including ‘religious employers’ who appear to share
the same religious objection as Conestoga and the Hahns,
leaving tens of millions of employees and their families
untouched by it.”). As the District Court observed, this
scheme of sporadic application has spawned “dozens of
lawsuits … challeng[ing] both the Contraceptive Mandate and
the dimensions of its exemptions.” (JA 11.) This is just the
latest episode.

8
II. Discussion 6

A. RFRA

RFRA was enacted “to provide very broad protection
for religious liberty.” Hobby Lobby, 134 S. Ct. at 2760. It
was passed in the wake of the Supreme Court’s decision in
Employment Division, Department of Human Resources of
Oregon v. Smith, 494 U.S. 872 (1990), which upended

6
As stated in the Majority Opinion, the District Court
had jurisdiction and so do we. See (Maj. Op. at 15 n.5). The
government conceded at oral argument that the Individual
Plaintiffs have standing to challenge the Mandate. See
http://www2.ca3.uscourts.gov/oralargument/audio/16-
1275RealAlternativesInc,etalv.SecretaryDeptofHealthandHu
manServices,etal..mp3, at 22:32 (argued November 3, 2016)
(counsel for the government recognizing that there “probably
was standing”). I agree. But for the Mandate, the Individual
Plaintiffs would be able to purchase a health plan that does
not include the contraceptives to which they object. Cf.
March for Life v. Burwell, 128 F. Supp. 3d 116, 123 n.6
(D.D.C. 2015) (“At the request of the Court, plaintiffs
submitted a letter received from March for Life’s insurance
carrier, CareFirst BlueCross BlueShield. The letter states that
CareFirst would be willing to offer March for Life or its
employees a plan omitting the contraceptive coverage that
they are objecting to [i]f a legal exemption from [the
Mandate] is obtained.” (alteration in original) (citation
omitted)). The Majority acknowledges this when they state
that “[t]he existence of an alternative plan is … relevant to
standing[.]” (Maj. Op. at 58. n.36.)

9
decades of precedent by “virtually eliminat[ing] the
requirement that the government justify burdens on religious
exercise.” Conestoga Wood, 724 F.3d at 407 (Jordan, J.,
dissenting) (quoting 42 U.S.C. § 2000bb(a)(4)). RFRA was
supported by an “extraordinary ecumenical coalition in the
Congress[,]” id., and has been hailed as “the most important
congressional action with respect to religion since the First
Congress proposed the First Amendment.” Id. (quoting
Douglas Laycock & Oliver S. Thomas, Interpreting the
Religious Freedom Restoration Act, 73 Tex. L. Rev. 209, 243
(1994)). 7

Most importantly for present purposes, RFRA restored
in religious liberty cases “the judicial standard of review
known as ‘strict scrutiny,’ which is ‘the most demanding test
known to constitutional law.’” Id. at 408 (quoting City of
Boerne v. Flores, 521 U.S. 507, 534 (1997)). According to
RFRA, the government is generally forbidden to
“substantially burden a person’s exercise of religion even if
the burden results from a rule of general applicability.” 42

7
The Majority focuses its telling of this history on
Smith’s rejection of a test that allowed broad protection for
religious liberty. See (Maj. Op. at 35 (quoting Smith, 494
U.S. at 888 for the proposition that a balancing test “would
open the prospect of constitutionally required religion
exemptions from civic obligations of almost every
conceivable kind”)). That misses the key point that RFRA
was passed for the very purpose of overruling Smith to the
fullest extent of Congress’s power. See 42 U.S.C.
§ 2000bb(b) (declaring that the purpose of RFRA was “to
restore the compelling interest test” from pre-Smith
jurisprudence).

10
U.S.C. § 2000bb-1(a). If the government does substantially
burden an individual’s exercise of religion, then that
individual is entitled to an exemption from the government
action, unless the government can show that the “application
of the burden to the person-- (1) is in furtherance of a
compelling government interest; and (2) is the least restrictive
means of furthering that compelling governmental interest.”
42 U.S.C. § 2000bb-1(b). No one disputes that the strict
scrutiny required by RFRA applies to the Contraceptive
Mandate, if the Mandate substantially burdens religious belief
and practice. 8

B. Substantial Burden

The Majority says that the Individual Plaintiffs have
not demonstrated a substantial burden on their religious
beliefs, but my colleagues reach that conclusion by a route
that amounts to questioning the validity of those beliefs – an
indulgence that we are forbidden. The Supreme Court has
made it clear that, when applying RFRA and analyzing a
burden on religion, our role is confined. “[F]ederal courts
have no business addressing … whether the religious belief
asserted in a RFRA case is reasonable … .” Hobby Lobby,
134 S. Ct. at 2778 (internal quotations omitted); cf. Smith,

8
RFRA as passed by Congress also applied to the
States, but, in City of Boerne v. Flores, 521 U.S. 507, 534
(1997), the Supreme Court held that the attempt to apply the
statute to the States exceeded Congress’s power under
Section Five of the Fourteenth Amendment. See Hobby
Lobby, 134 S. Ct. at 2761 (discussing City of Boerne, 521
U.S. at 533-34.)

11
494 U.S. at 887 (“Repeatedly and in many different contexts,
we have warned that courts must not presume to determine
the place of a particular belief in a religion or the plausibility
of a religious claim.”); Thomas v. Review Bd. of Ind. Emp’t
Sec. Div., 450 U.S. 707, 715 (1981) (recognizing that “it is
not for us to say that the line [a religious observer] drew was
an unreasonable one” and that courts cannot “dissect religious
beliefs”). Instead of weighing the reasonableness of deeply-
held religious convictions (and inevitably passing judgment
on their value), we have a “narrow function.” Hobby Lobby,
134 S. Ct. at 2779. We ask only “whether the line drawn [by
the adherent] reflects an honest conviction.” Id. (citation
omitted).

Once we have determined that an adherent has an
honest conviction, we ask if the government regulation
imposes a substantial burden on adherence to that conviction.
In this instance, we must decide “whether the [Contraceptive]
[M]andate imposes a substantial burden on the ability of the
objecting parties to [live] in accordance with their religious
beliefs[.]” Hobby Lobby, 134 S. Ct. at 2778. A “substantial
burden” exists where: (1) “a follower is forced to choose
between following the precepts of his religion and forfeiting
benefits otherwise generally available to other [persons]
versus abandoning one of the precepts of his religion in order
to receive a benefit”; or (2) “the government puts substantial
pressure on an adherent to substantially modify his behavior
and to violate his beliefs.” Washington v. Klem, 497 F.3d
272, 280 (3d Cir. 2007).

The Individual Plaintiffs attest in their Verified
Complaint that paying for insurance coverage for

12
contraception violates their religious beliefs. 9 See Verified
Complaint at ¶ 46 (JA 99-100) (“[T]he Real Alternatives
employees and their families object, on the basis of their
sincerely held ethical and religious beliefs, to participating in,
and/or paying a portion of the premium for, a health insurance
plan which provides coverage for objectionable items for
themselves and their family members.”). Because of the
Contraceptive Mandate, they are faced with two choices:
purchase a plan with the offending coverage (either through
their employer or on the exchanges) or decline to purchase a
plan, face a tax penalty, and leave their families uninsured.
See 26 U.S.C. § 5000A (codifying the ACA’s individual
mandate which requires individuals without employer-
coverage to purchase insurance or pay a penalty).
Notwithstanding the Majority’s protestations to the contrary,
that is a prime example of a substantial burden on religion. It
manages to satisfy both of the alternative tests for a
substantial burden: a believer is forced to choose whether to
follow the precepts of his religion and be penalized by the
government, or to abandon his convictions, 10 and the

9
A Verified Complaint is treated as an affidavit in the
summary judgment posture. See, e.g., Reese v. Sparks, 760
F.2d 64, 67 (3d Cir. 1985) (treating a verified complaint as an
affidavit for purposes of summary judgment).
10
Supreme Court jurisprudence demonstrates that the
imposition of a government penalty is at least as onerous as
the withholding of a government benefit. See Hobby Lobby,
134 S. Ct. at 2775 (noting that it had “little trouble
concluding” that forcing plaintiffs to choose between
honoring their religious convictions or facing severe
economic penalties was a substantial burden).

13
“government [thus] puts substantial pressure on [the follower]
to substantially modify his behavior and to violate his
beliefs.” Washington, 497 F.3d at 280.

The Supreme Court has long since declared that a
Hobson’s choice like the one forced upon the Individual
Plaintiffs is indeed a substantial burden on the exercise of
religion. See Thomas, 450 U.S. at 716 (“More tha[n] 30 years
ago, the Court held that a person may not be compelled to
choose between the exercise of a First Amendment right and
participation in an otherwise available public program.”).
And that principle remains in full force today. See Trinity
Lutheran Church of Columbia, Inc. v. Comer, 137 S. Ct.
2012, 2021-22 (2017) (finding a burden on religion under the
Free Exercise clause where a state statute required a church to
choose between “participat[ing] in an otherwise available
benefit program or remain[ing] a religious institution”). It
reflects an understanding that predates RFRA but rings
throughout the statute. See Sherbert v. Verner, 374 U.S. 398,
404 (1963) (finding a burden on a plaintiff where a state
unemployment law “force[d] her to choose between following
the precepts of her religion and forfeiting benefits, on the one
hand, and abandoning one of the precepts of her religion in
order to accept work, on the other hand”); Wisconsin v.
Yoder, 406 U.S. 205, 218 (1972) (finding a substantial burden
where a “Wisconsin [compulsory education] law
affirmatively compel[led] [the plaintiffs] under threat of
criminal sanction, to perform acts undeniably at odds with
fundamental tenets of their religious beliefs”). The avoidance
of such dilemmas is a key purpose of RFRA. See 42 U.S.C.
§ 2000bb(b) (stating that a purpose of RFRA is “to provide a
claim or defense to persons whose religious exercise is
substantially burdened by [the] government”).

14
 Hobby Lobby considered in depth whether the
Contraceptive Mandate imposed a substantial burden on
religiously devout persons who were being forced to make a
choice very like the one at issue here. 11 134 S. Ct. at 2775-
77. In that case, “family-run businesses” whose owners had
strongly-held religious beliefs against contraception were
forced to face severe economic fines if they chose to honor
their beliefs. 12 Id. The Supreme Court reasoned that

11
The Majority asserts that my saying the decisions
faced by the employers in Hobby Lobby and the employees
here are similar “misstates the applicability of Hobby
Lobby[.]” (Maj. Op. at 34 n.17.) Not so. The comparison is
apt because the claimants in Hobby Lobby and the Individual
Plaintiffs in this case were both forced by the United States to
take nearly-identical action: purchase of and participation in a
plan that covers a form of contraception that they believe is
antithetical to the sanctity of life.
12
The Supreme Court also rejected an argument made
by the government that the plaintiffs could avoid the
substantial burden by simply declining to provide health
insurance to their employees. See Hobby Lobby, 134 S. Ct. at
2777 (“We doubt that the Congress that enacted RFRA – or,
for that matter, ACA – would have believed it a tolerable
result to put family-run businesses to the choice of violating
their sincerely held religious beliefs or making all of their
employees lose their existing healthcare plans.”). Here, the
burden is likewise substantial because the Individual
Plaintiffs’ only alternative to purchasing the offending
insurance plans is to forego insurance and pay the associated
penalty.

15
“[b]ecause the Contraceptive Mandate forces them to pay an
enormous sum of money … if they insist on providing
insurance coverage in accordance with their religious beliefs,
the [M]andate clearly imposes a substantial burden on those
beliefs.” 13 Id. at 2779.

Two other courts have considered the precise question
before us today: whether the Mandate imposes a substantial
burden on the exercise of religious beliefs when individuals
are required to purchase insurance coverage through their
employer or on the open market, and all available plans
(because of government action) are required to contain
coverage at odds with those individuals’ faith. Both courts
held that the Contraceptive Mandate does, in that context,
impose a substantial burden on the exercise of religion. In
March for Life v. Burwell, 128 F. Supp. 3d 116, 130 (D.D.C.
2015), the court said “[e]mployee plaintiffs are … caught
between the proverbial rock and a hard place: they can either
buy into and participate in a health insurance plan that
includes the coverage they find objectionable and thereby
violate their religious beliefs, or they can forgo health
insurance altogether and thereby subject themselves to
penalties for violating the ACA’s individual mandate.”
Similarly, in Wieland v. United States Department of Health
& Human Services, 196 F. Supp. 3d 1010, 1017 (E.D. Mo.

13
The immediate financial cost to the employees is
less here, but not insignificant. See 26 U.S.C. § 5000A
(imposing a penalty of the higher of either 2.5% of household
income or $695/adult and $347.50/child, the latter capped at
$2,085). Of course, that cost does not account for the very
serious risk that must be absorbed if one is forced to go
without health insurance.

16
2016), the court observed that the Mandate’s “ultimate impact
is that Plaintiffs must either maintain a health insurance plan
that includes contraceptive coverage, in violation of their
sincerely-held religious beliefs, or they can forgo healthcare
altogether, which will result in the imposition of significant
penalties (not to mention the potentially crippling costs of
uninsured health care).”

The Majority here, though, sees things differently. It
claims that the Contraceptive Mandate cannot possibly
impose a substantial burden on anyone, relying on six general
arguments to bolster that conclusion. Those reasons,
however, look like nothing more than a rejection of where the
Individual Plaintiffs’ consciences have led them to draw the
line against being complicit in what their religions tell them is
wrong. It is the legitimacy of their conscientious religious
objections that my colleagues call into question, contrary to
the explicit direction of the Supreme Court. See Hobby
Lobby, 134 S.Ct. at 2778 (refusing to delve into “difficult and
important question[s] of religion and moral philosophy”).

1. The Precedential Effect of Geneva
College

First, the Majority relies on the now-vacated decision
of our court in Geneva College v. Secretary, United States
Department of Health & Human Services, 778 F.3d 422 (3d
Cir. 2015), vacated and remanded sub nom. Zubik v. Burwell,
136 S. Ct. 1557 (2016), to emphasize that courts can, and
should, assess the substantiality of a claimant’s asserted
burden. In that case, a panel considered the religious
exemption to the Mandate and determined that requiring non-
profit religious employers to register their objection to the

17
Contraceptive Mandate by filling out a form was not a
substantial burden under RFRA. See id. at 442 (“Because we
find that the self-certification procedure does not cause or
trigger the provision of contraceptive coverage, appellees are
unable to show that their religious exercise is burdened.”).
That opinion was deprived of any precedential effect by the
Supreme Court’s decision in Zubik v. Burwell, 136 S. Ct at
1561. Nevertheless, the Majority contends that Geneva is
persuasive and was not vacated because it was incorrect.
(Maj. Op. at 37 n.18.) I have my doubts about Geneva’s
reasoning, 14 but no doubt that it is not controlling. 15

14
The opinion in Geneva reflects, I think, an admirable
effort to explain why the form-filling exercise should not give
the faithful concern that they are complicit in actions contrary
to their religion. But there is a different and persuasive
discussion in the dissent from the order denying rehearing en
banc review in the Tenth Circuit in Little Sisters of the Poor
Home for the Aged v. Burwell, 799 F.3d 1315, 1317 (10th Cir.
2015) (Hartz, J., dissenting) (“When a law demands that a
person do something the person considers sinful, and the
penalty for refusal is a large financial penalty, then the law
imposes a substantial burden on that person’s free exercise of
religion. All the plaintiffs in this case sincerely believe that
they will be violating God’s law if they execute the
documents required by the government. And the penalty for
refusal to execute the documents may be in the millions of
dollars. How can it be any clearer that the law substantially
burdens the plaintiffs’ free exercise of religion?”).
15
In claiming that I mischaracterize their argument,
my colleagues agree that Geneva is not controlling and lacks
precedential force. (Maj. Op. at 37 n.18.) (“Geneva is no

18
longer controlling[.]”). But the Majority claims that “Zubik
vacated our judgment in Geneva but did not attack the
reasoning” and suggests that the Supreme Court’s vacatur has
no impact on “the view of our Court” as set forth in that case.
(Id.) It is inaccurate to claim that a vacatur has no effect on
the strength of an opinion – indeed, we have repeatedly
emphasized in our case law that, when an opinion is vacated,
“it carries no precedential force.” 1621 Route 22 W.
Operating Co., LLC v. Nat’l Labor Relations Bd., 825 F.3d
128, 141 n.6 (3d Cir. 2016); see also Leader v. Apex Hosiery
Co., 108 F.2d 71, 81 (3d Cir. 1939) (holding that a decree
considered to be vacated “is no longer binding as a precedent,
as the law of the case, or as res judicata”). Other Circuits are
in general agreement on this point. See, e.g., Durning v.
Citibank, N.A., 950 F.2d 1419, 1424 (9th Cir. 1991) (rejecting
an argument that a decision still had precedential value
because it was vacated on alternative grounds because while
“[a] decision may be reversed on other grounds … a decision
that has been vacated has no precedential authority
whatsoever”). The Majority cites no case law for the
extraordinary proposition that an appellate court’s reasoning
and judgment, after it has been vacated by the Supreme
Court, should carry weight in future cases, especially when
applied to litigants who were not parties to the original
dispute.
Geneva’s holding was vacated after the Supreme Court
received supplemental briefing indicating that the government
and non-profit religious employers could potentially reach a
compromise position that did not infringe on the rights of the
latter. See Zubik, 136 S. Ct at 1560 (“Given the gravity of the
dispute and the substantial clarification and refinement in the
positions of the parties, the parties on remand should be

19
 But even if Geneva were binding or persuasive
precedent, it does not lead to the result the Majority reaches
in this case. There are significant factual differences between
the burdens alleged in Geneva and those at issue here.
Notably, the panel in Geneva reasoned that the claimed
burden – the requirement to fill out and file a form – was
actually a means to register and affirm the employer’s
objection to providing contraceptive coverage. See Geneva,
778 F.3d at 438-39 (“If anything, because the appellees
specifically state on the self-certification form that they object
on religious grounds to providing such coverage, it is a
declaration that they will not be complicit in providing
coverage.”). According to Geneva, filling out the form was,
in effect, the organization’s chance to “wash[] its hands of
any involvement in contraceptive coverage[,]” leaving it to
“the insurer and the third party administrator [to] tak[e] up the
slack under compulsion of federal law.” Id. at 441 (internal
quotation omitted). 16 Here, the Individual Plaintiffs are

afforded an opportunity to arrive at an approach going
forward that accommodates petitioners’ religious exercise
while at the same time ensuring that women covered by
petitioners’ health plans ‘receive full and equal coverage,
including contraceptive coverage.’” (citation omitted)). Thus,
the Supreme Court did not in any way endorse the conclusion
that the college did not face a substantial burden under
RFRA. See id. (“[T]he Court does not decide whether
petitioners’ religious exercise has been substantially burdened
… .”)
16
The response to that reasoning, of course, is that
saying something does not make it so. If the government says
“file this paperwork so that we can give abortifacients to your

20
compelled to do much more than fill out and file a form. Far
from distancing themselves from the objectionable coverage,
the Individual Plaintiffs are forced to sign up and pay for it,
unless they want themselves and their families to be
uninsured and to pay fines. They must actually provide
financial support for the objectionable contraceptive
coverage, just like the plaintiffs in Hobby Lobby. The
Majority does nothing to address that distinction between
Geneva and this case.

2. Using, Supporting, or Advocating the
Use of Contraceptives

The Majority next turns to the words of the Individual
Plaintiffs’ Complaint, where they object to “using,
supporting, or otherwise advocating, the use of abortifacients,
or participating in a health insurance plan that covers such
items for themselves or their families.” (Verified Compl.
¶ 158.) 17 The Majority first claims that signing up for an

employees,” it may not help to add “and don’t worry, you will
not be complicit in what’s going to happen as soon as you file
that paperwork.”
17
The Majority ignores this statement of the Individual
Plaintiffs in their Verified Complaint defining their burden:
“the Real Alternatives employees and their families object, on
the basis of their sincerely held ethical and religious beliefs,
to participating in, and/or paying a portion of the premium
for, a health insurance plan which provides coverage for
objectionable items for themselves and their family
members.” See Verified Complaint at ¶ 46. That particular
iteration of the burden focuses on the financial aspect of

21
insurance plan that covers contraceptive coverage does not
involve the “use, support, or advocacy of contraceptives”
because “[c]hecking off a box to be eligible for
reimbursement of services … of the employee’s choosing …
in no way indicates, let alone suggests, support or advocacy
for that service.” (Maj. Op. at 44.) That conclusion relies on
the Majority’s perception of how insurance coverage works:

The plan deems the employee eligible to be
reimbursed for hundreds of different services,
and that employee can take advantage of that
eligibility as he or she sees fit. Should the
employee opt to use a particular service, he or
she fills out a form and asks to be paid back for
costs incurred. In the end, the employee uses a
covered service, or not; either way, there is no
requirement to support or advocate for whatever
service he or she, or others, selects.

(Id.)

As my colleagues see it, because the Individual
Plaintiffs can elect not to use the covered contraceptives, they
are not burdened by having to pay for the coverage. The
message is “get over it.” And that seems to me to be only a
“thinly-veiled attack” on sincerely-held religious beliefs.

paying into a plan that supports contraceptive services. The
Majority has completely ignored the financial consequences
of the Mandate, so it is perhaps not surprising that they have
chosen to attend solely to a portion of the Complaint that does
not mention those consequences.

22
March for Life, 128 F. Supp. 3d at 129. When the Individual
Plaintiffs say in their Verified Complaint that it is at odds
with their religious beliefs to purchase a plan which uses their
money to offer products and services they believe to be
morally abhorrent, I think we are supposed to believe them.

And we should, because their concern that their money
will be used to support contraceptives is perfectly logical. It
is the Majority’s characterization of how the insurance market
functions that is confused. It overlooks two essential truths:
money is fungible and insurance is based on the pooling of
risk. Cf. Nat’l Fed’n of Indep. Bus. v. Sebelius, 132 S. Ct.
2566, 2585 (2012) (The requirement that everyone must
purchase health insurance “forces into the insurance risk pool
more healthy individuals, whose premiums on average will be
higher than their health care expenses. This allows insurers to
subsidize the costs of covering the unhealthy individuals the
[ACA] reforms require them to accept.”). In the
government’s own words, the system “works through ‘risk
pooling in the group market’ which ‘results in sharing …
costs … across an entire plan or employee group.’”
(Responding Br. at 23-24 (quoting 75 Fed. Reg. at 41,730).)
Thus even when the Individual Plaintiffs elect not to use
contraceptive coverage, they still pay for and thus support it.
See id. at 24 (noting that the plans “cover a wide array of
services … [and insurers] set rates based on standardized
policies [that] ensure[] that medical costs are spread across
the entire pool of plan beneficiaries”). It is peculiar, then, for
the Majority to claim that purchasing an insurance plan that
includes contraception “does not assure the availability of
specific services.” (Maj. Op. at 45.) While an individual
must seek out and use a particular service, the point of health

23
insurance is in fact to help facilitate and support access to
each service for everyone in the risk pool.

Taken to its logical conclusion, my colleagues’
position means that the Contraceptive Mandate could only be
a “substantial burden” on the exercise of religion if the
government forced religious objectors not only to buy plans
with contraceptive coverage, but also to buy the covered
contraceptives. That idea was rejected by the Supreme Court
in Hobby Lobby, when it determined that providing coverage
to employees, who may or may not elect to use the
contraceptive coverage, was a substantial burden on the
exercise of religion. 18 See Hobby Lobby, 134 S. Ct. at 2778
(“The Hahns and Greens believe that providing the coverage
demanded by the HHS regulations is connected to the
destruction of an embryo in a way that is sufficient to make it
immoral for them to provide the coverage.”). The Majority
here may prefer the position taken by the dissent in Hobby
Lobby, see 134 S. Ct. at 2799 (Ginsburg, J., dissenting) (“I

18
The Majority makes an artificial distinction and says
that while the Contraceptive Mandate “requires nothing of the
employees that implicates their religious beliefs” the Mandate
did affect the employers in Hobby Lobby because it “literally
required” them to “‘arrange for’ contraceptive coverage in a
way that effectively amounted to sponsorship.” (Maj. Op. at
51 (quoting Hobby Lobby, 134 S. Ct. at 2775)). That
purported difference is meaningless – in both Hobby Lobby
and here the claimants were forced to financially support
others’ use of contraceptives, an action that was antithetical to
their religious beliefs. As in Hobby Lobby, “it is not for us to
say [whether] the[] religious beliefs are mistaken or
insubstantial.” 134 S. Ct. at 2778.

24
would conclude that the connection between the families’
religious objections and the contraceptive coverage
requirement is too attenuated to rank as substantial. The
requirement carries no command that Hobby Lobby or
Conestoga purchase or provide the contraceptives they find
objectionable.”), but that was the losing argument, as it
should have been.

3. Participating in a Plan
Containing Contraceptives

The Majority next focuses its attention on the
Individual Plaintiffs’ claim that participating in a health
insurance plan containing coverage for contraceptives is a
substantial burden on their free exercise. In doing so, my
colleagues reduce the Individual Plaintiffs to non-entities in
the calculus of harm. The Majority’s argument is that the
Individual Plaintiffs do not meaningfully “participate” in the
acquisition of their health insurance coverage, so the market
regulation forcing all but exempt plans to carry contraceptive
coverage cannot be a substantial burden on them. There is,
however, no sound legal or logical foundation for that
position.

To begin, the Majority claims that there is no “active
‘participation’” by an individual in subscribing to an
insurance plan. (Maj. Op. at 45.) Their argument is that the
concepts of “buy into” and “participate in” are not
“interchangeable[,]” therefore when individuals purchase an
insurance plan, they do not participate in it. (Id.) This is a
semantic distinction without difference. And even assuming
that the “active” participation requirement had a basis in our
case law (which it does not), being an insurance plan

25
participant should fit the bill. “[H]ealth insurance does not
exist independently of the people who purchase it,” March for
Life, 128 F. Supp. 3d at 129, and the purchasers are not
designated by the insurers as “plan participants” for nothing.
As already explained, the Individual Plaintiffs are not simply
paying for the services they elect to use; they are participants
in a plan that pools risk and provides comprehensive
coverage. They are paying for all services, even those they
individually decline to use, thus their participation in the plan
directly subsidizes the use of contraceptives. That is hardly
“remote facilitation.” (Maj. Op. at 48) (citation omitted).
What’s more, the Majority completely ignores that
“participation” in the insurance market is compelled – and
enforced with a significant monetary penalty. 19 Being
required to associate with and subsidize an organization or
activity that one disagrees with does indeed impose a
substantial burden on religion.

The Majority also makes a nearly identical argument
using a slightly different term – directness. But, deploying a
synonym does not improve the argument that the Individual
Plaintiffs can be ignored as playing no “active ‘role’” in their

19
By ignoring the fact that the Individual Plaintiffs are
forced to buy health insurance, the Majority attempts to make
the central question whether or not a health insurance
purchaser meaningfully “participates” in their insurance plan.
But that is not the question RFRA asks. The proper inquiry is
whether government action substantially burdens religion.
Where the “participation” is abhorrent to the claimants’
religion – and it is compelled – it should be plain that their
religious exercise is substantially burdened.

26
health plans. (Maj. Op. at 45.) Even if the Individual
Plaintiffs’ burden or participation could rightly be
characterized as “indirect” in some way, nothing requires a
burden to be “direct” to be cognizable under RFRA. See
Lyng v. Nw. Indian Cemetery Protective Ass’n, 485 U.S. 439,
450 (1988) (“It is true that this Court has repeatedly held that
indirect coercion or penalties on the free exercise of religion,
not just outright prohibitions, are subject to scrutiny under the
First Amendment.”). 20

20
The Majority chides me for not including a fuller
quotation in the parenthetical to this citation to Lyng v.
Northwest Indian Cemetery Protective Association, 485 U.S.
439 (1988). (Maj. Op. at 49-50 n.29.) But the Supreme
Court itself, in Trinity Lutheran Church of Columbia, Inc. v.
Comer, recently relied on Lyng for that precise principle,
noting that “the Free Exercise clause protects against ‘indirect
coercion or penalties on the free exercise of religion, not just
outright prohibitions.’” 137 S. Ct. 2012, 2022 (2017)
(quoting Lyng, 485 U.S. at 450).
Also, according to my colleagues, the March for Life
opinion that they spend so much time belittling actually
supports them on the significance of “direct vs. indirect”
burdens. But in the quotation the Majority borrows, the
March for Life opinion simply acknowledged the fact that the
Contraceptive Mandate regulated insurers in the first instance.
See March for Life, 128 F. Supp. 3d at 129 (noting that
individuals are not “the direct objects” of the Contraceptive
Mandate). That does not mean that purchasers are unaffected.
In reality, the labels “direct” and “indirect” are too malleable
to be of any real use in this context. By regulating the types
of plans insurance companies can offer, and then forcing
individuals to purchase those plans, the government is, in a

27
 My colleagues also draw a number of analogies in an
effort to demonstrate why this case does not involve a
substantial burden. None ring true. They hypothesize
someone alleging a substantial burden when subscribing to a
magazine, or joining an organization, or acquiring a credit
card. According to the Majority, “there is no active
‘participation’” in any of these scenarios because “[t]hese are
all packages that involve a one-time enrollment, followed by
essentially passive eligibility for certain services that the
member opts in or out of.” (Maj. Op. at 45.) My colleagues
again completely ignore that purchasing health insurance in
the era of the ACA is far from voluntary – it is compelled and
enforced with a monetary fine. Their examples are therefore
meaningless. Of course subscribing to a magazine would be a
substantial burden under RFRA if it were abhorrent to the
subscriber’s religious beliefs and were forced upon him by
the government. 21 The same is true for compelled
membership in an organization.

very real sense, directly acting on individuals. Moreover, the
Majority has ignored the rest of the discussion in March for
Life, which lays out why the Mandate is a substantial burden.
See id. (“While it is true that an asserted burden is also not an
actionable substantial burden when it falls on a third party,
not the religious adherent … health insurance does not exist
independently of the people who purchase it.” (internal
citations and quotations omitted)).
21
Jefferson was right: “to compel a man to furnish
contributions of money for the propagation of opinions which
he disbelieves, is sinful and tyrannical.” Abood v. Detroit Bd.
of Ed., 431 U.S. 209, 235 n.31 (1977) (quoting I. Brant,
James Madison: The Nationalist 354 (1948)).

28
 The credit card and banking analogy fares no better.
My colleagues say that, as in the insurance market,
“accountholders [at a bank] have no say in lending decisions
(what rates to charge, which borrowers to lend to) and no
direct control over the bank.” (Maj. Op. at 46 n.25.)
Accordingly, if we were to “[a]ssume that the individual’s
bank account is mandated by the Government under a
privatized Social Security regime” and “an accountholder had
a religious objection to the bank’s practices” such as “lending
money at interest[,]” that accountholder could not
“successfully vindicate his or her religious beliefs through
RFRA.” (Id.) I disagree. In that hypothetical, it would
undoubtedly impose a substantial burden on religion to force
such believers to put their money into an interest bearing
account contrary to their religious beliefs. 22 The
consequence of determining that there was a substantial
burden would not be to prevent the government from
instituting a privatized Social Security regime. The
consequence would be to force the government to satisfy
strict scrutiny before forcing the religious objector to

22
Those who desire to follow prohibitions on usury in
the Torah and Quran often avoid traditional banking or enter
into alternative arrangements with banks. See, e.g., Naureen
S. Malik, Interest-Free Financing for U.S. Muslims, ABC
News http://abcnews.go.com/Business/story?id=87070
(explaining that there are two relevant Islamic prohibitions,
“[o]ne against the use of ribaa or ribit, also known as usury;
and the other against gharar, the unbundled sale of risk, such
as gambling, insurance or derivatives[,]” and noting that
banks will offer alternative arrangements to comply with
those prohibitions).

29
participate. 23 See United States v. Lee, 455 U.S. 252, 260-61
(1982) (analyzing whether Social Security satisfied strict
scrutiny with respect to an Amish objector).

23
The Majority’s analogies are troubling not only
under RFRA, but also under the Constitution. See W.
Virginia State Bd. of Educ. v. Barnette, 319 U.S. 624, 642
(1943) (“If there is any fixed star in our constitutional
constellation, it is that no official, high or petty, can prescribe
what shall be orthodox in politics, nationalism, religion, or
other matters of opinion or force citizens to confess by word
or act their faith therein.”). The proposals suggested by the
Majority would violate the First Amendment in more ways
than one. In the association context, the Supreme Court has
repeatedly found that being forced to subsidize and affiliate
with an organization one disagrees with clearly burdens
freedom of association and expression, even when the cost of
membership is de minimis and there is no additional
requirement to participate. See, e.g., Harris v. Quinn, 134 S.
Ct. 2618, 2639, 189 L. Ed. 2d 620 (2014) (striking down
compelled labor union membership requirements); Keller v.
State Bar of California, 496 U.S. 1, 6, (1990) (finding
mandatory bar dues that were used for ideological or political
educational programs violated the First Amendment).
Likewise, free speech protections prevent the government
from compelling an individual to subsidize or facilitate
expression of speech that one disagrees with. See, e.g.,
Miami Herald Pub. Co. v. Tornillo, 418 U.S. 241 (1974)
(holding that the government cannot require a newspaper to
provide space for the expression of certain viewpoints); Pac.
Gas & Elec. Co. v. Pub. Utilities Comm’n of California, 475
U.S. 1 (1986) (rejecting a law that forced a power company to

30
 In the end, the Majority’s claim that the Individual
Employees do not meaningfully participate in their health
care plans cannot be saved by the hypotheticals on which they
rely. Each ignores that the government has coerced the
Individual Plaintiffs to purchase health insurance with
provisions deeply offensive to their sincerely held religious
beliefs. The hypotheticals thus serve only to underscore the
weakness of the Majority’s argument.

4. “Incidental” Effects and Lyng

To bolster its arguments regarding “direct” and
“active” participation, the Majority tries to link this case and
Geneva to Lyng v. Northwest Indian Cemetery Protective
Association. In that case the Supreme Court concluded that
the government’s building of a road on public land used for
religious purposes by Native Americans was not a violation
of their right to Free Exercise. 485 U.S. at 447-53. My
colleagues also make passing reference (Maj. Op. at 40) to
Bowen v. Roy, 476 U.S. 693, 699-700 (1986), a Supreme
Court case relied upon in Lyng which held that requiring the
use of social security numbers to participate in federal food
stamp and aid programs was not a significant burden on
religious beliefs. See Bowen, 476 U.S. at 699-700.

Bowen and Lyng are distinguishable. Both cases
recognized the difference between challenges to “certain
forms of governmental compulsion” and policies that
amounted to the “Government’s internal procedures.”

allow public interest groups to share a message on its billing
envelopes).

31
Bowen, 476 U.S. at 700. In Bowen, because the assignment
of a social security number did not require the religious
objectors to do anything, the Court found that the law fell into
the latter category. Id. (recognizing that religious protections
under the Free Exercise clause extend to “what the
government cannot do to the individual, not in terms of what
the individual can extract from the government”) (quotation
omitted). And in Lyng, the same was true: the claimants
sought to stop government action that affected their religious
practice (i.e. building a road through their lands), but did not
compel their own behavior. Lyng, 485 U.S. at 449 (“In both
cases, the challenged Government action would interfere
significantly with private persons’ ability to pursue spiritual
fulfillment according to their own religious beliefs. In neither
case, however, would the affected individuals be coerced by
the Government’s action into violating their religious beliefs;
nor would either governmental action penalize religious
activity by denying any person an equal share of the rights,
benefits, and privileges enjoyed by other citizens.”). By
contrast, the ACA forces the Individual Plaintiffs to engage in
certain behavior – purchasing health insurance – and enforces
that compulsion with the threat of a significant fine.

My colleagues, however, fixate on the Supreme
Court’s observation in Lyng that an incidental effect of a
government program with “no tendency to coerce individuals
into acting contrary to their religious beliefs” is not violative
of the First Amendment. (Maj. Op. at 49 n.29.) (quoting
Lyng, 485 U.S. at 450-51). Of course, I do not disagree with
that limitation. But the Majority is fighting a losing battle
with common sense when it argues that imposing financial
penalties on individuals who fail to take action that violates
their religion has “no tendency to coerce.” The Mandate

32
coerces the Individual Plaintiffs into violating their beliefs by
forcing them to purchase a health care plan at odds with their
religious convictions. See Verified Complaint at ¶ 46 (JA 99-
100) (“[T]he Real Alternatives employees and their families
object, on the basis of their sincerely held … religious beliefs,
to participating in, and/or paying a portion of the premium
for, a health insurance plan which provides coverage for
objectionable items for themselves and their family
members.”). Thus, the Individual Plaintiffs here are not
challenging the way the government “conduct[s] its own
internal affairs,” (Maj. Op. at 40 (quoting Bowen, 476 U.S. at
699)); they are challenging what a government regulation
requires them to do. 24 As the Supreme Court’s recent holding

24
The Majority’s collection of out-of-Circuit cases is
also not persuasive. I will spare the reader an extensive
response to each and every one of the cases cited by the
Majority in its lengthy footnote 33. Broadly speaking,
however, the cases cited are simply inapplicable or
distinguishable. For instance, some of the cases involve
situations where the government offered an accommodation
for religious belief. See, e.g., United States v. Friday, 525
F.3d 938, 947-48 (10th Cir. 2008) (upholding a law creating a
general prohibition on hunting bald eagles but allowing
Native Americans to apply for a permit if they need to hunt
for religious reasons); Berman v. Bd. of Elections, 420 F.2d
684, 685–86 (2d Cir. 1969) (concluding a case was moot
where the municipality permitted an individual to vote in
another polling place when voting in a Church violated his
religious beliefs). Others recognized that claimants could
exercise their religion, without government penalty, in closely
analogous circumstances that did not impose a burden. See,
e.g., Henderson v. Kennedy, 253 F.3d 12, 17 (D.C. Cir. 2001)

33
in Trinity Lutheran demonstrates, laws that coerce religious
claimants to disavow their religion in order to receive a
government benefits are inconsistent with our constitutional
traditions. Cf. Trinity Lutheran, 137 S. Ct. at 2022 (finding
“express discrimination” under the First Amendment where a
church was denied the opportunity to compete for a
government benefit “solely because it is a church). 25

(upholding a law preventing sale of t-shirts on the National
Mall against a RFRA challenge because the religious
individuals seeking to sell their shirts on the Mall had not
shown why selling on the Mall rather than a few blocks away
was required by their religious beliefs); Lakewood, Ohio
Congregation of Jehovah’s Witnesses, Inc., v. City of
Lakewood, 699 F.2d 303, 307–08 (6th Cir. 1983) (upholding
a zoning ordinance that limited locations where churches
could be built where there was no suggestion that building in
the approved zone would impose a prohibitive cost or
interfere with the religious mission of the Church). Most
fundamentally, none of these cases involved challenges to
government action that forced individuals to act contrary to
their religious beliefs under threat of government fine.
25
The Majority says that Trinity Lutheran is not
relevant because it is “not a RFRA case[.]” (Maj. Op. at 41 n.
29.) But First Amendment cases based on the Free Exercise
clause certainly are relevant to understanding the meaning
and application of RFRA. See Hobby Lobby, 134 S. Ct. at
2778-79 (relying on pre-RFRA cases to analyze a substantial
burden under RFRA); (Maj. Op. at 43 (relying on Lyng, a pre-
RFRA case)). And whether the Supreme Court was
“[s]ignaling its intent to confine its holding” (Maj. Op. at 50
n.29) in Trinity Lutheran with a footnote is far from clear.

34
 The Majority also dresses up the “incidental” point in
different language and says that RFRA bars claims arising out
of burdens on third parties. The faulty logic is that, because
the Contraceptive Mandate only regulates the insurer and not
the Individual Plaintiffs, it cannot be a substantial burden
under RFRA. See (Maj. Op. at 55 (“The Supreme Court has
consistently rejected the argument that an independent
obligation on a third party can impose a substantial burden on
the exercise of religion in violation of RFRA.” (quoting
Geneva, 778 F.3d at 440-41))). But the Individual Plaintiffs
do not object to insurance companies offering plans with
contraceptive coverage, making the cases the Majority relies
on about third parties irrelevant. The Individual Plaintiffs are
only asking the government to allow them to purchase a plan
that does not include the offending coverage – a request that,
according to this record, would not impose any harm or
burden on any third parties. There evidently are insurers
prepared to fill that market demand, just as there were before
the ACA told all insurers that they had to eliminate that
choice.

5. Opening the Floodgates

The Majority also relies on a floodgates argument to
hold that the Individual Plaintiffs have not experienced a

See Trinity Lutheran, 137 S. Ct. at 2026 (Gorsuch, J.,
concurring) (voicing concern that courts would “mistakenly
read” that footnote to narrow the scope of the court’s holding
and pointing out that doing so was “unreasonable” because
cases are “governed by general principles, rather than ad hoc
improvisations.” (citation omitted)).

35
substantial burden on their free exercise of religion. My
colleagues worry that allowing the Individual Plaintiffs to
maintain a RFRA claim would open the way to myriad
challenges to the ACA, because “the categories of services
that could offend religious beliefs [are] wide-ranging.” (Maj.
Op. at 52.) Thus, “denying … such services to all on the
basis of the religious objections of some would be neither
desirable nor administrable.” Id.

Of course, that fails to address the burden issue at all.
It is merely an assertion that, regardless of the burden on
religious belief, it could be difficult for the government to do
what it wants if any accommodation for religious believers
must be made. Sadly, this argument is “the classic rejoinder
of bureaucrats throughout history: If I make an exception for
you, I’ll have to make one for everybody, so no exceptions.”
Gonzales v. O Centro Espirita Beneficente Uniao do Vegetal,
546 U.S. 418, 436 (2006). It is also the reasoning of the
dissent in Hobby Lobby, which worried that allowing a RFRA
challenge to one part of the ACA “would open the prospect of
constitutionally required religious exemptions from civic
obligations of almost every conceivable kind.” 134 S. Ct. at
2785 (quoting Smith, 494 U.S. at 888-89). But the Majority
in Hobby Lobby rightly rejected that hyperbolic concern,
recognizing that the judiciary is bound to apply the balancing
test set forth by RFRA to adjudicate such claims. See id.
(“But Congress, in enacting RFRA, took the position that ‘the
compelling interest test as set forth in prior Federal court
rulings is a workable test for striking sensible balances
between religious liberty and competing prior governmental
interests.’” (quoting 42 U.S.C. § 2000bb(a)(5))). The
command “to enforce RFRA as written[,]” id., requires us to
avoid imagining a speculative “parade of horribles” as a

36
counterweight to the real burden on real people. March for
Life, 128 F. Supp. 3d at 132; see also Gonzales, 546 U.S. at
434 (“RFRA, however, plainly contemplates that courts
would recognize exceptions—that is how the law works.”
(citing 42 U.S.C. § 2000bb-1(c))). 26

Nevertheless, because the Majority has cited concern
for the insurance markets as a reason to walk away from
RFRA, it bears emphasis that there is a simple answer to that
concern. 27 It was given by Judge Richard Leon of the United

26
The Majority criticizes my reliance on Gonzales
because that opinion does not address the Majority’s
“concerns regarding the end-run on legislation” that would be
“unleash[ed]” by adjudicating the Individual Plaintiffs’
RFRA claim. (Maj. Op. at 57 n.34.) But courts are bound to
adjudicate the substantial burden inquiry under RFRA based
on the facts before them. Cf. Morrow v. Balaski, 719 F.3d
160, 201 (3d Cir. 2013), as amended (June 14, 2013)
(Fuentes, J., dissenting) (“[W]e ought not refuse to grant
relief that is warranted simply to stem future litigation.”).
And Gonzales did address an argument that the
“effectiveness” of the regulation at issue would be
“necessarily undercut” by granting an exception. 546 U.S. at
434. The Court rejected that speculation. See id. at 435
(finding that there was “no evidence” that allowing a RFRA
exemption for claimants would “undercut the Government’s
ability to enforce” the law with respect to non-claimants).
27
My colleagues claim that the existence of an
alternative plan is only relevant to “standing and questions of
redressability” (Maj. Op. at 58 n.36), and yet they emphasize
concerns about the workability of the insurance market in

37
States District Court for the District of Columbia in his
thoughtful rebuttal of the “parade of horribles” argument in
the March for Life case. He demonstrated that the Majority’s
argument has no real weight because “[i]nsurance companies
have every incentive to maintain a sustainable and
functioning market … .” 28 128 F. Supp. 3d at 132. Thus,

their substantial burden analysis. See (Id. at 56-57 (worrying
that “a finding that coverage for one set of objectionable
services constitutes a substantial burden would imply that
coverage for all such services imposes a substantial burden –
an implication that would render the health care system
totally unworkable” (citations and internal quotation marks
omitted))). As ought to be clear from their own concerns, the
likely availability of alternative insurance plans is relevant to
the merits aspects of the case.
28
My colleagues try to rebut this point by dragging a
red herring across the trail: they argue that my position
means I am hostile to all regulation. See (Maj. Op. at 58
n.35) (characterizing my position as concluding that “any
regulation of any market is unnecessary”). Even if that were
true, and it most assuredly is not, it is irrelevant to the
discussion. To be clear, I am not arguing that all regulation is
devoid of value. I am simply stating that, if we conclude that
individual religious adherents are substantially burdened by
the regulation, granting them an exemption will not take
down the system. The fact that they will have to find an
insurer – one which is subject to market forces – to provide
them with their desired plan demonstrates that their request
will not unravel the system. If it did present that threat, no
insurer would offer such a plan. And we know that at least
one insurer is likely to offer such a plan. See March for Life,

38
“the government’s interest in the same would not be
undermined by simply making it legal for a third-party
provider to offer, without penalty, a plan consistent with
[Individual] [P]laintiffs’ religious beliefs.” Id. In the event
that “offering an insurance plan that does not include a
service or services to which a potential purchaser objects on
religious grounds would be ‘an impossible administrative

128 F. Supp. 3d at 123 n.6 (recognizing that an insurer was
willing to offer a contraception-free plan). What’s more, we
know that, before the ACA forbade markets to respond to
consumer demand, many insurers offered such plans and the
United States managed to have a functioning health insurance
market.
The seatbelt analogy the Majority offers is passing
strange but must, I suppose, have an answer. First,
automobile regulations recognize safety concerns for the
public generally – who knows who will ride in a vehicle; it
could be any number of people, and protecting them has been
deemed wise, so the government did not wait for market
forces to work. That safety concern is unlike anything related
to the Contraceptive Mandate and the insurance market.
Strangers do not get in and out of your policy as they can get
in and out of your car. Moreover, if one were to imagine an
anti-seatbelt religious sect (a thought exercise which seems to
demean the religious concerns actually at issue in this case),
there is no warrant for fearing that, if the government
permitted members of that sect to buy a car without seatbelts
or to remove the belts after buying the car, the U.S.
automobile industry would cease to function.

39
undertaking,’ insurance companies will not do it.” 29 Id.
When we leave to the insurance companies themselves the
decision of what coverage options they can profitably
provide, it is obvious that the “parade of horribles” will not
begin to march. See id. The market managed to provide
coverage options before the ACA and it is a good bet it can
do so again.

6. “Substantial” Burden

My colleagues repeatedly highlight that government
action must substantially burden religion in order to be
cognizable under RFRA, citing Geneva and legislative history
as proof that the weight of the burden is “the very essence of
a RFRA claim[.]” (Maj. Op. at 42.) They say that, even if
there is a burden on the Individual Plaintiffs, it is not enough
to be considered “substantial.” That is the comfortable
rationale. “No matter how sincerely held [the Individual
Plaintiffs’] beliefs may be, we cannot accept at face value that
subscribing to the plan imposes a ‘substantial burden.’” (Id.
at 59.) In articulating that conclusion, my colleagues

29
The Majority argues that I am operating on a false
premise because “[i]nsurance companies have an interest in a
sustainable and functioning insurance market only to the
extent that it is profitable for them.” (Maj. Op. at 57-58.) But
that is exactly my premise and it is not false. Long-term
profits are only realizable in a sustainable and functioning
market. If the cost of providing plans with carve-outs for
conscience threatens the viability of the insurance market,
such plans will not be offered, and they will not be offered
precisely because insurance companies are motivated by
profit.

40
recognize that their characterization is “a matter of
subjectivity,” as indeed it is. Id. at 48.

They are, of course, correct that the plain language of
RFRA forbids the government from “substantially
burden[ing]” a claimant’s exercise of religion. 42 U.S.C.
§ 2000bb-1(a); Cf. Paek v. Attorney Gen. of the U.S., 793
F.3d 330, 334 (3d Cir. 2015) (“[W]e do not resort to
legislative history to cloud a statutory text that is clear.”
(quoting Ratzlaf v. United States, 510 U.S. 135, 147-48
(1994))). But as I have already endeavored to show, the
compelled action here is indeed a substantial burden. It
seems to me that the disagreement we have in this case is not
fundamentally about the burden; it is about the underlying
belief.

The Majority claims that I have made the misstep of
“conflat[ing]” the duty to analyze whether a burden is
substantial with our obligation to accept the validity of a
claimant’s religious belief. (Maj. Op. at 43 n.24.) But, I have
addressed the two questions distinctly, see supra at 12 (“Once
we have determined that an adherent has an honest
conviction, we ask if the government regulation imposes a
substantial burden on adherence to that conviction.”),
knowing that caution is needed because an evaluation of the
substantiality of a burden can easily cross into the forbidden
territory of opining on the merits of a claimant’s beliefs. See
Hobby Lobby, 134 S. Ct. at 2777-78 (recognizing that
focusing on the closeness of “connection between what the
objecting parties must do … and the end they find to be
morally wrong” in reality “dodges the question that RFRA
presents … and instead addresses a very different question
that the federal courts have no business addressing”).

41
 It is the Majority’s approach that runs afoul of binding
precedent, 30 and my colleagues’ rejection of the deference we

30
In support of its interpretation of the substantiality
requirement, the Majority repeatedly cites to the dissenting
opinion in Hobby Lobby, see (Maj. Op. at 46-47; 51 n.30), as
well as secondary sources arguing against the majority
opinion in Hobby Lobby, see, e.g., Frederick Mark Gedicks,
“Substantial” Burdens: How Courts May (and Why They
Must) Judge Burdens on Religion Under RFRA, 85 Geo.
Wash. L. Rev. 94, 101 (2017) (lauding the dissent in Hobby
Lobby for “question[ing]” what it characterizes as “a doctrinal
regime that renders RFRA’s substantial burden element
functionally nonjusticiable”); Matthew A. Melone,
Corporations and Religious Freedom: Hobby Lobby
Stores—A Missed Opportunity to Reconcile a Flawed Law
with a Flawed Health Care System, 48 Ind. L. Rev. 461, 503
(2015) (taking the position that “an imposition on
conscience” that arises from the Mandate “is not a burden on
exercise at all”). Those writings may make for interesting
reading but they are not the law, no matter how earnestly the
Majority wishes they were. Legal academics are free to
disregard Supreme Court precedent, but we are not. On many
difficult issues, including this one, there are law review
articles with varying perspectives, compare Gedicks, supra,
with Scott W. Gaylord, RFRA Rights Revisited: Substantial
Burdens, Judicial Competence, and the Religious Nonprofit
Cases, 81 MO. L. REV. 655 (2016) (arguing that Hobby Lobby
precludes courts from considering the weight of the burden
imposed on religious claimants), but that intellectual variety
does not mean courts can adopt the reasoning they find most
appealing, rather than abiding by controlling Supreme Court

42
owe to the Individual Plaintiffs’ convictions is at odds with
the respect that has historically governed our approach to
expressions of religious belief. “The religious views
espoused by respondents might seem incredible” to some
people, “[b]ut if those doctrines are subject to trial … [to
determine] their truth or falsity, then the same can be done
with the religious beliefs of any sect.” United States v.
Ballard, 322 U.S. 78, 87 (1944). When judges wade into
those waters, “they enter a forbidden domain.” Id.; see also
James Madison, Memorial and Remonstrance against
Religious Assessments, Papers 8:298—304 (June 20, 1785)
(critiquing the notion that a civil judge can be “a competent
Judge of Religious Truth”). My friends in the Majority do
not simply wade in; they dive in with gusto, commenting that
their analysis allows them to “enumerate[] [each] allegation
in turn, and … conclude that the Real Alternatives Employees
have failed to demonstrate that the Contraceptive Mandate
forces them to violate their religious beliefs.” (Maj. Op. at
43.)

I sincerely wish that this were not the Majority’s
analytical approach. In a powerful dissent from the denial of
en banc review in the Little Sisters of the Poor case, Judge
Harris Hartz of the Tenth Circuit pointed out how fraught
with ill-consequences it can be. 31 Calling it a “dangerous

precedent. We are required to follow Hobby Lobby, and I am
pleased to do so since its reasoning is entirely persuasive.
31
The Majority criticizes my reliance on the dissenting
opinion in Little Sisters. I cite it as persuasive, not binding,
authority. And I note that the majority opinion in Little
Sisters was vacated by the Supreme Court in Zubik v.

43
approach to religious liberty,” Judge Hartz asked whether our
country could “really tolerate letting courts examine the
reasoning behind a religious practice or belief and decide
what is core and what is derivative?” Little Sisters, 799 F.3d
at 1317 (Hartz, J., dissenting). He used two examples to
demonstrate the serious problems raised by a what’s-the-big-
deal approach. First, it could require a Christian “to work on
December 25 because, according to a court, his core belief is
that he should not work on the anniversary of the birth of
Jesus but a history of the calendar and other sources show that
Jesus was actually born in March.” Id. Next, he said it would
allow the government to provide a Jewish prisoner with “only
non-kosher food because the real purpose of biblical dietary
laws is health, so as long as the pork is well-cooked, etc., the
prisoner’s religious beliefs are not substantially burdened.”
Id. at 1317-18. Such reasoning is “contrary to all precedent
concerning the free exercise of religion.” Id. at 1318.

I agree with Judge Hartz and decline to question the
Individual Plaintiffs’ religious beliefs under the guise of
adjudicating “substantial burden.” I respect their convictions
and conclude that the Contraceptive Mandate – which forces
them, under threat of monetary penalty, to sign up for and
participate in a system that violates their devoutly held beliefs
about human life – is a substantial burden on their exercise of
religion.

Burwell, 136 S. Ct. 1557 (2016). Moreover, my colleagues in
the Majority are not consistent in their rejection of dissenting
opinions. See (Maj. Op. at 46-47 (favorably describing the
dissent in Hobby Lobby).

44
 C. Strict Scrutiny

Because the Individual Plaintiffs are “substantially
burdened” by the Contraceptive Mandate, I turn to the “strict
scrutiny” questions the Majority does not address: whether
the government action is “in furtherance of a compelling
government interest” and is the “least restrictive means” of
achieving that interest. See 42 U.S.C. § 2000bb-1(b). That
standard is “exceptionally demanding,” Hobby Lobby, 134 S.
Ct. at 2780, and the government’s arguments are inadequate.
Along with many others who have considered the matter, I do
not believe that the Contraceptive Mandate can survive strict
scrutiny. 32 See id. (“HHS has not shown that it lacks other
means of achieving its desired goal without imposing a
substantial burden on the exercise of religion by the objecting
parties in these cases.”); March for Life, 128 F. Supp. 3d at
131 (“The final question the Court must ask under RFRA is
whether the current Mandate is the least restrictive means of
serving this governmental interest. Assuredly, it is not!”);
Wieland, 196 F. Supp. 3d at 1019 (holding that the
“government has not met its burden” to satisfy RFRA). I

32
The Majority avoids this step in the analysis by
holding that the Individual Plaintiffs are not substantially
burdened by the Mandate. My colleagues claim that I am
wrong to recognize that the Mandate has not survived strict
scrutiny on repeated occasions because only two courts have
“addressed the precise question before us today.” (Maj. Op.
at 59 n.37.) It is true that only two courts have faced the
identical dilemma, see supra p. 16-17, but I am confident that
the outcomes in the avalanche of related litigation the
Mandate has spawned are relevant, see, e.g., Hobby Lobby,
134 S. Ct. at 2779-81, and that is what I have referred to.

45
consider in turn both the interest initially advanced by the
government – access to contraception – and the government’s
newly discovered interest – a universal health care system.

1. Access to Contraception

In Hobby Lobby, the majority opinion assumed
without deciding that one interest proffered by the
government was compelling: “ensuring that all women have
access to all FDA-approved contraceptives without cost
sharing.” 134 S. Ct. at 2779. If that is a given, the question
becomes whether the Mandate is the least restrictive means of
furthering that compelling government interest. The test is
sometimes framed as an inquiry into whether the means is
“precisely tailored” to meet the compelling interest. Id. at
2783; see also Church of the Lukumi Babalu Aye, Inc. v. City
of Hialeah, 508 U.S. 520, 531–32 (1993) (recognizing in the
Free Exercise context that a burden on religion “must be
justified by a compelling governmental interest and must be
narrowly tailored to advance that interest”). The Hobby
Lobby Court concluded that there were several other options
available to the government to meet that interest, the most
straightforward of which would be for “the government to
assume the cost of providing the contraceptives at issue to
any women who are unable to obtain them under their health-
insurance policies due to their employers’ religious
objections.” 134 S. Ct. at 2780. Here, the government could
surely do the same thing, defraying the cost of contraceptive
coverage to the extent necessary to make up for the absence

46
of people in the insurance pool who decline the coverage. 33 If
allowing some people to opt out of the Mandate ended up
costing any significant amount, the government – whose
interest it is – could absorb the cost. See id. at 2781 (“RFRA
… may in some circumstances require the government to
expend additional funds to accommodate citizens’ religious
beliefs.”).

In Hobby Lobby, the Supreme Court observed that the
government had “already established an accommodation for
nonprofit organizations with religious objections[,]” id. at
2782, and the very existence of that accommodation proved
that less restrictive means could be used to reach the
government’s ends. See id. (considering that the
accommodation would “not impinge on the plaintiffs’
religious belief … and it serves HHS’s stated interests equally
well”). The same is true here. The government briefly argues
that an accommodation cannot be possible for individual
buyers of insurance, saying that exemptions to the Mandate
“apply only to employers … not individuals.” See (Gov. Br.
at 29 (quoting District Court Opinion, JA 76)). But the
government does not justify why employers deserve an
accommodation and individuals do not. Indeed, the argument
is quite an about-face from the position the government took
in Hobby Lobby and Conestoga Wood, when it contended
loudly that only individuals could have religious scruples and
the companies who employed them could not. Hobby Lobby,
134 S. Ct. at 2774 (“HHS contends that Congress could not

33
It is not clear that the government would need to
step in at all, since the number of people wanting to avoid
such coverage is unknown.

47
have wanted RFRA to apply to for-profit corporations
because it is difficult as a practical matter to ascertain the
sincere ‘beliefs’ of a corporation.”); Conestoga Wood, 724
F.3d at 403 (Jordan, J., dissenting) (noting the irony in the
idea “religious belief takes shape within the minds and hearts
of individuals” while denying religious liberty to “an entity
that is nothing more than the common vision of five
individuals from one family who are of one heart and mind
about their religious belief”). The Individual Plaintiffs have
proposed a number of ways the government could satisfy its
interest in providing contraceptive coverage. 34

Parties to the Zubik litigation also suggested ways that
access to contraceptives could be provided without trampling
on religious beliefs. See Zubik v. Burwell, Supp. Reply Brief
For Petitioners, 2016 WL 1593773 at *1 (filed April 20,
2016) (“The government concedes … that its existing
regulatory scheme ‘could be modified’ to eliminate the self-
certification requirement for petitioners with insured plans
without sacrificing its professed objective of ‘ensuring that
the affected women receive contraceptive coverage
seamlessly.’”) (quoting Respondent’s Supp. Br. at 14-15).
The Petitioners in Zubik outlined a solution in which “the
insurance company [could be made to] make available to plan
beneficiaries a separate plan providing the excluded

34
The government could include religiously-objecting
families in “existing federal family planning programs” that
provide coverage for free or reduced rates; provide objecting
families with “federal subsidies” to offset the cost of the
coverage; or require the government to pay insurance
companies directly for the added cost of contraceptive
coverage. See (Opening Br. at 54-55).

48
contraceptive coverage” and separately “contact beneficiaries
to inform them of that plan and how to enroll.” See Zubik,
Supp. Br. for Petitioners, 2016 WL 1445914 at *4 (filed
April 12, 2016). The distinct plans would be akin to dental or
vision insurance that are “truly” independent of general health
insurance and have a separate enrollment process, insurance
card, and payment source. Id. at *1. That same option could
be provided to individuals purchasing health care on the open
market.

The wisdom of those options may be debated, but not
their existence, so the government’s decision to simply refuse
to engage in the discussion is telling. It appears that the
government “has open to it [several] less drastic way[s] of
satisfying its legitimate interest[]” and has made “no showing
that any of the [Individual Plaintiffs’] alternative ideas would
be unworkable.” See Conestoga Wood, 724 F.3d at 414-15
(Jordan, J., dissenting) (citations omitted). Thus, the
government’s position cannot withstand strict scrutiny.

2. A Uniform Health Care System

Evidently recognizing that it cannot win if its interest
is described as providing contraceptive coverage, the
government actually abandons that position and declares it to
be “irrelevant,” (Responding Br. at 27), which is remarkable
given how intensely it insisted that that interest was
compelling before. Nevermind. It has a new set of interests
now. In its words, “a compelling interest in the provision of
health care and the functioning of the insurance market …
[and] a corresponding ‘interest in the uniformity of the health
care system the ACA puts in place, under which all eligible
citizens receive the same minimum level of coverage’” are

49
the only rationales we should consider. (Responding Br. at
25 (quoting Priests for Life v. U.S. Department of Health &
Human Services, 772 F.3d 229, 265 (D.C.Cir. 2014)).) 35

These sweeping claims fly in the face of the Supreme
Court’s command in Hobby Lobby that compelling
government interests must be precisely defined. The Court
there rejected the government’s attempt to assert interests that
were “couched in very broad terms, such as promoting
‘public health’ and ‘gender equality.’” Hobby Lobby, 134 S.
Ct. at 2779. Instead, it said, judges are “to loo[k] beyond
broadly formulated interests and to scrutinize[e] the asserted
harm of granting specific exemptions to particular religious
claimants—in other words, to look to the marginal interest in
enforcing the Contraceptive Mandate in these cases.” Id.
(alterations in original) (quotations omitted). A generalized
interest in health care and insurance is too abstract to be
compelling in a legal sense when addressing the Individual
Plaintiffs’ request for relief.

No more compelling is the government’s claimed
interest in uniformity of the Mandate’s application. That
claim cannot be given credence because millions of people
have already been excepted. “The Mandate is a classic …

35
There is more than a whiff of gamesmanship about
the government’s newly claimed compelling interest. In a
strict scrutiny analysis, we ordinarily reject “post hoc
rationalizations” for government action and instead rely on
the “basis [for the regulation] articulated by the agency
itself.” See Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State
Farm Mut. Auto. Ins. Co., 463 U.S. 29, 50 (1983).

50
example of … arbitrary underinclusiveness.” Conestoga
Wood, 724 F.3d at 414-15 (Jordan, J., dissenting). As the
Supreme Court observed, “[a]ll told, the Contraceptive
Mandate presently does not apply to tens of millions of
people.” Hobby Lobby, 134 S. Ct. at 2764 (internal quotation
and citation omitted). “A law cannot be regarded as
protecting an interest of the highest order when it leaves
appreciable damage to that supposedly vital interest
unprohibited.” Conestoga Wood, 724 F.3d at 413 (Jordan, J.,
dissenting) (quoting Church of Lukumi Babalu, 508 U.S. at
547). 36 The government cannot persuasively declare that it
has an interest in universality and uniformness – only to, at
the same time, make the means decidedly not universal and

36
In addition to the variety of exemptions from the
employer mandate discussed in Hobby Lobby, 134 S. Ct. at
2764, there are also a wide variety of exemptions from the
individual mandate. Significantly, there is an exemption for
those who have membership in a religious sect that objects to
insurance. 26 U.S.C. § 5000A(d)(2)(A). Additionally,
individuals who participate in a previously-existing “health
care sharing ministry” are exempted from the Mandate. 26
U.S.C. § 5000A(d)(2)(B) (defining a health care sharing
ministry as a tax-exempt organization of members who “share
a common set of ethical or religious beliefs and share medical
expenses among members in accordance with those beliefs”
that has existed since December 31, 1999). Those additional
exemptions further underscore the Mandate’s
underinclusiveness.

51
uniform. Because of this incongruity, the claimed interest
cannot credibly be characterized as compelling. 37

37
The United States Court of Appeals for the District
of Columbia Circuit in Priests For Life v. United States
Department of Health & Human Services, 772 F.3d 229 (D.C.
Cir. 2014), vacated and remanded sub nom. Zubik v. Burwell,
136 S. Ct. 1557 (2016), accepted a number of government
interests as compelling, including “a sustainable system of
taxes and subsidies under the ACA to advance public health.”
Id. at 258. In considering the Mandate’s furtherance of that
interest, the court concluded that “[t]he government’s interest
in a comprehensive, broadly available system is not undercut
by the other exemptions in the ACA, such as the exemptions
for religious employers, small employers, and grandfathered
plans.” Id. at 266.
That holding is not binding on us and, in any event, is
an assertion rather than a reasoned conclusion. The scope of
the exceptions here is far more significant than the “narrow
category” exempted from Social Security in United States v.
Lee, 455 U.S. 252, 261 (1982). See Hobby Lobby Stores,
Inc. v. Sebelius, 723 F.3d 1114, 1143 (10th Cir. 2013), aff'd
sub nom. Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct.
2751 (2014) (“[T]he interest here cannot be compelling
because the contraceptive-coverage requirement presently
does not apply to tens of millions of people.”). I am
persuaded that the number of “congressional exemptions” to
the Mandate demonstrate that the ACA does “not preclude
exceptions altogether” and “RFRA makes clear that it is the
obligation of the courts to consider whether exceptions are
required under the test set forth by Congress.” Gonzales v. O
Centro Espirita Beneficente Uniao do Vegetal, 546 U.S. 418,
421, 434 (2006).

52
 Charitably assuming that the government’s interest is
better understood as a functioning and comprehensive
insurance market, the Mandate is again not the least
restrictive means of achieving that interest. The government
points to United States v. Lee, 455 U.S. at 257-258, to say a
universal system can be the least restrictive means to achieve
a compelling interest. True enough. But in Lee, the Supreme
Court was considering taxation to provide a “comprehensive
national social security system.” Id. at 258. As Judge Leon
pointed out in March for Life, there is a “critical distinction”
between that scheme and the ACA:

Unlike in Lee, the government does not provide
the insurance at issue here, and there is no
single “comprehensive national [health
insurance] system.” See Lee, 455 U.S. at 258,
102 S. Ct. 1051. Instead, the government
regulates a host of third party insurers. The
Mandate burdens employee plaintiffs’ religious
exercise by restricting the form in which those
third parties can offer something that plaintiffs,
for all intents and purposes, must buy.

March for Life, 128 F. Supp. 3d at 131–32.

Understood from that perspective, there is an obvious
solution to further the government’s interest: refrain from
penalizing insurers who offer plans in accordance with the
Individual Plaintiffs’ beliefs. Id. at 132 (“The government
need not require an insurer offer such a plan at plaintiffs’
request in order to avoid burdening plaintiffs’ religious
exercise.”). Because insurance companies would offer such
plans as a result of market forces, doing so would not

53
undermine the government’s interest in a “sustainable and
functioning market.” Id. And that remedy would also
necessarily be limited in scope; it would not “enable
[insurance companies] to refuse to provide [contraceptive]
coverage to others who do not share those religious
objections.” Id. Because the government has failed to
demonstrate why allowing such a system (not unlike the one
that allowed wider choice before the ACA) would be
unworkable, it has not satisfied strict scrutiny.

III. Conclusion

To the Majority, this is all much ado about nothing: the
burden of signing forms and paying money in support of
drugs, devices, and procedures that affect the well springs of
human life is so slight it cannot be called substantial, so the
Individual Plaintiffs should simply sign and pay and stop
complaining. What my colleagues fail to appreciate is that
coercing financial support for something deeply objectionable
is a real and substantial burden, and a forced signature alone
can be problematic. In matters of conscience, the signing of
one’s name is more than a scrawl on paper. Robert Bolt gave
these compelling words to Sir (and Saint) Thomas More in
the play “A Man for All Seasons”: “When a man takes an
oath, … he’s holding his own self in his hands. Like water.
And if he opens his fingers then – he needn’t hope to find
himself again.”

The Individual Plaintiffs do not want to lose
themselves. They have demonstrated the seriousness of the
burdens forced upon them by the Contraceptive Mandate.
Under RFRA, it thus became incumbent on the government to
show that its actions are narrowly tailored to achieve a

54
compelling purpose. In my estimation, the government has
failed to meet that exacting standard. I thus respectfully
dissent and concur only in the judgment as to Real
Alternatives, not in the judgment against the Individual
Plaintiffs.

55