Francis A. GILARDI, et al., Appellants v. UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES, et al., Appellees, 733 F.3d 1208

Opinion for the Court filed by Circuit Judge BROWN, with whom Senior Circuit Judge EDWARDS joins except as to parts VI, VII, and VIII, and with whom Senior Circuit Judge RANDOLPH joins except as to parts III and IV. Opinion concurring in part and concurring in the judgment filed by Senior Circuit Judge RANDOLPH. Opinion concurring in part and dissenting in part filed by Senior Circuit Judge EDWARDS. BROWN, Circuit Judge. Two years after our decision Seven-Sky v. Holder, 661 F.3d 1 (D.C.Cir.2011), we are asked to revisit the behemoth known as the Affordable Care Act.

This time, however, we are not confronted with a question of constitutional authority. Instead, we must determine whether the contraceptive mandate imposed by the Act trammels the right of free exercise — a right that lies at the core of our constitutional liberties — as protected by the Religious Freedom Restoration Act. We conclude it does. I Two brothers, Francis and Philip Gilardi, are equal owners of Freshway Foods and Freshway Logistics — both companies are closely-held corporations that have elected to be taxed under Subchapter S of the Internal Revenue Code.

The two companies collectively employ about 400 employees and operate a self-insured health plan through a third-party administrator and stop-loss provider. As adherents of the Catholic faith, the Gilardis oppose contraception, sterilization, and abortion. Accordingly, the two brothers' — exercising their powers as owners and company executives — excluded coverage of products and services falling under these categories. But along came the Affordable Care Act.

Part of the Act directs all group health plans and health insurance issuers to provide, without cost-sharing requirements, preventive care as determined by the Health Resources and Services Administration. 42 U.S.C. § 300gg-13(a)(4). In turn, the Administration issued guidelines requiring coverage for “all Food and Drug Administration-approved contraceptive methods, sterilization procedures, and patient education and counseling for all women with reproductive capacity,” as prescribed by a healthcare provider. Women’s Preventive Services Guidelines, Health Res. & Servs.

Admin., http://www.hrsa.gov/womensguidelines/; see Group Health Plans and Health Insurance Issuers Relating to Coverage of Preventive Services Under the Patient Protection Affordable Care Act, 77 Fed.Reg. 8725, 8725-26 (Feb. 15, 2012) (citing the online HRSA Guidelines); see also 29 C.F.R. § 2590.715-2713(a)(l)(iv); 45 C.F.R. 147.131(c). There are exceptions — some ephemeral, some permanent — for grandfathered plans, religious organizations, and small businesses. See 26 U.S.C. § 4980H(a); id. § 4980H(c)(2)(A); 42 U.S.C. § 18011; 45 C.F.R. §§ 147.130(a)(l)(iv)(A)-(B).

But the Freshway companies do not fall into any of these categories. As a result, the Gilardis were faced with two choices: adjust their companies’ plans to provide the mandated contraceptive services in contravention of their religious beliefs, or pay a penalty amounting to over $14 million per year.

Finding themselves on the horns of an impossible dilemma, the Gilardis and their companies filed suit in district court, alleging the contraceptive mandate violated their rights under the Religious Freedom Restoration Act (RFRA), 42 U.S.C. § 2000bb et seq., the Free Exercise Clause, the Free Speech Clause, and the Administrative Procedure Act. The plaintiffs moved for a preliminary injunction, but the district court denied their request. With respect to the Freshway companies, the court determined they could not “exereise” religion and thus no substantial burden on religious exercise was demonstrable under RFRA.

As for the Gilardis, the court found any burden on the Gilardis’ religious beliefs was indirect. The plaintiffs timely filed an interlocutory appeal and moved for an injunction pending appeal. After having initially denied their motion, we issued, sua sponte, an order giving them a temporary reprieve from the mandate. II Our standard of review for a denial of a preliminary injunction rests upon what aspect of the district court’s decision we are examining.

Insofar as our review concerns the district court’s consideration of the preliminary-injunction factors and the ultimate decision to grant or deny the injunction, we review for an abuse of discretion. See In re Navy Chaplaincy, 697 F.3d 1171, 1178 (D.C.Cir.2012). But we review the legal conclusions underlying the decision de novo and review findings of fact for clear error. Id.; Chaplaincy of Full Gospel Churches v.

England, 454 F.3d 290, 297 (D.C.Cir.2006). “In ruling on a preliminary injunction a key issue — often the dispositive one — is whether the movant has shown a substantial likelihood of success on the merits.” Greater New Orleans Fair Hous. Action Ctr. v. U.S. Dep’t of Hous. & Urban Dev., 639 F.3d 1078, 1083 (D.C.Cir.2011). To determine this likelihood, we must answer whether the contraceptive mandate of 45 C.F.R. § 147.130(a)(1)(iv), as applied to the Appellants, violates their free-exercise rights as protected by RFRA.

As the parties have made eminently clear, we must separately examine the claims by the Freshway companies and their owners. III We begin with the Freshway companies. Before addressing the merits of their RFRA claim, we must first ask whether they may bring the challenge at all. The statute allows “[a] person whose religious exercise has been burdened” to seek judicial relief, but leaves us bereft of guidance on who a “person” is. See 42 U.S.C. § 2000bb-l(c) (emphasis added). For at least one of our sister circuits (as well as the Appellants), the Dictionary Act, 1 U.S.C. § 1, dispositively answers the question.

See Hobby Lobby Stores, Inc. v. Sebelius, 723 F.3d 1114, 1129, 1132 (10th Cir.2013) (en banc). Under the Act, the definition of “person” extends to “corporations, companies, associations, firms, partnerships, societies, and joint stock companies”—in other words, it encompasses the corporeal and the incorporeal. 1 U.S.C. § 1. The Freshway companies largely depend on the Dictionary Act’s elision of the differences in identity, hoping it applies to their RFRA claim. But the focus on personhood is too narrow; instead, we must construe the term “person” together with the phrase “exercise of religion.” See Rasul v.

Myers, 512 F.3d 644, 668 (D.C.Cir.2008) (“Because RFRA prohibits the Government from ‘substantially burdening] a person’s exercise of religion’ instead of simply the exercise of religion, 42 U.S.C. § 2000bb-1(a), we must construe ‘person’ as qualifying ‘exercise of religion.’ ” (emphasis in original)), vacated and remanded on other grounds by 555 U.S. 1083, 129 S.Ct. 763, 172 L.Ed.2d 753 (2008); see also 42 U.S.C. § 2000bb-1(c) (“A person whose religious exercise has been burdened in violation of this

section may ... obtain appropriate relief against a government.” (emphasis added)). And RFRA provides us with no helpful definition of “exercise of religion”; all we can glean from the statute is that “ ‘religious exercise’ includes any exercise of religion, whether or not compelled by, or central to, a system of religious belief.” 42 U.S.C. §§ 2000bb-2(4), 2000cc-5(7)(A). We must therefore turn to the full body of our free-exercise caselaw to discern whether the Freshway companies are persons capable of religious exercise under the statute. See Rasul, 512 F.3d at 671; see also Autocam Corp. v.

Sebelius, 730 F.3d 618, 625-26 (6th Cir.2013); Hobby Lobby, 723 F.3d at 1167 (Briscoe, C.J., concurring in part and dissenting in part). IV The query is simple: do corporations enjoy the shelter of the Free Exercise Clause? Or is the free-exercise right a “purely personal” one, such that it is “unavailable to corporations and other organizations because the ‘historic function’ of the particular guarantee has been limited to the protection of individuals”? First Nat’l Bank of Bos. v. Bellotti, 435 U.S. 765, 778 n. 14, 98 S.Ct. 1407, 55 L.Ed.2d 707 (1978) (quoting United States v.

White, 322 U.S. 694, 698-701, 64 S.Ct. 1248, 88 L.Ed. 1542 (1944)). We turn to the “nature, history, and purpose” of the Clause for our answer. Id. At the time of the Framing, a great debate raged on the precise formulation of what we now know as the Free Exercise Clause. See Michael W. McConnell, The Origins and Historical Understanding of Free Exercise of Religion, 103 Harv. L.Rev. 1409, 1480-85 (1990).

The earliest drafts from the House of Representatives focused on the protection of conscience, rather than the “exercise of religion.” See id. at 1482; see also 1 Annals of Congress 729 (1789) (noting a later amendment to change the Clause’s prototype to read: “no religion shall be established by law, nor shall the equal rights of conscience be infringed”). And as the debates went on, strong concerns arose that the rights of religious sects would not be “well secured under the ... Constitution.” 1 Annals of Congress 730 (remarks of Daniel Carroll, Aug. 15, 1789).

To address these concerns and others, the House continued to tinker and toil; once the dust had settled, it eventually proposed a constitutional amendment barring Congress from “prevent[ing] the free exercise” of religion and “infring[ing] the rights of conscience.” 1 id. at 766. But the Senate had different ideas, and in the end, it was the free exercise of religion — standing alone — that was sent to the states for ratification. See McConnell, supra, at 1488; see also Louis Fisher, Religious Liberty in America: Political Safeguards 56 (2002).

This history reveals two things about the Clause’s purpose relevant to our inquiry today. First, the constitutional guarantee “extended the broader freedom of action to all believers,” allowing for the inclusion of “conduct as well as belief.” McConnell, supra, at 1490. Second, the adopted formulation encompassed both individual judgment, as well as “the corporate or institutional aspects of religious belief.” Id.

Because the word religion “connotes a community of believers,” the prohibition against the impingement on religious free exercise must be understood to cover the activities of both individuals and religious bodies. See id. And these two groups have been the beneficiaries of the Supreme Court’s free-exercise jurisprudence. To be sure, the right has largely been understood as a personal one. Before incorporation, the Court described the free-exercise right as an individual one — “the indefeasible right to worship God according to the dictates of conscience.” Cummings v.

Missouri, 71 U.S. 277, 304, 4 Wall. 277, 18 L.Ed. 356 (1866). Incorporation did nothing to alter that sentiment; shortly after Cantwell v. Connecticut, 310 U.S. 296, 60 S.Ct. 900, 84 L.Ed. 1213 (1940), the Court reaffirmed the personal nature of the right as part of “the mind and spirit of man.” See Jones v. City of Opelika, 316 U.S. 584, 594, 62 S.Ct. 1231, 86 L.Ed. 1691 (1942), overruled on other grounds by 319 U.S. 103, 63 S.Ct. 890, 87 L.Ed. 1290 (1943); see also Sch. Dist. of Abington Twp. v.

Schempp, 374 U.S. 203, 223, 83 S.Ct. 1560, 10 L.Ed.2d 844 (1963) (“[The purpose of the Free Exercise Clause] is to secure religious liberty in the individual by prohibiting any invasions thereof by civil authority.” (emphasis added)). And that understanding still resonates with the modern Court. See Zelman v. Simmons-Harris, 536 U.S. 639, 679 n. 4, 122 S.Ct. 2460, 153 L.Ed.2d 604 (2002) (Thomas, J., concurring) (“In particular, these rights inhere in the Free Exercise Clause, which unlike the Establishment Clause protects individual liberties of religious worship.”).

That is not to say the Court views organizations as constitutional outliers — indeed, its jurisprudence reflects the foundational principle that religious bodies— representing a communion of faith and a community of believers — are entitled to the shield of the Free Exercise Clause. The Court has heard free-exercise challenges from religious entities and religious organizations. See Jimmy Swaggart Ministries v. Bd. of Equalization of Cal., 493 U.S. 378, 381, 110 S.Ct. 688, 107 L.Ed.2d 796 (1990); Corp. of the Presiding Bishop of the Church of Jesus Christ of Latter-day Saints v.

Amos, 483 U.S. 327, 330, 107 S.Ct. 2862, 97 L.Ed.2d 273 (1987); Tony & Susan Alamo Found. v. Sec’y of Labor, 471 U.S. 290, 292, 105 S.Ct. 1953, 85 L.Ed.2d 278 (1985); Kedroff v. St. Nicholas Cathedral of Russian Orthodox Church in N. Am., 344 U.S. 94, 107-08, 73 S.Ct. 143, 97 L.Ed. 120 (1952). It has listened to the grievances of religious sects and member congregations. See Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, — U.S. -, 132 S.Ct. 694, 699, 181 L.Ed.2d 650 (2012); Gonzales v.

O Centro Espirita Beneficente Uniao do Vegetal, 546 U.S. 418, 425, 126 S.Ct. 1211, 163 L.Ed.2d 1017 (2006); Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 524, 113 S.Ct. 2217, 124 L.Ed.2d 472 (1993). It has even entertained claims by religious and educational institutions. See Bob Jones Univ. v. United States, 461 U.S. 574, 579-80, 103 S.Ct. 2017, 76 L.Ed.2d 157 (1983). Beyond these cases involving religious organizations, however, we glean nothing from the Court’s jurisprudence that suggests other entities may raise a free-exercise challenge.

But that the Court has never seriously considered such a claim by a secular corporation or other organizational entity is not to say it never will. For the nonce, only one aberrational case comes to mind. In Gallagher v. Crown Kosher Super Market of Massachusetts, Inc., 366 U.S. 617, 81 S.Ct. 1122, 6 L.Ed.2d 536 (1961), a corporation operated by members of the Orthodox Jewish faith challenged the constitutionality of Massachusetts’ Sunday closing laws. See id. at 618, 81 S.Ct. 1122. The Court summarily disposed of the corporation’s free-exercise claim, tersely noting that Braunfeld v.

Brown, 366 U.S. 599, 81 S.Ct. 1144, 6 L.Ed.2d 563 (1961), obviated the need for further discussion. See Crown Kosher, 366 U.S. at 631, 81 S.Ct. 1122. Technically speaking, the Court did rule on the merits of the case. But it remained dubitante about standing — perhaps the novelty of a secular corporation bringing a free-exercise challenge was too novel. See id. (“Since the decision in [.Braunfeld ] rejects the contentions presented by these appellees on the merits, we need not decide whether appellees have standing to raise these questions.”); cf. Hobby Lobby, 723 F.3d at 1150 (Hartz, J., concurring).

Meanwhile, we need not base a right of free exercise for nonreligious organizations on so thin a reed of caselaw, especially as both we and the Supreme Court have expressed strong doubts about that proposition. See, e.g., Harris v. McRae, 448 U.S. 297, 321, 100 S.Ct. 2671, 65 L.Ed.2d 784 (1980) (explaining that a free-exercise challenge is “one that ordinarily requires individual participation”); Holy Land Found. for Relief & Dev. v.

Ashcroft, 333 F.3d 156, 167 (D.C.Cir.2003) (noting “the dubious proposition that a charitable corporation not otherwise defined can exercise religion as protected in the First Amendment”). Citing Citizens United v. FEC, 558 U.S. 310, 130 S.Ct. 876, 175 L.Ed.2d 753 (2010), the Freshway companies argue that corporations — religious or otherwise — are entitled to the full array of First Amendment protections, including the right to free exercise. They are not the only proponents of this position.

See Hobby Lobby, 723 F.3d at 1135 (majority opinion) (“Because Hobby Lobby and Mardel express themselves for religious purposes, the First Amendment logic of Citizens United, where the Supreme Court has recognized a First Amendment right of for-profit corporations to express themselves for political purposes, applies as well.” (citation omitted)); see also Conestoga Wood Specialties Corp. v. Sec’y of the U.S. Dep’t of Health & Human Servs., 724 F.3d 377, 400 (3d Cir.2013) (Jordan, J., dissenting) (citing Citizens United, 558 U.S. at 342, 130 S.Ct. 876).

There is an appeal to this simple reasoning; after all, the free-exercise and free-speeeh rights are enshrined in the same constitutional provision, separated only by a semicolon. Perhaps Appellants’ constitutional arithmetic, Citizens United plus the Free Exercise Clause equals a corporate free-exercise right, will ultimately prevail. But we must be mindful that Citizens United represents the culmination of decades of Supreme Court jurisprudence recognizing that all corporations speak. See Conestoga Wood, 724 F.3d at 384.

When it comes to the free exercise of religion, however, the Court has only indicated that people and churches worship. As for secular corporations, the Court has been all but silent. Consider Bellotti — the progenitor of Citizens United.

When the Bellotti Court declared “political speech does not lose First Amendment protection ‘simply because its source is a corporation,’” Citizens United, 558 U.S. at 342, 130 S.Ct. 876 (quoting Bellotti, 435 U.S. at 784, 98 S.Ct. 1407), it reviewed many cases in which the Court invalidated a state law because it “infringe[d on] protected speech by corporate bodies.” Bellotti, 435 U.S. at 778 n. 14, 98 S.Ct. 1407. In other words, Bellotti crystallized a robust body of caselaw giving rise to the constitutional right of corporate political speech, which the Citizens United Court could rely on as a firm foundation.

No such corpus juris exists to suggest a free-exercise right for secular corporations. Thus, we read the “nature, history, and purpose” of the Free Exercise Clause as militating against the discernment of such a right. When it comes to corporate entities, only religious organizations are accorded the protections of the Clause. And we decline to give credence to the notion that the for-profit/non-profit distinction is dispositive, as that, too, is absent from the Clause’s history.

Fortunately, we need not opine here on what a “religious organization” is, as the Freshway companies have conceded they do not meet that criterion. The Freshway companies alternatively assert they can vindicate the free-exercise rights of their owners. They reason that if “a company is owned and controlled by a few like-minded individuals who share the same religious values and run the company pursuant to those values,” the company may serve as the owners’ surrogate. Appellants’ Br. at 50.

This pass-through theory of corporate standing is logically and structurally appealing in light of the government’s shell game. And EEOC v. Townley Engineering & Manufacturing Co., 859 F.2d 610 (9th Cir.1988) provides longstanding, if illusory, support. In Townley, the Ninth Circuit concluded— without much in the way of legal substantiation — that the corporation was “merely the instrument through and by which [the owners] expressed] their religious beliefs.” Id. at 619. Admittedly, there is a certain theological congruence to Townley’s characterization.

The Bible says “faith without works is dead.” James 2:26 (King James). As amici point out, not only are Catholic employers morally responsible for the management of their companies, “instructing or encouraging someone else to commit a wrongful act is itself a grave moral wrong — i.e., ‘scandal’ — under Catholic doctrine.” Br. of Catholic Theologians at 3. Thus, amici reason, “the Mandate thrusts Catholic employers into a ‘perfect storm’ of moral complicity in the forbidden actions.” Br. of Catholic Theologians at 5; see also Br. of the Archdiocese of Cincinnati at 16-17 nn. 6, 7.

When even attenuated participation may be construed as a sin, see, e.g., United States v. Lee, 455 U.S. 252, 261 n. 12, 102 S.Ct. 1051, 71 L.Ed.2d 127 (1982), it is not for courts to decide that the corporate veil severs the owner’s moral responsibility. But dogma does not dictate justiciability. Though Townley’s conclusion is theologically defensible, its standing bona fides, supported only by a reference to a footnote in Tony & Susan Alamo Foundation v. Secretary of Labor, 471 U.S. 290, 105 S.Ct. 1953, 85 L.Ed.2d 278 (1985), are more dubious.

The Alamo Foundation Court relied on a theory of religious associational standing; in other words, the organization could raise a free-exercise defense on behalf of one of its executives because the executive was an adherent to the group’s religious creed. See id. at 303 n. 26, 105 S.Ct. 1953. How this supports standing for a secular corporation to vindicate its owners’ free-exercise rights is unclear. Townley’s misconception of religious assoeiational standing has spread from one free-exercise case to another, even creeping its way into the current contraceptive mandate challenges. See Stormans, Inc. v.

Selecky, 586 F.3d 1109, 1120 (9th Cir.2009); see also Monaghan v. Sebelius, 931 F.Supp.2d 794, 800-02 (E.D.Mich.2013); Geneva Coll. v. Sebelius, 929 F.Supp.2d 402, 428 (W.D.Pa.2013). While we decline the Freshway companies’ invitation to accept Townley’s ipse dixit that closely held corporations can vindicate the rights of their owners, we understand the impulse. The free exercise protection — a core bulwark of freedom — should not be expunged by a label. But for now, we have no basis for concluding a secular organization can exercise religion. V That leaves the Gilardis.

Obviously, they have no difficulty satisfying the threshold inquiry to which their enterprises succumbed; they are, most assuredly, “persons” under RFRA. See also Rasul v. Myers, 563 F.3d 527, 533 (D.C.Cir.2009) (Brown, J., concurring) (“RFRA does not define ‘person,’ so we must look to the word’s ordinary meaning. There is little mystery that a ‘person’ is ‘an individual human being ... as distinguished from an animal or a thing.’ ” (quoting Webster’s New International Dictionary 1606 (1981))).

And there is no dispute that the mandate, as directed to the Gilardis, is a palpable and discernible infringement of free exercise. All that stands between the Gilardis and the hope of vindication is the uncertain barrier of the shareholder-standing rule and an inchoate concern about prudential standing — a “jurisdictional issue which cannot be waived or conceded” in this circuit. Ass’n of Battery Recyclers, Inc. v. EPA, 716 F.3d 667, 674 (D.C.Cir.2013).

The shareholder-standing rule gives us little pause; we are satisfied that the Gilardis have been “injured in a way that is separate and distinct from an injury to the corporation.” See Crosby v. Beam, 47 Ohio St.3d 105, 548 N.E.2d 217, 219 (1989); see also Rawoof v. Texor Petroleum Co., 521 F.3d 750, 757 (7th Cir.2008) (employing the state-law derivative action rule to address shareholder standing in a federal question case).

If the companies have no claim to enforce — and as nonreligious corporations, they cannot engage in religious exercise — we are left with the obvious conclusion: the right belongs to the Gilardis, existing independently of any right of the Freshway companies. Thus, the Gilardis’ injury — which arises therefrom — is “separate and distinct,” providing us with an exception to the shareholder-standing rule. VI We now reach the heart of the Gilardis’ RFRA claim. The Act requires the Gilardis to “allege! ] a substantial burden on [their] religious exercise.” Kaemmerling v. Lappin, 553 F.3d 669, 677 (D.C.Cir.2008).

Religious exercise is broadly defined as “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); see also id. § 2000bb-2. A “substantial burden” is “substantial pressure on an adherent to modify his behavior and to violate his beliefs.” Kaemmerling, 553 F.3d at 678 (quoting Thomas v. Review Bd. of Ind. Emp’t Sec. Div., 450 U.S. 707, 718, 101 S.Ct. 1425, 67 L.Ed.2d 624 (1981)). We begin with the peculiar step of explaining what is not at issue.

This case is not about the sincerity of the Gilardis’ religious beliefs, nor does it concern the theology behind Catholic precepts on contraception. The former is unchallenged, while the latter is unchallengeable. See id. at 716, 101 S.Ct. 1425 (“Particularly in this sensitive area, it is not within the judicial function and judicial competence to inquire whether the petitioner or his fellow worker more correctly perceived the commands of their common faith. Courts are not arbiters of scriptural

interpretation.”); see also United States v. Ballard, 322 U.S. 78, 86, 64 S.Ct. 882, 88 L.Ed. 1148 (1944) (“Men may believe what they cannot prove. They may not be put to the proof of their religious doctrines or beliefs.”). Equally uncontroverted is the nature of the Gilardis’ religious exercise: they operate their corporate enterprises in accordance with the tenets of their Catholic faith. See Hobby Lobby, 723 F.3d at 1189 (Matheson, J., concurring in part and dissenting in part). The only dispute touches on the characterization of the burden.

The burden is too remote and too attenuated, the government says, as it arises only when an employee purchases a contraceptive or uses contraceptive services. We disagree with the government’s foundational premise. The burden on religious exercise does not occur at the point of contraceptive purchase; instead, it occurs when a company’s owners fill the basket of goods and services that constitute a healthcare plan. In other words, the Gilardis are burdened when they are pressured to choose between violating their religious beliefs in managing their selected plan or paying onerous penalties.

See Thomas, 450 U.S. at 717-18, 101 S.Ct. 1425; Wisconsin v. Yoder, 406 U.S. 205, 218, 92 S.Ct. 1526, 32 L.Ed.2d 15 (1972) (“The impact of the compulsory-attendance law on respondents’ practice of the Amish religion is not only severe, but inescapable, for the Wisconsin law affirmatively compels them, under threat of criminal sanction, to perform acts undeniably at odds with fundamental tenets of their religious beliefs.”); Kaemmerling, 553 F.3d at 678. The Framers of the Constitution clearly embraced the philosophical insight that government coercion of moral agency is odious.

Penalties are impertinent, according to Locke, if they are used to compel men “to quit the light of their own reason, and oppose the dictates of their own consciences.” John Locke, A Letter Concerning Toleration 13-14 (J. Brook ed., 1792) (1689). Madison described conscience as “the most sacred of all property,” James Madison, Property, Nat’l Gazette, Mar. 29, 1792, at 174, reprinted in James Madison’s “Advice to My Country” 25, 83-84 (David B. Mattern ed., 1997), and placed the freedom of conscience prior to and superior to all other natural rights.

Religion, he wrote, is “the duty which we owe to our Creator ... being under the direction of reason and conviction only, not of violence or compulsion,” 1 Madison Papers 174 (1962), “precedent” to “the claims of Civil Society,” James Madison, Memorial and Remonstrance Against Religious Assessments (1785); see also United States v. Macintosh, 283 U.S. 605, 633-34, 51 S.Ct. 570, 75 L.Ed. 1302 (1931) (Hughes, C.J., dissenting) (“[I]n the forum of conscience, duty to a moral power higher than the state has always been maintained....

The essence of religion is belief in a relation to God involving duties superior to those arising from any human relation.”). From thence sprang the idea that the right to free exercise necessarily prohibits the government from “compelling] a man to furnish contributions of money for the propagation of opinions which he disbelieves.” Thomas Jefferson, The Virginia Act for Establishing Religious Freedom (1786). And that prohibition has plainly manifested itself throughout the years as an integral component of the free-exercise guarantee. Justice Brennan, writing for the Court in Sherbert v.

Verner, 374 U.S. 398, 83 S.Ct. 1790, 10 L.Ed.2d 965 (1963), put it well: “Government may neither compel affirmation of a repugnant belief, nor penalize or discriminate against individuals because they hold religious views abhorrent to the authorities.” Id. at 402, 83 S.Ct. 1790 (citations omitted). The contraceptive mandate demands that owners like the Gilardis meaningfully approve and endorse the inclusion of contraceptive coverage in their companies’ employer-provided plans, over whatever objections they may have.

Such an endorsement — procured exclusively by regulatory ukase — is a “compelled] affirmation of a repugnant belief.” See id. That, standing alone, is a cognizable burden on free exercise. And the burden becomes substantial because the government commands compliance by giving the Gilardis a Hobson’s choice. They can either abide by the sacred tenets of their faith, pay a penalty of over $14 million, and cripple the companies they have spent a lifetime building, or they become complicit in a grave moral wrong.

If that is not “substantial pressure on an adherent to modify his behavior and to violate his beliefs,” we fail to see how the standard could be met. See Thomas, 450 U.S. at 718, 101 S.Ct. 1425. In suggesting that no substantial burden lies with the Gilardis, the government invokes the principles undergirding the bargain for the corporate veil.

True, it is an elementary principle of corporate law that “incorporation’s basic purpose is to create a distinct legal entity, with legal rights, obligations, powers, and privileges different from those of the natural individuals who created it, who own it, or whom it employs.” Cedric Kushner Promotions, Ltd. v. King, 533 U.S. 158, 163, 121 S.Ct. 2087, 150 L.Ed.2d 198 (2001). And as part of that fiction, shareholders forgo certain rights pertaining to the corporation. See Grote v. Sebelius, 708 F.3d 850, 858 (7th Cir.2013) (Rovner, J., dissenting). But we cannot simply stop there.

Shareholders make such a sacrifice because the corporation can generally exercise some analogue of the forgone right. As a corporation is “capable of making and executing contracts, possessing and owning real and personal property in its own name, suing and being sued,” a shareholder cannot expect to exercise the right to take these actions in his or her personal capacity. See 1 W. Fletcher Cyclopedia of the Law of Corporations § 25 (2006). This is no less true with constitutional rights. See Franks v. Rankin, Nos. 11AP-934, 11AP-962, 2012 WL 1531031, at *10 (Ohio Ct.App.

May 1, 2012) (rejecting a shareholder’s due process claim brought on behalf of the corporation). Mindful of these principles, consider the ramifications of the government’s argument. It contends free exercise is an individual right. If the Gilardis had run their businesses as sole proprietorships, they would presumably have a viable RFRA claim under the government’s theory. Cf. Braunfeld, 366 U.S. at 601, 81 S.Ct. 1144 (describing individual merchants who challenged a Sunday closing law under the Free Exercise Clause).

But the government, relying on what is perhaps an incomplete understanding of corporate law, argues the Gilardis lose the ability to make such a claim by taking advantage of state incorporation law. And as a corollary to the government’s expansive theory, the party being regulated — the corporation— cannot make a free-exercise claim, as it is not an individual capable of exercising religion. So, in the government’s view, there is no corporate analogue, and the individual right disappears into the ether. This

interpretation is perplexing and troubling. It is perplexing because we do not believe Congress intended important statutory rights to turn on the manner in which an individual operates his businesses. The government’s logic is also quite troubling because it would eventually reach First Amendment free-exercise cases. The same language, “exercise” “of religion,” appears both in the Constitution and RFRA. Compare U.S. Const. Amend.

I (“Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof....”), with 42 U.S.C. § 2000bb-1(a) (“Government shall not burden a person’s exercise of religion ____”). Thus, if the government is correct, the price of incorporation is not only the loss of RFRA’s statutory free-exercise right, but the constitutional one as well. And that would create a risk of an unconstitutional condition in future cases. See Perry v.

Sindermann, 408 U.S. 593, 597, 92 S.Ct. 2694, 33 L.Ed.2d 570 (1972) (“[T]his Court has made clear that even though a person has no ‘right’ to a valuable governmental benefit and even though the government may deny him the benefit for any number of reasons, there are some reasons upon which the government may not rely. It may not deny a benefit to a person on the basis that infringes on his constitutionally protected interests (emphasis added)). A parade of horribles will descend upon us, the government exclaims, if religious beliefs could serve as a private veto for the contraceptive mandate.

Hyperbole aside, we note it was Congress, and not the courts, that allowed for an individual’s religious conscience to prevail over substantially burdensome federal regulation. In Gonzales v. O Centro Espirita Beneficente Uniao do Vegetal, 546 U.S. 418, 126 S.Ct. 1211, 163 L.Ed.2d 1017 (2006), the Supreme Court provided an apt response: The Government’s argument echoes 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.

But RFRA operates by mandating consideration, under the compelling interest test, of exceptions to “rule[s] of general applicability.” 42 U.S.C. § 2000bb-l(a). Congress determined that the legislated test [of RFRA] “is a workable test for striking sensible balances between religious liberty and competing governmental interests.” § 2000bb(a)(5). Id. at 436, 126 S.Ct. 1211 (alteration in original). VII As the Gilardis have demonstrated the substantial nature of their burden, we now turn to strict scrutiny, a “searching examination” where the onus is borne exclusively by the government. Fisher v.

Univ. of Tex., — U.S. -, 133 S.Ct. 2411, 2419, 186 L.Ed.2d 474 (2013); see also 42 U.S.C. § 2000bb-1(b). “[U]nless the government demonstrates a compelling governmental interest, and uses the least restrictive means of furthering that interest,” the mandate must be set aside. See Holy Land, 333 F.3d at 166 (internal quotation marks omitted); see also 42 U.S.C. § 2000bb. While “strict scrutiny must not be strict in theory, but fatal in fact,” neither should it be “strict in theory but feeble in fact.” Fisher, 133 S.Ct. at 2421 (internal quotation marks omitted).

A It is difficult to divine precisely what makes an interest “compelling,” but a few reliable metrics exist. The interest cannot be “broadly formulated” — the test demands particularity. See O Centro, 546 U.S. at 431, 126 S.Ct. 1211 (citing Yoder, 406 U.S. at 213, 221, 92 S.Ct. 1526). The “compelling” nature of the interest is contingent on its context. See id. (citing Grutter v. Bollinger, 539 U.S. 306, 327, 123 S.Ct. 2325, 156 L.Ed.2d 304 (2003); Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 228, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995)).

And the interest must be “of the highest order,” Yoder, 406 U.S. at 215, 92 S.Ct. 1526, meaning it cannot leave “appreciable damage to [a] supposedly vital interest unprohibited,” Lukumi, 508 U.S. at 547, 113 S.Ct. 2217 (quoting Florida Star v. B.J.F., 491 U.S. 524, 541-42, 109 S.Ct. 2603, 105 L.Ed.2d 443 (1989) (Scalia, J., concurring in part and concurring in the judgment)). The government cites several concerns to bolster its claim that the contraceptive mandate serves a compelling interest (or interests), but its recitation is sketchy and highly abstract.

Perhaps the government thought it best to focus on justiciability, hoping its ipse dixit would be sufficient to survive strict scrutiny. After all, if no one has standing to object, the state avoids the searching inquiry into its means. Here, the articulated concerns range from “safeguarding the public health” to “protecting a woman’s compelling interest in autonomy” and promoting gender equality. But the government does little to demonstrate a nexus between this array of issues and the mandate.

For example, as a standalone principle, “safeguarding the public health” seems too broadly formulated to satisfy the compelling interest test. It has been used to justify all manner of government regulations in other contexts. See Roe v. Wade, 410 U.S. 113, 154, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973) (abortion laws); Loxley v. Chesapeake Hosp. Auth., No. 97-2539, 1998 WL 827285, at *4 (4th Cir. Dec. 1, 1998) (competence of medical personnel); Dunagin v. City of Oxford, 718 F.2d 738, 747 (5th Cir.1983) (en banc) (liquor advertisement rules).

And here, the government relies on the broad sweep of that interest once more, citing Mead v. Holder, 766 F.Supp.2d 16, 43 (D.D.C.2011), an individual-mandate case in which a district court found the public health interest sufficient. But the invocation of the interest in Mead seems empty, reflexive, and talismanic. The government cites Mead as if to say, “once a compelling interest, always a compelling interest.” It fails to recognize that “safeguarding the public health” is such a capacious formula that it requires close scrutiny of the asserted harm. See O Centro, 546 U.S. at 431, 126 S.Ct. 1211.

We cannot be satisfied with the government’s representation as to the compelling nature of the interest simply because other courts have reached that conclusion in the generality of cases. See Yoder, 406 U.S. at 221, 92 S.Ct. 1526. The nebulousness of the government’s interest, however, prevents us from engaging in the type of exacting scrutiny warranted here. What exactly is the government trying to ameliorate? Is it the integrity of “the health and insurance markets”? Surely, that cannot be the answer; the comprehensive sweep of the Affordable Care Act will remain intact with or without the mandate.

Or is it a need to provide greater access to contraceptive care? If so, as we note below, the reasons underpinning that need are tenuous at best. If we are to assess whether an exemption for the Gilardis would pose an “impediment to [a governmental] objective[],” we must first be able to discern what that objective is. See id. at 221, 236, 92 S.Ct. 1526. Simply reciting Mead is not enough. The government’s invocation of a “woman’s compelling interest in autonomy” is even less robust. The wording is telling. It implies autonomy is not the state’s interest to assert.

Nevertheless, the government, quoting Eisenstadt v. Baird, 405 U.S. 438, 92 S.Ct. 1029, 31 L.Ed.2d 349 (1972), claims the mandate protects a woman’s ability to decide “whether to bear or beget a child.” See id. at 453, 92 S.Ct. 1029. Our difficulty in accepting the government’s rationale stems from looking at the Eisenstadt quote in its entirety: “If the right of privacy means anything, it is the right of the individual, married or single, to be free from unwarranted governmental intrusion into matter so fundamentally affecting a person as the decision to bear or beget a child.” Id. (emphasis added).

Regardless of what this observation means for us today, it is clear the government has failed to demonstrate how such a right&wkey;whether described as noninterference, privacy, or autonomy — can extend to the compelled subsidization of a woman’s procreative practices. Again, our searching examination is impossible unless the government describes its purposes with precision. As with Mead, simply invoking Eisenstadt is not enough.

Equally unconvincing is the government’s assertion that the mandate averts “negative health consequences for both the woman and the developing fetus.” From the outset, we note the science is debatable and may actually undermine the government’s cause. For the potential mother, as one amicus notes, the World Health Organization classifies certain oral contraceptives as carcinogens, marked by an increased risk for breast, cervical, and liver cancers. Br. of the Breast Cancer Prevention Institute, at 8-9.

On the other hand, the contraceptives at issue have been approved by the Food and Drug Administration, supported by research touting their benefits. See Op. of Edwards, J., at 1240. This tug-of-war gives us pause because the government has neither acknowledged nor resolved these contradictory claims. Even giving the government the benefit of the doubt, the health concerns underpinning the mandate can be variously described as legitimate, substantial, perhaps even important, but it does not rank as compelling, and that makes all the difference. Cf. Hutchins v.

District of Columbia, 188 F.3d 531, 541 (D.C.Cir.1999) (en banc). Caselaw concerning the analogous context of abortion is particularly illuminating in this regard. Time and again, the government’s interest in such cases has been deemed legitimate and substantial. See Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 846, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992); see also Gonzales v. Carhart, 550 U.S. 124, 145, 127 S.Ct. 1610, 167 L.Ed.2d 480 (2007). But it has never been compelling.

See Casey, 505 U.S. at 932, 112 S.Ct. 2791 (Blackmun, J., concurring in part and dissenting in part) (“[W]hile a State has ‘legitimate interests from the outset of the pregnancy in protecting the health of the woman and the life of the fetus that may become a child,’ legitimate interests are not enough. To overcome the burden of strict scrutiny, the interests must be compelling.”). And we have no reason to believe otherwise here.

While we do not exclude the possibility that the state’s interest in safeguarding maternal or fetal health sometimes may be compelling, we cannot draw such a conclusion in this particular context. See O Centro, 546 U.S. at 431, 126 S.Ct. 1211. Finally, we note “gender equality” is a bit of a misnomer; perhaps the government labeled it as such for the veneer of constitutional importance attached to the term. More accurately described, the interest at issue is resource parity — which, in the analogous abortion context, the Supreme Court has rejected as both a fundamental right and as an equal-protection issue.

See Harris, 448 U.S. at 317-18, 100 S.Ct. 2671 (“Although the liberty protected by the Due Process Clause affords protection against unwarranted government interference with freedom of choice in the context of certain personal decisions, it does not confer an entitlement to such funds as may be necessary to realize all the advantages of that freedom.”); Maher v. Roe, 432 U.S. 464, 471, 97 S.Ct. 2376, 53 L.Ed.2d 484 (1977) (“But this Court has never held that financial need alone identifies a suspect class for purposes of equal protection analysis.”). The government cites Roberts v. U.S.

Jaycees, 468 U.S. 609, 104 S.Ct. 3244, 82 L.Ed.2d 462 (1984), to advance its “gender equality” interest. There, the Court observed that “[assuring women equal access to such goods, privileges, and advantages clearly furthers compelling state interests.” Id. at 626, 104 S.Ct. 3244. But when that observation is put into context, it fails to support the government’s case. U.S. Jaycees concerned an organization that had shut women out entirely from a superior class of membership; it did not involve disparate membership fees or any resource-parity issue that may sustain the government’s argument.

See id. at 613, 628, 104 S.Ct. 3244. This case is quite different. Beyond the question of access, it is difficult for the government to suggest the interests of women are monolithic, and unlike U.S. Jaycees, the government’s proposed solution clearly impinges on other core prerogatives. B Let us assume, however, the government has a compelling interest. Even then, we cannot see how the mandate is “the least restrictive means of furthering that ... interest.” See 42 U.S.C. § 2000bb-l. It suffers from two flaws that cannot be overcome.

First, there are viable alternatives — presented by the Gilardis and others — that would achieve the substantive goals of the mandate while being sufficiently accommodative of religious exercise. See Appellants’ Br. at 61; see also Conestoga Wood, 724 F.3d at 414-15 (Jordan, J., dissenting). The government could defeat these alternatives by proving they would “present an administrative problem of such magnitude, or ... afford the exempted class so great a competitive advantage, that such a requirement would ... render[ ] the entire statutory scheme unworkable.” Sherbert, 374 U.S. at 408-09, 83 S.Ct. 1790.

But it has made no such case; for all we know, a broader religious exemption would have so little impact on so small a group of employees that the argument cannot be made. Moreover, the mandate is self-defeating. When a government regulation “fail[s] to prohibit nonreligious conduct that endangers [its asserted] interests in a similar or greater degree” than the regulated conduct, it is underinclusive by design. See Lukumi, 508 U.S. at 543, 113 S.Ct. 2217.

And that underinclusiveness can suggest an inability to meet the narrow-tailoring requirement, as it raises serious questions about the efficacy and asserted interests served by the regulation. In this case, small businesses, businesses with grandfathered plans (albeit temporarily), and an array of other employers are exempt either from the mandate itself or from the entire scheme of the Affordable Care Act. Therefore, the mandate is unquestionably underinclusive.

See Hobby Lobby, 723 F.3d at 1143; see also Conestoga Wood, 724 F.3d at 414 (Jordan, J., dissenting) (“It cannot legitimately be said to vindicate a compelling governmental interest because the government has already exempted from its reach grandfathered plans, employers with under 50 employees, and what it defines as ‘religious employers’, thus voluntarily allowing millions upon millions of people — by some estimates 190 million — to be covered by insurance plans that do not satisfy the supposedly vital interest of providing the public with free contraceptives.” (internal citations omitted)). A word on Lee.

We would be remiss if we omitted this observation by the Court: When followers of a particular sect enter into commercial activity as a matter of choice, the limits they accept on their own conduct as a matter of conscience and faith are not to be superimposed on the statutory schemes which are binding on others in that activity. 455 U.S. at 261, 102 S.Ct. 1051. The government understands this quote to foreclose free-exercise claims by employers like the Gilardis. But once again, context matters. We mention Lee in our narrow-tailoring discussion because that is where it belongs.

The Court made the statement quoted above while evaluating whether the “limitation on religious liberty ... [was] essential to accomplish an overriding governmental interest.” Id. at 257, 102 S.Ct. 1051. In engaging in that inquiry, the Court examined “whether accommodating the Amish belief [would] unduly interfere with fulfillment of the governmental interest [of assuring contribution to the Social Security scheme].” Id. at 259, 102 S.Ct. 1051.

Lee was a rare case in which the government fended off a strict-scrutiny challenge by proving exemptions would “present an administrative problem of such magnitude ... that such a requirement would have rendered the entire statutory scheme unworkable.” Sherbert, 374 U.S. at 408-09, 83 S.Ct. 1790; see Lee, 455 U.S. at 258, 102 S.Ct. 1051 (“Moreover, a comprehensive national social security system providing for voluntary participation would be almost a contradiction in terms and difficult, if not impossible, to administer.”).

Proving the incompatibility of the requested religious exemption was necessary to prove that the government had employed the least-restrictive means. See Lee, 455 U.S. at 259-60, 102 S.Ct. 1051 (“Unlike the situation presented in [Yoder], it would be difficult to accommodate the comprehensive social security system with myriad exceptions flowing from a wide variety of religious beliefs.”). Congress had carefully determined the breaking point of Social Security — any uncontemplated exemptions could render the statutory scheme unworkable. See id. at 258, 102 S.Ct. 1051.

In contrast, the government has not proven — nay, even asserted — statutory unworkability here. Its “private veto” concern is somewhat on point, but without substance or substantiation, is nowhere near enough. If we found narrow tailoring satisfied by mere ipse dixit, the strict-scrutiny inquiry would become feeble indeed.

And unlike Lee, where the government successfully asserted the Social Security system required every contribution that Congress did not otherwise exempt, there is nothing to suggest the preventive-care statute would become unworkable if employers objecting on religious grounds could opt out of one part of a comprehensive coverage requirement. The Gilardis’ employees will still receive an array of services such as well-woman visits, gestational-diabetes screenings, HPV testing, counseling for sexually-transmitted infections, support for breastfeeding, and counseling for interpersonal and domestic violence.

See Women’s Preventive Services Guidelines, Health Res. & Servs. Admin., http://www.hrsa.gov/womensguidelines/. The provision of these services — even without the contraceptive mandate- — by and large fulfills the statutory command for insurers to provide gender-specific preventive care. At the very least, the statutory scheme will not go to pieces.

VIII We conclude the district court erred in denying a preliminary injunction for the Gilardis on the grounds that their case was unlikely to succeed on the merits; therefore, we reverse the district court’s denial of a preliminary injunction for the individual owners. Because the court premised its decision entirely on a question of law, we must remand for consideration of the other preliminary-injunction factors. See Chaplaincy of Full Gospel Churches, 454 F.3d at 304. We affirm the district court’s denial of a preliminary injunction with respect to the Freshway companies. So ordered. .

For ease of reference, we will refer to this provision as “the contraceptive mandate.” . The Gilardis could have dropped their healthcare coverage altogether, but they regard this option as morally unthinkable. See J.A. at 41 ¶ 10; J.A. at 52 ¶ 10. . We agree with Judge Edwards that the Gilardis’

Article III standing is indisputable. See Op. of Edwards, J., at 1227-28. . We assume, without deciding, that Congress did not intend to abrogate the prudential-standing requirement in enacting RFRA. We share Judge Edwards’ concerns about whether prudential-standing principles apply to RFRA challenges and whether the shareholder-standing rule is part of the prudential-standing equation. See Op. of Edwards, J., at 1229-31. But it would be imprudent to decide these questions without the benefit of full briefing on this issue, especially as the Gilardis can easily surmount the shareholder-standing hurdle. .

Our conclusion is buttressed by other considerations. First, Ohio caselaw does not treat the derivative-action rule as an unyielding one; to the contrary, some flexibility has been shown when it comes to close corporations such as the Freshway companies. See, e.g., Yackel v. Kay, 95 Ohio App.3d 472, 642 N.E.2d 1107, 1109-10 (1994). Moreover, none of the principles underlying the shareholder-standing rule is offended by allowing the Gilardis' suit to proceed — there is no danger of multiple lawsuits, and no creditor or shareholder interests will be compromised as a result of their RFRA challenge. See 12B W.

Fletcher Cyclopedia of the Law of Corporations § 5911.50 (2006). We recognize that shareholders of large public corporations will be subject to different constraints and will likely find the burden threshold insuperable. . Much ink has already been spilled on how the government's interests leave “appreciable damage” unprohibited. See Conestoga Wood, 724 F.3d at 413-14 (Jordan J., dissenting); Hobby Lobby, 723 F.3d at 1143-44. We share these concerns, but need not repeat them here. . See A. Raymond Randolph, Before Roe v. Wade: Judge Friendly’s Draft Abortion Opinion, 29 Harv. J.L. & Pub.

Pol’y 1035, 1048 (2006) (describing the transformation of the right described in Griswold and Roe). . Underinclusiveness is generally a relevant consideration of the narrow-tailoring inquiry. See Lukumi, 508 U.S. at 546, 113 S.Ct. 2217 ("First, even were the governmental interests compelling, the ordinances are not drawn in narrow terms to accomplish those interests. As we have discussed, all four ordinances are overbroad or underinclusive in substantial respects.” (emphasis added)).

We recognize the considerable overlap between narrow-tailoring under-inclusiveness and the "appreciable damage” component of the compelling-interest prong. Nevertheless, we need not reconcile the distinctions between the two— under both formulations, the government falls short. See supra at 1219 n. 6. . Lee also reinforces our doubts about whether the government’s asserted interests are sufficiently compelling.

The proper functioning of the tax system was perceived as an interest of "a high order,” see 455 U.S. at 260, 102 S.Ct. 1051, akin to interests of "public safety, peace, or order” that justified government intrusions on religious exercise in the past, see Sherbert, 374 U.S. at 403, 83 S.Ct. 1790. Given our many doubts about the interests posited, we are skeptical about whether the mandate is designed to address "the gravest abuses, endangering paramount interest.” See Sherbert, 374 U.S. at 406, 83 S.Ct. 1790.