A “VERY SPECIFIC” HOLDING: ANALYZING THE EFFECT OF HOBBY LOBBY ON RELIGIOUS LIBERTY CHALLENGES TO HOUSING DISCRIMINATION LAWS

A “VERY SPECIFIC” HOLDING: ANALYZING THE EFFECT OF HOBBY LOBBY ON RELIGIOUS LIBERTY CHALLENGES TO HOUSING DISCRIMINATION LAWS

The Supreme Court’s 2014 decision in Hobby Lobby v. Burwell sent shockwaves through the legal community. While many praised its broad interpretation of the Religious Freedom Restoration Act (RFRA) as a milestone in protecting religious liberty, others expressed concern that it would essentially turn RFRA and similar legislation on the state level into a “license to discriminate” against LGBT individuals in areas such as employment, public accommoda­tions, and housing. This Note focuses on housing, analyzing the possibility that landlords would use Hobby Lobby’s broad view of religious liberty law to claim religious exemptions from laws forbidding housing discrimination against LGBT people. The Note applies Hobby Lobby to a hypothetical claim by a landlord, exploring possible argu­ments from both sides on the question of whether a housing antidiscrimination law creates a “substantial burden” on the landlord’s belief, whether the government has a “compelling interest” in preventing housing discrimination against LGBT people, and whether an antidiscrimination law is the “least restrictive means” of fulfilling this interest. The Note concludes that, even under Hobby Lobby’s broad reading of RFRA, such a claim for a religious exemption would not be successful.

Introduction

On June 30, 2014, the U.S. Supreme Court handed down its decision in Burwell v. Hobby Lobby Stores, Inc., ruling that the Patient Protection and Affordable Care Act’s mandate that employers provide contraceptive coverage unlawfully burdened the plaintiff employers’ free exercise of religion. 1 134 S. Ct. 2751, 2785 (2014) (“The contraceptive mandate, as applied to closely held corporations, violates RFRA [Religious Freedom Restoration Act].”). In applying the Religious Freedom Restoration Act (RFRA), the federal law governing religious liberty rights, the Court held that the contraception mandate substantially burdened the plaintiff companies’ religious beliefs by forcing them to pay for contraceptive drugs the plaintiffs’ owners sincerely believed were immoral. 2 Id. at 2779 (“Because the contraceptive mandate forces [plaintiffs] to pay an enormous sum of money . . . if they insist on providing insurance coverage in accordance with their religious beliefs, the mandate clearly imposes a substantial burden on those beliefs.”). Further, while the government may very well have had a compelling interest in providing access to contraception, the mandate was not the “least restrictive means” of furthering that interest, as there were other possible poli­cies that could further the exact same goal. 3 Id. at 2780. Specifically, the government could require insurance companies to provide contraception if an em­ployer had a religious objection (a policy already in place for religious nonprofits) or the government could provide contraception directly. 4 See id. at 2780–83.

The Hobby Lobby Court took great pains to emphasize that its holding was limited to the specific facts before it and that the holding would not allow individuals to opt out of other types of laws, specifically laws banning hiring discrimination on the basis of race, by claiming a religious exemption. 5 See id. at 2783 (stating Hobby Lobby’s holding cannot be used to challenge racial discrimination laws because “[g]overnment has a compelling interest in providing an equal opportunity to participate in the workforce without regard to race, and prohibitions on racial discrimination are precisely tailored to achieve that critical goal”). Indeed, on the narrow set of facts before the Court, and given the apparent existence of a viable and obvious alterna­tive mechanism to achieve the goal of providing women access to contraception, the Hobby Lobby decision seems reasonable, striking a delicate balance between respecting religious liberty and achieving important social goals. However, the analytical framework the Court laid out may have much broader implications than Justice Alito’s assurances suggest. As Justice Ginsburg wrote in her dissent:

[T]he Court holds that commercial enterprises, including corporations . . . can opt out of any law (saving only tax laws) they judge incompatible with their sincerely held religious beliefs [,] . . . at least when there is a “less restrictive alternative.” And such an alternative, the Court suggests, there always will be whenever, in lieu of tolling an enterprise claiming a religion-based exemption, the government, i.e., the general public, can pick up the tab. 6 Id. at 2787 (Ginsburg, J., dissenting).

While this may be a bit of a hyperbole, one can certainly envision individuals and corporations petitioning for exemptions from numerous types of laws that might burden their religious beliefs, including laws designed to protect LGBT individuals from discrimination. In addition, while RFRA only applies to federal laws, many states have their own versions of RFRA or interpret their state free exercise clauses with RFRA-like tests. 7 At least thirty-two states have adopted these protections. See 2015 State Religious Freedom Restoration Legislation, Nat’l Conf. St. Legislatures (Sept. 3, 2015), http:// www.ncsl.org/research/civil-and-criminal-justice/2015-state-rfra-legislation.aspx [http:// perma.cc/M9KY-9HZX] [hereinafter 2015 State Legislation] (providing list of states proposing to add or amend state religious freedom laws). Compare State Religious Freedom Restoration Acts, Nat’l Conf. of St. Legislatures (Oct. 15, 2015), http://www. ncsl.org/research/civil-and-criminal-justice/state-rfra-statutes.aspx [http:// perma.cc/ 23AH-XCNA] (providing map showing twenty-one states have enacted state RFRA legislation), with Juliet Eilperin, 31 States Have Heightened Religious Freedom Protections, Wash. Post (Mar. 1, 2014), http://www.washingtonpost.com/blogs/the-fix/wp/2014 /03/01/where-in-the-u-s-are-there-heightened-protections-for-religious-freedom/ [http:// perma.cc/QP9B-3M65] (providing map showing at least eleven states that have not adopted legislation but whose courts apply RFRA-like test to state free exercise claims). Moreover, in 2015, an additional twelve states proposed legislation that would create new RFRA laws, although these proposals are still pending. Therefore, it is conceivable that both state and federal courts struggling to decide religious-exemption cases may use Hobby Lobby’s language as guidance. 8 Indeed, state courts often use Supreme Court jurisprudence as guidance in analyzing RFRA-like claims, even if they are not directly applying RFRA. See, e.g., Swanner v. Anchorage Equal Rights Comm’n, 874 P.2d 274, 282–83 (Alaska 1994) (applying Supreme Court cases in analyzing claim under state free exercise clause).

In the wake of Hobby Lobby, scholars have speculated about the possibility that individuals or companies could use Hobby Lobby’s rationale to gain religious exemptions allowing them to discriminate against LGBT individuals in areas such as employment and public accommodations. 9 See, e.g., Mary Anne Case, Legal Protections for the “Personal Best” of Each Employee: Title VII’s Prohibition on Sex Discrimination, the Legacy of Price Waterhouse v. Hopkins, and the Prospect of ENDA, 66 Stan. L. Rev. 1333, 1376 (2014) (anticipating religious challenges to LGBT-targeted employment discrimination laws post–Hobby Lobby); Ira C. Lupu, Hobby Lobby and the Dubious Enterprise of Religious Exemptions, 38 Harv. J.L. & Gender 35, 92–100 (2015) (analyzing possibility of religious exemptions from employment and public-accommodations discrimination laws in light of Hobby Lobby). This Note addresses another important area in which LGBT individuals face discrimination: housing. Courts have not yet addressed religious liberty claims by landlords seeking to discriminate against LGBT indi­viduals. However, several states addressed a similar question in the 1990s, when a number of landlords sought to discriminate against un­married, cohabitating heterosexual couples on the grounds that renting to couples “living in sin” would violate their religious beliefs. 10 See infra section I.C (discussing cases in which landlords sought religious exemptions to discriminate against unmarried couples). Results in those states were mixed, with some finding that the anti-marital-status discrimination laws did not substantially burden landlords’ beliefs, some finding that they did burden those beliefs but that the state’s interest in preventing discrimination was sufficiently compelling to override this burden, and some finding that neither of these were true and that the landlords deserved an exemption. 11 See, e.g., Swanner, 874 P.2d at 284 (holding compelling interest in preventing housing discrimination overrode substantial burden on landlord’s beliefs); Smith v. Fair Emp’t & Hous. Comm’n, 913 P.2d 909, 929 (Cal. 1996) (holding housing antidiscrimina­tion law did not substantially burden landlord’s religious beliefs but not deciding compelling interest question); Donahue v. Fair Emp’t & Hous. Comm’n, 2 Cal. Rptr. 2d 32, 46 (Dist. Ct. App. 1993) (holding antidiscrimination law was substantial burden on landlord and preventing housing discrimination against unmarried couples was not sufficiently “compelling interest” to justify this burden); Attorney Gen. v. Desilets, 636 N.E.2d 233, 241 (Mass. 1994) (denying summary judgment on grounds that antidiscrimination law was “substantial burden” and government needed more evidence to prove preventing housing discrimination against unmarried couples was “compelling interest”).

Given the disagreement among courts on the housing discrimina­tion issue, and a recent Supreme Court decision that, as this Note will show, likely decimated some of the best arguments in favor of the tenants in these situations, 12 See infra section II.A (analyzing impact of Hobby Lobby on arguments employed by state courts in previous housing discrimination cases). the possibility that landlords will invoke religion to seek license to discriminate against LGBT people is very real. Further­more, the fact that courts have grappled with similar questions in the unmarried-cohabitant context makes housing distinct from other areas that have been more thoroughly explored, such as employment. While it is difficult to see a less restrictive means of preventing employment dis­crimination than passing an employment antidiscrimination law, the housing context contains ample room for RFRA arguments, as courts have recognized in cases involving unmarried cohabitants. Therefore, this Note seeks to contribute to the literature by discussing the unique area of housing, which has heretofore been underexplored despite being one of the areas where RFRA-like arguments are possibly most likely to succeed.

This Note analyzes how such a claim by a landlord would fare under the rationale laid out in Hobby Lobby. Part I details the history of religious liberty law in the United States, including the Hobby Lobby decision and a string of state court decisions in cases where landlords sought religious exemptions to discriminate against unmarried couples. Part II discusses how a landlord might use Hobby Lobby in connection with the older housing discrimination cases to argue that an antidiscrimination law imposes a “substantial burden” on the landlord’s religious beliefs and provides possible responses to those arguments. Part III addresses the “compelling interest and least restrictive means” portion of the analysis, concluding that Hobby Lobby does not support religious exemptions from such laws and that a careful RFRA analysis leads to the conclusion that such exemptions should not be granted.

I. The History of Religious Liberty Law in the United States

This Part describes the development of U.S. religious liberty law up to the Hobby Lobby decision in an effort to place Hobby Lobby in its historical context and better evaluate how it shifted from, or built upon, previous understandings of the law. Section I.A discusses the development of religious liberty case law from the earliest cases up to the Supreme Court’s controversial decision in Employment Division v. Smith. Section I.B describes the enactment of the Religious Freedom Restoration Act and the Religious Land Use and Institutionalized Persons Act (RLUIPA), as well as state-level statutes and court decisions designed to provide greater religious liberty protection than the Smith decision allowed. Section I.C examines the post-RFRA history of religious liberty challenges to antidiscrimination provisions in cases involving unmarried couples, a closely analogous situation to LGBT-housing discrimination, to get a sense of how courts might have analyzed a landlord’s challenge to an LGBT antidiscrimination law before Hobby Lobby. Section I.D discusses the Hobby Lobby decision itself and how it both built off and departed from previous standards. Finally, section I.E discusses the legal landscape surrounding anti-LGBT housing discrimination and how it runs up against religious liberty law.

A. The Development of Religious Liberty Law

The Free Exercise Clause of the First Amendment to the U.S. Constitution reads, “Congress shall make no law . . . prohibiting the free exercise [of religion].” 13 U.S. Const. amend. I. The Supreme Court was first called to consider this text in 1878, in the context of a constitutional challenge to a federal antibigamy law by a Mormon whose religious practices mandated polygamous marriage. 14 Reynolds v. United States, 98 U.S. 145 (1878). The Court upheld the law and refused to grant the appellant an exemption, holding that while Congress could not pass laws interfering with religious belief, it could outlaw religious practices that were deemed “in violation of social duties or subversive of good order.” 15 Id. at 164. In the Court’s words, “To permit [a religious exemption] would be to make the professed doctrines of religious belief superior to the law of the land, and in effect to permit every citizen to become a law unto himself. Government could exist only in name under such circumstances.” 16 Id. at 167.

For nearly a century after this case, the Court generally read the Free Exercise Clause narrowly, as a clause designed to protect individuals from religion-based discrimination rather than give religion-based conduct heightened protection in the secular world. The Court would regularly strike down laws that directly or indirectly penalized individuals for their religious beliefs 17 See, e.g., Fowler v. Rhode Island, 345 U.S. 67, 69–70 (1953) (holding ordinance preventing Jehovah’s Witness minister from giving speech in public park but not prohibiting other types of church services was unconstitutional); W. Va. State Bd. of Educ. v. Barnette, 319 U.S. 624, 642 (1943) (striking down law requiring children to salute American flag in school under penalty of expulsion as applied against Jehovah’s Witnesses whose beliefs prohibit this); Cantwell v. Connecticut, 310 U.S. 296, 307 (1940) (overturning ordinance outlawing soliciting money for religious purposes without license, which impermissibly interfered with Jehovah’s Witnesses’ primary method of evangelization). or limited the right of religious people to evangelize their faith. 18 See, e.g., Follet v. Town of McCormick, 321 U.S. 573, 576 (1944) (striking down ordinance requiring Jehovah’s Witnesses to pay business-license tax for disseminating religious books and pamphlets); Murdock v. Pennsylvania, 319 U.S. 105, 110–11 (1943) (same); Cantwell, 310 U.S. at 307 (same). In contrast, the Court would uphold laws regulating or prohibiting conduct in the secular realm that was considered against the public interest, even when such laws placed a burden on a practice or behavior based in religion. 19 See, e.g., Braunfield v. Brown, 366 U.S. 599, 603, 605 (1961) (upholding law forbidding retail sale on Sundays as applied against Orthodox Jews whose religion required rest on Saturday, as law did not “make criminal the holding of any religious belief or opinion” but simply made “practice of their religious beliefs more expensive”); Cleveland v. United States, 329 U.S. 14, 20 (1946) (upholding anti-human-trafficking law against Mormon who imported foreign woman to make her one of his plural wives, finding Mormon practice of polygamy “immoral purpose” under statute); Prince v. Massachusetts, 321 U.S. 158, 169–70 (1944) (upholding child labor law forbidding minors from selling religious pamphlets as applied to Jehovah’s Witnesses).

However, this changed in 1963 with the case of Sherbert v. Verner, 20 374 U.S. 398 (1963). in which the Court for the first time applied a balancing test to a claim for a religious exemption from a secular law, specifically a law denying unemployment benefits as a consequence of the plaintiff’s refusal to work on her holy day. 21 The case involved a Seventh-day Adventist who was fired by her employer for refusing to work on Saturday, her Sabbath day, and was subsequently denied benefits under her state’s Unemployment Compensation Act because she “failed . . . to accept available suitable work when offered” to her. Id. at 399–401. She alleged that this statute impermissibly burdened her free exercise of religion by interfering with her ability to observe her Sabbath day. Id. The Court held that a burden on an individual’s exercise of religion must be justified by a “compelling state interest in the regulation of a subject within the State’s constitutional power to regulate.” 22 Id. at 403. Applying this test to the case at bar, the Court decided that the law placed a burden on the plaintiff’s right to exercise her religion and was not justified by any compelling interest. Id. at 408–09. The Court said that, even if the law served a compelling interest, it would “plainly be incumbent upon the appellees to demon­strate that no alternative forms of regulation would combat such abuses without infringing First Amendment rights.” 23 Id. at 407. This can be seen as the earliest articulation of the “least restrictive means” prong of the analysis in the religious-freedom context, which later became codified in RFRA. See infra section I.B (describing RFRA test). The Court also articulated important limits on its holding, suggesting that a religious belief that infringes on another person’s religious liberties, or that makes the believer a “nonproductive member of society,” may not be given similar deference. 24 Sherbet, 374 U.S. at 409–10.

In articulating its balancing test, the Sherbert Court did not purport to be breaking new ground; indeed, it merely applied the same strict scrutiny analysis that had been applied in challenges to legislation that discriminated against racial minorities. 25 See id. at 403 (citing NAACP v. Button, 371 U.S. 415, 438 (1963)) (articulating balancing test). However, it was clear that secular laws that were facially neutral but indirectly burdened a particular religion would no longer be given deference by the Court as they had since Reynolds. 26 See supra notes 13–18 and accompanying text (discussing Reynolds and its progeny). Rather, the Court would carefully weigh the burden on the petitioner’s religion against the policy justification for the law to determine whether interference with the petitioner’s religious practice was justified. Later, in Wisconsin v. Yoder, 27 406 U.S. 205 (1972). the Court explicitly made clear that this balancing test applied not only to laws targeted at certain religious practices but to laws of “general applicability” that were facially neutral toward religion with the incidental effect of burdening a religious practice. 28 Id. at 220. The Court held that a compulsory formal-education law impermissibly burdened Amish parents’ free exercise of their religion, which required them to educate their children according to their own beliefs and values rather than those of the formal education system. Id. at 234. While the state obviously had a compelling interest in ensur­ing the education of its children, this interest was less compelling than the Amish’s right to direct the upbringing of their children according to their religious principles. Id. Because the plaintiffs had demonstrated that the Amish education system adequately served the purpose of ensuring Amish children were properly educated, the compulsory education law did not meet the “compelling interest” test. Id. In the Court’s words, “only those interests of the highest order and those not otherwise served can overbalance legitimate claims to the free exercise of religion.” 29 Id. at 215.

Until 1990, the Court continued to occasionally use this balancing test to protect religious believers from laws that burdened their beliefs. 30 See, e.g., Thomas v. Review Bd. of the Ind. Emp’t Sec. Div., 450 U.S. 707, 720 (1981) (striking down law denying unemployment benefits to Jehovah’s Witness who was fired for his refusal to help create turrets for military tanks). However, despite the bold rhetoric of Sherbert and Yoder, the Court was inconsistent in applying the test, leaving the status of the Sherbert test in a state of ambiguity and confusion. 31 See Robert M. O’Neil, Religious Freedom and Nondiscrimination: State RFRA Laws Versus Civil Rights, 32 U.C. Davis L. Rev. 785, 789 (1999) (“During the quarter century between Sherbert and Smith, the evolution of the free exercise doctrine was in fact somewhat less smooth than one might have expected.”). Compare Bowen v. Roy, 476 U.S. 693, 708 (1986) (declining to apply “compelling interest” test in challenge by Native American plaintiff to law requiring him to obtain social security number for his daughter, holding government could meet its burden by showing program, “neutral and uniform in its application, is a reasonable means of promoting a legitimate public interest”), with United States v. Lee, 455 U.S. 252, 261 (1982) (applying “compelling interest” test to challenge by Amish plaintiff to social security taxes but holding tax system satisfied test and justified substantial burden). Many other cases from this period analyze claims without either mentioning the Verner test or explicitly rejecting it, leaving the state of the law unclear. See, e.g., Lyng v. Nw. Indian Cemetery Protective Ass’n, 485 U.S. 439, 450–51 (1988) (upholding road construction on sacred Native American land because “incidental effects of government programs, which may make it more difficult to practice certain religions . . . [do not] require government to bring forward a compelling justification”); O’Lone v. Estate of Shabazz, 482 U.S. 342, 348–53 (1987) (sustaining prison’s refusal to excuse inmates from work to attend worship services without mentioning Sherbert test or requiring prison officials to show their policy was least restrictive means of accomplishing compelling interest). Regardless of Sherbert’s apparent promise of heightened protection for religious freedom, the vast majority of cases ended with a rejection of the religious objector’s claim. 32 James E. Ryan, Note, Smith and the Religious Freedom Restoration Act: An Iconoclastic Assessment, 78 Va. L. Rev. 1407, 1412 (1992) (“[D]espite the apparent protection afforded claimants by the language of the compelling interest test, courts overwhelmingly sided with the government when applying that test.”).

Finally, in 1990, the Court directly addressed the status of the Verner test in Employment Division v. Smith, 33 Emp’t Div., Dep’t of Human Res. v. Smith, 494 U.S. 872 (1990). which involved a claim by Native Americans who had been denied unemployment compensation after being fired from their job for using an illegal substance in a religious ritual. In ruling against the plaintiffs, the Court expressly rejected the Sherbert test for claims against neutral, generally applicable laws, depart­ing from Yoder. 34 Id. at 885–89; see also supra notes 27–29 and accompanying text (discussing Yoder). The Smith decision can also be seen as a shift from a “constitutional exemption” model, in which the Free Exercise Clause itself mandates that religious people be given exemptions from laws that impose a burden on their beliefs or practices, to a “statutory exemption” model, in which any special religious exemptions to otherwise generally applicable laws must be explicitly codified in the statute. See Eugene Volokh, A Common-Law Model for Religious Exemptions, 46 UCLA L. Rev. 1465, 1467–68 (1999) (distinguish­ing pre-Smith constitutional exemption regime from post-Smith statutory exemption regime). The Court explained that the “compelling interest” test was inappropriate in assessing generally applicable statutes, as the broad diversity of religious beliefs and practices and the high burden on the government set by the “compelling interest” test would open the door to all kinds of religious exemptions from otherwise valid laws, ultimately “courting anarchy.” 35 Smith, 494 U.S. at 888. Such logic reflected the classic concern with religious exemptions first stated in Reynolds, that exemptions would allow “every citizen to become a law unto himself.” Reynolds v. United States, 98 U.S. 145, 167 (1878). Thus, the state of religious freedom law was now clarified: While the law would not permit laws expressly inhibiting or discriminating against a person’s religion, it would permit laws that were generally applicable and facially neutral toward religion, even if those laws placed a significant burden on a religious practice.

B. RFRA and RLUIPA: The Statutory Approach to Religious Liberty Law

The Smith decision sparked bipartisan outrage. Though in reality, religious liberty doctrine had been in flux and the Sherbert test was only occasionally seriously applied, 36 See supra notes 31–32 and accompanying text (discussing ambiguity of status of Sherbert test in years following Sherbert and Yoder). Congress perceived Smith’s explicit renunciation of the Sherbert test as a departure from settled law that required the government to show a compelling interest before burdening an individual’s religion. 37 See 42 U.S.C. § 2000bb(a)(4) (2012) (stating congressional findings that Smith “virtually eliminated the requirement that the government justify burdens on religious exercise imposed by laws neutral toward religion”). Responding to Smith, a unanimous House and a near-unanimous Senate passed the Religious Freedom Restoration Act, 38 Michael W. McConnell, Comment, Institutions and Interpretation: A Critique of City of Boerne v. Flores, 111 Harv. L. Rev. 153, 160 (1997) (“[T]he House of Representatives passed RFRA unanimously and the Senate did so by a vote of 97-3.”). intending to reinstate the test laid out in Sherbert and Yoder for laws facially neutral toward religion. 39 42 U.S.C. § 2000bb(b)(1) (stating RFRA’s purpose is to “restore the compelling interest test as set forth in” Sherbert and Yoder). RFRA provided that the government “shall not substantially burden a person’s exercise of religion even if the burden results from a rule of general applicability,” unless the government can demonstrate that the law “is in furtherance of a compelling governmental interest” and is the “least restrictive means of furthering” that interest. 40 Id. § 2000bb-1.

RFRA was originally intended to apply to state and local governments as well as the federal government. 41 Id. § 2000bb-2(1) (defining “government” to include “United States, a State, or a subdivision of a State”). However, in City of Boerne v. Flores, the Supreme Court held RFRA unconstitutional as ap­plied to state and local governments, as Congress’s power to prevent states from depriving individuals of constitutional rights did not give it the power to alter the meaning of the Free Exercise Clause from the meaning the Court had declared. 42 See U.S. Const. amend. XIV, §§ 1, 5 (giving Congress power to “enforce, by appropriate legislation, the provisions of” Fourteenth Amendment, which specifies “[n]o state shall . . . deprive any person of life, liberty, or property without due process of law”); City of Boerne v. Flores, 521 U.S. 507, 519 (1997) (“Legislation which alters the meaning of the Free Exercise Clause cannot be said to be enforcing the Clause. Congress does not enforce a constitutional right by changing what the right is.”). In light of the Court’s ruling, Congress passed RLUIPA, a more constrained law that applied the RFRA rule to land-use regulations and policies affecting institutionalized individuals. To ensure the law’s constitutionality, RLUIPA’s reach was expressly limited to situations that Congress could claim authority over using its commerce or spending powers. 43 See 42 U.S.C. § 2000cc (applying RFRA test to land-use regulations but limiting it to regulations of programs that “receive[] Federal financial assistance” or that would “affect[] . . . commerce with foreign nations, among the several States, or with Indian tribes”); id. § 2000cc-1 (applying same test for programs in areas with institutionalized individuals). In contrast with RFRA, RLUIPA was ultimately held constitutional as applied to the states. See Cutter v. Wilkinson, 544 U.S. 709, 720 (2005) (holding RLUIPA constitutional as valid exercise of Congress’s power and not in conflict with Establishment Clause).

In addition to the federal RLUIPA, twenty-one states passed their own laws applying the RFRA analysis to state and local government action. 44 See 2015 State Legislation, supra note 7 (noting “[c]urrently, 21 states have Religious Freedom Restoration Acts”). Notably, these statutes differ in how “substantial” the burden on religious exercise must be before heightened scrutiny is triggered. Some, like the federal RFRA, require a substantial burden, while some specify that virtually any burden will require heightened scrutiny. See Natacha Lam, Clash of the Titans: Seeking Guidance for Adjudicating the Conflict Between Equality and Religious Liberty in LGBT Litigation, 23 Tul. J.L. & Sexuality 113, 119 (2014) (describing differences in “burden” standard between state RFRA provisions). Among the states that did not pass RFRAs, thirteen states have interpreted the free exercise provisions in their own constitutions as providing heightened scrutiny for burdens on religious liberty, in contrast with Smith’s anti-religious-exemption interpretation of the federal Free Exercise Clause. 45 See Eilperin, supra note 7 (showing thirteen states without RFRAs whose courts provide heightened religious liberty protection via state constitution); see also Lam, supra note 44, at 118 (describing state courts rejecting Smith standard and interpreting their own free exercise clauses according to Verner balancing test).

In light of both RFRA and RLUIPA at the federal level and the several ways in which heightened religious liberty protection is provided at the state level, many situations in which one might assert a free exercise claim will now be governed by some version of the RFRA compelling interest test. Notably, in the years following RFRA’s passage, the majority of claims under RFRA failed, as courts were consistently hesitant to find a burden substantial enough to justify a religious exemption. 46 See Ira C. Lupu, The Failure of RFRA, 20 U. Ark. Little Rock L.J. 575, 591, 604–17 (1998) [hereinafter Lupu, The Failure of RFRA] (noting 143 of 168 RFRA decisions prior to City of Boerne found against religious objector and majority of cases relied on “substantial burden” prong because of prong’s ambiguity). In addition, since City of Boerne and the subsequent passage of state RFRAs, there have been very few cases explicitly addressing such claims at the state level, creating what one author calls an “imposing and unknown legal specter.” 47 Lam, supra note 44, at 119.

C. The History of Religious Challenges to Housing Antidiscrimination Laws

As of this writing, there have been no cases addressing a landlord’s religious objection to a housing antidiscrimination law related to sexual orientation or gender identity. 48 See Jack M. Battaglia, Religion, Sexual Orientation, and Self-Realization: First Amendment Principles and Anti-Discrimination Laws, 76 U. Det. Mercy L. Rev. 189, 307 (1999) (“No opinions address religious objection to the application of sexual orientation anti-discrimination laws in the context of housing.”). Therefore, how a state court might analyze such a claim remains unclear. However, there is a closely analogous line of cases that can help shed light on the question. In the 1990s, several state courts entertained religious liberty claims by land­lords seeking to discriminate against unmarried, cohabitating heterosex­ual couples. 49 See infra note 50 (listing notable housing discrimination cases). These landlords claimed that laws prohibiting discrimina­tion on the basis of marital status burdened their religious beliefs, as by mandating that the landlords give the couples a place in which to engage in premarital sex, the laws forced the landlords to facilitate sinful behavior in violation of their deeply held religious convictions. 50 See, e.g., Swanner v. Anchorage Equal Rights Comm’n, 874 P.2d 274, 277 (Alaska 1994) (“Under [landlord’s] religious beliefs, even a non-sexual living arrangement by roommates of the opposite sex is immoral and sinful because such an arrangement suggests the appearance of immorality.”); Smith v. Fair Emp’t & Hous. Comm’n, 913 P.2d 909, 912 (Cal. 1996) (“‘Respondent believes that sex outside of marriage is sinful, and that it is a sin for her to rent her units to people who will engage in nonmarital sex on her property.’”); Donahue v. Fair Emp’t & Hous. Comm’n, 2 Cal. Rptr. 2d 32, 38 (Ct. App. 1993) (“[The landlords] are devout Roman Catholics who believe that sexual intercourse outside of marriage is a mortal sin and that to assist or facilitate such behavior also constitutes a sin.”); Attorney Gen. v. Desilets, 636 N.E.2d 233, 234–35 (Mass. 1994) (“The defendants, who are Roman Catholics, believe that they should not facilitate sinful conduct, including fornication.”). Such claims are very similar to a claim that could be brought in a sexual-orientation discrimination case, as the stated “burden” on the landlord’s religious belief would be essentially the same: By being forced to rent to a person or couple who engages in conduct or self-identifies in a way the landlord is morally opposed to, the landlord is being coerced into facilitating their “sin.” 51 See Scott A. Johnson, Note, The Conflict Between Religious Exercise and Efforts to Eradicate Housing Discrimination Against Nontraditional Couples: Should Free Exercise Protect Landlord Bias?, 53 Wash. & Lee L. Rev. 351, 379–80 (1996) (stating “very similar” situation to unmarried cohabitant cases is “refusal of a landlord to rent to a homosexual or lesbian couple on the basis of the landlord’s religious belief that facilitating such relationships is a sin”). An examination of how courts have analyzed these claims in the unmarried-cohabitant context can provide a framework for understanding how a claim in the sexual-orientation context may have come out pre–Hobby Lobby.

Smith v. Fair Employment & Housing Commission, 52 913 P.2d 909. a California Supreme Court case, is an example of a case where a landlord failed to obtain a religious exemption. The court held that the landlord could not gain a religious exemption because the law did not “substantially burden” the landlord’s religious beliefs. The court made three key arguments in support of this holding: Any “burden” on the plaintiff’s religious beliefs resulted from her voluntary participation in the rental market; 53 Id. at 925. The court explicitly distinguished between the case at bar and the unemployment-compensation cases in which religious exemptions had traditionally been granted: “[In unemployment-compensation cases], one can avoid the conflict between the law and one’s beliefs about the Sabbath only by quitting work and foregoing compensation. To do so, however, is not a realistic solution for someone who lives on the wages earned through personal labor.” Id. granting a religious exemption in this case would unduly burden third parties by subjecting potential tenants to discrimination; 54 Id. and simply making religious exercise more expensive, as opposed to outright banning it, did not constitute a “substantial burden.” 55 Id. at 927. Free exercise cases had traditionally held that “mak[ing] religious beliefs more expensive,” as opposed to “mak[ing] criminal the holding of any religious belief or opinion,” is not a substantial burden for free exercise purposes. Braunfield v. Brown, 366 U.S. 599, 603, 605 (1961) (upholding law requiring businesses to close on Sunday as applied to Orthodox Jewish retailers, whose religion mandated day of rest on Saturday, even though obeying both law and religious beliefs would force them to close all weekend and harm their business). As the court found that the landlord’s belief was not substantially burdened, it declined to reach the question of whether preventing housing discrimination against unmarried couples was a “compelling government interest.” 56 Smith, 913 P.2d at 929. In dissent, Judge Joyce Kennard argued that the antidiscrimination law did substantially burden the landlord’s religious beliefs. Judge Kennard argued that a substantial burden exists wherever:
“(1) [A] religious adherent engages in a particular activity; (2) a governmental command relating to the activity conflicts with the adherent’s religious beliefs concerning the activity; (3) the conflict is irreconcilable (that is, to satisfy the governmental command the adherent must either abandon the activity or violate his or her religious beliefs); and (4) the detriment to the adherent from abandoning the activity creates substantial secular pressure on the adherent to violate his or her religious beliefs rather than abandon the activity.”
Id. at 943 (Kennard, J., dissenting). The law in Smith, then, constituted a “substantial burden” because the claimant was forced to choose between: (1) following her religious beliefs—and, consequently, either giving up her livelihood, in which she had invested time and effort, or facing stiff penalties including civil fines and imprisonment—or (2) obeying the law and violating her religious teachings. Id. at 946 (Kennard, J., dissenting). Judge Kennard also reasoned that the plurality’s concern with a religious exemption’s impact on third parties should not have been considered in the “substantial burden” analysis but rather at the “compelling interest” stage, where the impact on third parties can shed light on the government’s interest in the law. Id. at 947–48 (Kennard, J., dissenting). Finally, Judge Kennard wrote that, to meet the “compelling interest” and “least restrictive means” prongs, the government would have to prove that exempting landlords from housing laws would “so reduce the stock of housing available to unmarried heterosexual couples, or otherwise be so infeasible, as to defeat or even substantially impair its goal of providing equal housing opportunities to unmarried heterosexual couples.” Id. at 953 (Kennard, J., dissenting).

The Alaska Supreme Court also denied a landlord’s request for a religious exemption; however, in contrast with the California court, the Alaska court did reach the compelling interest question. 57 Swanner v. Anchorage Equal Rights Comm’n, 874 P.2d 274, 282 (Alaska 1994) (addressing “compelling interest” prong). Like the California court, the Alaska court held that forcing the landlord to choose between his religious beliefs and obeying the law in the context of voluntary commercial activity was not a “substantial burden.” 58 Id. at 283 (“[The landlord] has made no showing of a religious belief which requires that he engage in the property-rental business. Additionally, the economic burden . . . of which he complains, is caused by his choice to enter into a commercial activity that is regulated by anti-discrimination laws.”). On the compelling interest question, the court distinguished between a “derivative” interest in ensuring access to housing for everyone and a “transactional” interest in preventing individual acts of discrimination based on “irrelevant” characteristics like marital status. 59 Id. at 282 (distinguishing derivative interest, in which “the State does not object to the particular activity in which the individual would like to engage, but is concerned about some other variable that the activity will affect,” from transactional interest, in which “the State objects to the specific desired activity itself”); see also Eugene Volokh, RFRA Strict Scrutiny: The Interest in Protecting Newly Created Private Rights, Volokh Conspiracy (Dec. 6, 2013), http://volokh.com/2013/12/06/5c-rfra-strict-scrutiny-interest-protecting-newly-created-private-rights/ [http://perma.cc/G3XQ-5TY3] [hereinafter Volokh, RFRA Strict Scrutiny] (describing two ways of framing housing antidiscrimination interest—as interest in protecting private right not to be discriminated against or interest in making sure people have ability to find place to live—and suggesting framing is often outcome determinative on whether religious exemption should be granted). While the derivative interest may require an evidentiary showing that cohabitating couples had experienced hardship in finding available housing, the transactional interest did not. 60 Swanner, 874 P.2d at 286 (Moore, C.J., dissenting) (“Because the interest is ‘transactional,’ the majority concludes that no evidentiary basis is required . . . . However, before the court would enforce the state’s ‘derivative’ interest . . . the AERC apparently would have to make an evidentiary showing that cohabiting couples have experienced hardship in finding available housing.”). Even if the landlord could show that most couples that faced housing discrimination were eventually able to find housing, the state’s interest in ensuring these couples faced no discrimination was itself sufficiently compelling to override any burden on the landlord’s belief. 61 Id. at 284 (Moore, C.J., dissenting) (“Because Swanner’s religiously impelled actions trespass on the private right of unmarried couples to not be unfairly discriminated against in housing, he cannot be granted an exemption from the housing anti-discrimination laws.”).

Other courts found that religious landlords could obtain an exemption from antidiscrimination laws. A representative example is Donahue v. Fair Employment & Housing Corp., 62 2 Cal. Rptr. 2d 32 (Dist. Ct. App. 1992). a California appellate court case decided before Smith. In contrast with the other courts, the Donahue court held that forcing a religious landlord to rent to an unmarried couple was a substantial burden. First, the law contained “an affirmative obligation” to engage in conduct the landlord considered immoral “combined with sanctions,” which constituted a serious burden on free exercise even in the context of voluntary commercial activity; 63 Id. at 42–44 (“[P]eople do not lose their freedom of religion and ‘liberty of conscience’ when they engage in worldly activities. The burden imposed upon the freedom to act consistent with one’s religious beliefs, even in a commercial or societal context, can constitute a burden on the free exercise of religion which is far from incidental.” (citation omitted)). and second, by exercising their religious beliefs, the landlords were not imposing their religious beliefs on others but merely exercising their own right not to participate in the tenants’ “immoral” behavior. 64 Id. at 41 (“The Donahues’ asserted religious exemption does not involve their imposition of beliefs upon others, but rather their own attempt to personally refrain from, as they see it, a sinful facilitation of impermissible behavior.”).

Moving to the compelling interest question, the court noted that the state had sanctioned discrimination against cohabitating couples in numerous other contexts. 65 Id. at 44–45 (describing contexts in which California “sanctioned and judicially enforced discrimination against cohabiting couples,” including right to sue for loss of consortium, unemployment compensation, conjugal prison visits, marital communication privilege, right to bring wrongful death action for death of partner, and college and university housing). It then concluded that the state had no interest in “providing prospective tenants with one rental unit as opposed to any other unit when both units are, in the language used by the Legislature, decent and not unsanitary, unsafe, overcrowded or congested.” 66 Id. at 45. Since unmarried couples would likely be able to find comparable housing even if they were denied housing by one landlord, the state had no interest in forcing a particular landlord to abide by the antidiscrimination laws. 67 Id. In Swanner’s terms, the court focused on the “derivative” interest in ensuring couples have access to housing, ignoring a possible “transactional” interest in ensuring that couples do not face discrimination. See supra notes 59–60 and accompanying text (describing Swanner court’s articulation of derivative and transactional interests).

The Massachusetts Supreme Court also found for the religious landlord but took a slightly different perspective. Like the court in Donahue, the Massachusetts court found that a law that made the landlord’s exercise of religion more difficult, and forced the landlord to choose between abiding by the law or abiding by his religion and facing costly consequences, was a substantial burden. 68 Attorney Gen. v. Desilets, 636 N.E.2d 233, 237–38 (Mass. 1994) (“[T]he government has placed a burden on defendants that makes their exercise of religion more difficult and costly. The statute affirmatively obliges the defendants to enter into a contract contrary to their religious beliefs and provides significant sanctions for its violation.”). It also found that the commercial context did not reduce the seriousness of the burden. 69 Id. at 238 (“The fact that the defendants’ free exercise of religion claim arises in a commercial context, although relevant when engaging in a balancing of interests, does not mean that their constitutional rights are not substantially burdened.”). On the compelling interest question, rather than outright rejecting the notion of a compelling interest in preventing housing discrimination against unmarried couples, as the Donahue court did, the Massachusetts court held that the court would need empirical evidence on the prevalence of housing discrimination against unmarried couples and whether it would be possible to allow religious exemptions without impeding the ability of unmarried couples to find homes. 70 Id. at 240 (“We have no sense . . . of the numbers of rental units that might be withheld from such people [referring to cohabitating couples] because of the religious beliefs of the owners of rental housing.”).

Taken together, these cases help clarify the kinds of factors a court might consider in deciding a landlord’s claim for a religious exemption to an antidiscrimination law pre–Hobby Lobby. On the substantial burden prong of the analysis, a court would think about three main questions. First, given that the landlord can choose to avoid the law’s burden by not engaging in the commercial real estate market at all, the court would decide whether the landlord’s voluntary participation in the market precludes any claim of a substantial burden. 71 See Rebecca A. Wistner, Cohabitation, Fornication and the Free Exercise of Religion: Landlords Seeking Religious Exemption from Fair Housing Laws, 46 Case W. Res. L. Rev. 1071, 1108–10 (1996) (describing case law suggesting “government’s interest in regulating secular commercial activities adds to the magnitude of its compelling interests” and “fact that this religious objection takes place in commercial activity for profit makes any burden on the landlord less onerous”); David B. Cruz, Note, Piety and Prejudice: Free Exercise Exemption from Laws Prohibiting Sexual Orientation Discrimination, 69 N.Y.U. L. Rev. 1176, 1234–35 (1994)(arguing religious exemptions from laws regulating commercial activities are unlikely under RFRA because “governmental authority is at its zenith, and religious authority at its nadir, in the commercial zone”). Second, the court would think about whether simply placing an economic cost on the exercise of one’s religion, as opposed to explicitly banning a religious practice, could constitute a substantial burden. 72 See supra note 55 and accompanying text (describing traditional tenet that simply making belief more expensive is not “substantial burden”). Third, the court might think about the possible impact of a religious exemption on third parties, deciding whether the discriminatory impact on potential tenants’ rights precludes any claim of a substantial burden. 73 See O’Neil, supra note 31, at 801 (“[C]reating a religiously based exemption for the devout landlord might . . . have a deleterious effect on the interests of third parties—a factor implicit throughout the U.S. Supreme Court’s analysis.”); Julie M. Ruhlin, Note, Beyond “Compelling”: Tenants’ Rights in the Conflict Between Religious Freedom and Laws that Prohibit Discrimination Based on Sexual Orientation, 6 S. Cal. Rev. L. & Women’s Stud. 613, 635 (1997) (arguing religious exemptions in housing cases unduly burden tenants’ rights to sexual and decisional privacy and freedom of association). Turning to the compelling interest prong, the court would first consider whether the government has a compelling interest in protecting LGBT individuals from discrimination in the first place. If the court decided this question affirmatively, the remainder of its analysis would largely turn on the nature of the state’s interest. If the state’s interest was in ensuring LGBT individuals had access to housing, then an empirical analysis on the prevalence of LGBT housing discrimination and the impact of a religious exemption would be required. On the other hand, if the state’s interest was in protecting individuals from any form of discrimination, the court would likely find a compelling interest sufficient to outweigh any burden on the landlord’s religious practice. 74 See Christopher E. Anders & Rose A. Saxe, Effect of a Statutory Religious Freedom Strict Scrutiny Standard on the Enforcement of State and Local Civil Rights Laws, 21 Cardozo L. Rev. 663, 673 (1999) (contrasting Desilets and Swanner approaches to compelling interest); see also Volokh, RFRA Strict Scrutiny, supra note 59 (describing two ways courts have framed housing antidiscrimination interest).

D. The Burwell v. Hobby Lobby Decision

In Burwell v. Hobby Lobby Stores, Inc., 75 134 S. Ct. 2751 (2014). the Supreme Court faced petitions by several closely held corporations to gain an exemption from a portion of the Patient Protection and Affordable Care Act (ACA) that required employers to cover contraception in their employees’ insurance plans. The companies did not object to all forms of contraception, but only four specific types that prevented an already fertilized egg from implanting into the uterus, which, according to the owners’ religious beliefs, is equivalent to abortion. 76 See id. at 2766 (“Like the Hahns, the Greens believe that life begins at concep­tion and that it would violate their religion to facilitate access to contraceptive drugs or devices that operate after that point.”). In applying RFRA, the Court had to decide: (1) whether corporations had standing to assert claims under RFRA; (2) whether the ACA imposed a “substantial burden” on the plaintiffs’ religious beliefs; (3) whether this burden was justified by a compelling governmental interest; and (4) if so, whether the ACA was the least restrictive means of achieving this interest.

The Court began by holding that closely held corporations were “persons” in the context of RFRA and thus had standing to bring suit under the law. 77 Id. at 2768. Writing for the majority, Justice Alito noted that RFRA left the term “persons” undefined but that the Dictionary Act’s definition of “person” includes corporations and that RFRA’s intention of broad protection for religious liberty beyond that required by the First Amendment suggests that Congress intended to adopt this broad definition. Id. He also noted that the Court had frequently entertained RFRA and free exercise suits by nonprofit corporations and had even once entertained a suit by retail merchants who were not officially incorporated and concluded that it would make no sense to include these groups in the definition of “person” but not include at least closely held corporations. Id. at 2769–70. See Gallagher v. Crown Kosher Super Mkt of Mass., Inc., 366 U.S. 617, 631 (1961) (dismissing claims of Orthodox Jewish businessmen that Sunday closing laws burdened their free exercise of religion on merits, without deciding whether they had standing to assert free exercise claim). For a more detailed textual constitutional argument justifying the application of RFRA to for-profit corporations, see generally Jeremy M. Christiansen, Note, “The Word[] ‘Person’ . . . Includes Corporations”: Why the Religious Freedom Restoration Act Protects Both For- and Nonprofit Corporations, 2013 Utah L. Rev. 623. After making this threshold finding, the Court found that the ACA imposed a substantial burden on the plaintiffs’ religious beliefs. In sorting through this difficult question, 78 The “substantial burden” analysis poses particular difficulties for courts, in part because RFRA and free exercise case law are ambiguous as to what precisely the analysis should entail. See Lupu, The Failure of RFRA, supra note 46, at 591–92 (describing how courts often use “substantial burden” prong as way to limit religious exemptions because of its ambiguity); Jonathan Knapp, Note, Making Snow in the Desert: Defining a Substantial Burden Under RFRA, 36 Ecology L.Q. 259, 281–84 (2009) (arguing RFRA’s “substantial burden” test is ambiguous as to whether burden must be coercive or simply have substantial impact on religious exercise but asserting Congress likely intended to maintain both tests from pre-Smith case law). Justice Alito’s opinion relied on two factors: the plaintiffs’ subjective beliefs about the burden on their religion and objective facts about the magnitude of the burden. In response to the argument that requiring employers to pay for insurance that covers contraception is too “attenuated” a burden to be considered “substantial” (as opposed to, say, being forced to use contra­ception themselves or directly provide it to employees), Justice Alito asserted that courts should be deferential to religious claimants on the substantial burden question. Inquiring into the legitimacy of a plaintiff’s claim that a particular government action substantially burdens her religious practice would essentially involve an inquiry into the legitimacy of the plaintiff’s religious belief. 79 To illustrate Justice Alito’s point in the contraception context: Justice Ginsburg and the government argued that requiring employers to cover contraception for their employees was too “attenuated” to constitute a substantial burden on the plaintiffs’ anticontraception beliefs since it did not require them to use or support contraception but merely provide insurance plans that covered it for those employees that wished to use it. Hobby Lobby, 134 S. Ct. at 2799 (Ginsburg, J., dissenting). However, since the plaintiffs assert that their religious beliefs require them not only to avoid contraception themselves, but also to avoid facilitating others’ use of contraception, stating that the burden complained of is too “attenuated” is tantamount to saying that the plaintiffs’ religious beliefs are incorrect or unreasonable. Id. at 2778 (majority opinion) (stating “federal courts have no business addressing . . . whether the religious belief asserted in a RFRA case is reasonable”). Courts, particularly the Supreme Court, have historically been extremely reluctant to engage in inquiries into the sincerity or validity of a religious belief, as the inherent subjectivity of religious viewpoints makes such an inquiry difficult and carries risk of allowing religious biases or prejudice to creep into the Court’s judgment. 80 See, e.g., Emp’t Div., Dep’t of Human Res. v. Smith, 494 U.S. 872, 887 (1990) (“Repeatedly and in many different contexts, we have warned that courts must not presume to determine . . . the plausibility of a religious claim.”); Presbyterian Church v. Mary Elizabeth Blue Hull Mem’l Presbyterian Church, 393 U.S. 440, 450 (1969) (holding state courts cannot decide question relating to local church’s separation from its parent church because “First Amendment forbids civil courts” from “determin[ing] matters at the very core of a religion”); United States v. Ballard, 322 U.S. 78, 86 (1944) (holding First Amendment precludes any inquiry into “truth or verity of respondents’ religious doctrines or beliefs”). However, Justice Alito’s discussion seemed to take this deference, which had historically been accorded plaintiffs on the question of whether a religious belief is “sincerely held,” and apply it in the context of the “substantial burden” question as well. As Justice Alito stated:

[The government’s] argument dodges the question that RFRA presents (whether the [Health and Human Services] HHS mandate imposes a substantial burden on the ability of the objecting parties to conduct business in accordance with their religious beliefs) and instead addresses a very different question that the federal courts have no business addressing (whether the religious belief asserted in a RFRA case is reasonable). 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. This belief implicates a difficult and important question of religion and moral philosophy, namely, the circumstances under which it is wrong for a person to perform an act that is innocent in itself but that has the effect of enabling or facilitating the commission of an immoral act by another. Arrogating the authority to provide a binding national answer to this religious and philosophical question, HHS and the principal dissent in effect tell the plaintiffs that their beliefs are flawed. For good reason, we have repeatedly refused to take such a step. 81 Hobby Lobby, 134 S. Ct. at 2778.

However, despite this deferential rhetoric, Justice Alito’s opinion also emphasized the economic magnitude of the burden, writing that the companies’ high financial penalties for not following the law, or for refusing to offer insurance at all in protest, constituted a substantial burden. 82 Id. at 2776 (noting taking such actions would “entail substantial economic consequences,” including “penalties of $2,000 per employee each year”). The Court did not consider an argument raised in amicus briefs that these penalties were actually lower than the average cost of providing health insurance, meaning it would not be significantly burdensome for the companies to simply not provide health insurance and pay the penalties instead, because this argument was not raised in the lower courts. The Court did note, however, that the argument ignores the importance of providing health insurance to the companies’ morals, religious beliefs, and business models. Id. at 2276–77. This suggests that objective facts about the magnitude of the burden are still relevant in determining whether the burden is “substantial.” 83 Indeed, the Court in the past has examined objective factors in answering the “substantial burden” question, even while purporting to avoid inquiry into a religious claimant’s sincerity. See, e.g., Hernandez v. Comm’r, 490 U.S. 680, 699 (1989) (suggesting denying tax deduction to donations made to Church of Scientology’s auditing and training sessions may not constitute “substantial burden,” regardless of sincerity of beliefs, because Scientology does not forbid payment of taxes and thus “burden” simply means adherents have less money available to access such sessions).

Since Hobby Lobby, the majority of the courts of appeals have addressed the apparent tension in Justice Alito’s reasoning by holding that courts must defer to a religious objector’s view on what her religious exercise is, but must decide for themselves whether the challenged law imposes a “burden” and if so, whether that burden is substantial. 84 See E. Tex. Baptist Univ. v. Burwell, 793 F.3d 449, 456 (5th Cir. 2015) (holding courts “defer to a religious objector’s view” on own “religious exercise,” but courts must determine whether “challenged law pressure[s] [the objector] to modify that exercise” and whether pressure is “substantial”); Univ. of Notre Dame v. Burwell, 786 F.3d 606, 612 (7th Cir. 2015) (“Although Notre Dame is the final arbiter of its religious beliefs, it is for the courts to determine whether the law actually forces Notre Dame to act in a way that would violate those beliefs.”); Geneva Coll. v. Sec’y U.S. Dep’t of Health & Human Servs., 778 F.3d 422, 435 (3d Cir. 2015) (“Without testing the appellees’ religious beliefs, we must nonetheless objectively assess whether the appellees’ compliance with the self-certification procedure does, in fact, trigger, facilitate, or make them complicit in the provision of contraceptive coverage.”), mandate recalled and stayed sub nom. Zubik v. Burwell, 135 S. Ct. 1544 (2015) (mem.); 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 . . . .”); see also Mich. Catholic Conference & Catholic Family Servs. v. Burwell, 755 F.3d 372, 385 (6th Cir. 2014) (“[A]lthough we acknowledge that the appellants believe that the regulatory framework makes them complicit in the provision of contraception, we will independently determine what the regulatory provisions require and whether they impose a substantial burden on appellants’ exercise of religion.”), cert. granted and judgement vacated sub nom. Mich. Catholic Conference v. Burwell, 135 S. Ct. 1914 (2015). However, at least one circuit seems to have followed a more deferential interpretation of Hobby Lobby, holding that a plaintiff’s sincere determination that a required behavior is a “substantial burden” is sufficient to meet that prong, even if the court itself finds the burden to be objectively attenuated or low. 85 See Sharpe Holdings, Inc. v. U.S. Dep’t. of Health & Human Servs., 801 F.3d 927, 941 (8th Cir. 2015) (“As Hobby Lobby instructs, however, we must accept [plaintiffs’] assertion that self-certification under the accommodation process . . . would violate their sincerely held religious beliefs . . . . [I]f one sincerely believes that [completing  aform for a religious exemption from contraception law] will result in conscience-violating consequences, what some might consider an otherwise neutral act is a burden too heavy to bear.”). Therefore, it remains unclear to what extent courts should come to their own, independent judgment about whether there is a “substantial burden” using facts beyond the plaintiff’s own assertion. This circuit split will likely be resolved soon, as the Supreme Court has recently granted certiorari on cases touching on this very question. 86 Geneva Coll., 778 F.3d at 435, cert. granted sub nom. Zubik v. Burwell 136 S. Ct. 529 (mem.) (Nov. 17, 2015).
, 778 F.3d 422, 435 (3d Cir. 2015), cert. granted sub nom., Zubik v. Burwell, 136 S. Ct. 529 (mem.) (U.S. Nov. 17, 2015) (No. 14-1418).

However, even if the Court ultimately determines that the “substantial burden” question is still for the court and does not require total deference to the plaintiff, it does seem that courts applying Hobby Lobby should be much more deferential to plaintiffs’ claims of substantial burden than they have been in the past. 87 See supra note 46 and accompanying text (describing courts’ historical reluctance to find “substantial burden” when applying RFRA). In other words, before Hobby Lobby, a court may have been reluctant to find a “substantial burden” in a case like Hobby Lobby on a number of grounds, including that the action required of the believer by law was too “attenuated” from the behavior the believer objects to, that a religious believer’s participation in the business world negates her right to object to regulations her secular competitors must follow, or that merely making adherence to the believer’s faith “more expensive” through a fine cannot constitute a “substantial burden.” Under Hobby Lobby, even if the court retains the ability to define for itself whether there is a substantial burden, it would seem that it can no longer use these considerations to find that such a burden exists; to use them would be inconsistent with the Hobby Lobby Court’s own finding of a substantial burden. 88 See infra section II.A (discussing ways in which Hobby Lobby has changed “substantial burden” analysis in favor of religious objectors).

As for the “compelling interest” prong of the analysis, the Court assumed, without deciding, that providing access to contraception was a “compelling interest.” However, Justice Alito raised doubts based on the fact that the government had already exempted small businesses and businesses covered by grandfathered plans from this requirement. In Justice Alito’s view, this may have indicated that the interest was not particularly “compelling.” 89 Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751, 2779–80 (2014).

The Court, instead, moved directly to the “least restrictive means” test, in which it concluded that the contraception mandate could not be applied to the plaintiffs because the law was not the least restrictive means of furthering the government’s goal. 90 Id. at 2779–83. Justice Alito suggested that the government could easily adopt a less restrictive means by simply directly providing contraception for women whose employers had religious objections to such treatment. 91 Id. at 2781. Justice Alito pointed out that, while RFRA “cannot be used to require creation of entirely new programs,” it “surely allows” modification of an existing program and RFRA may “in some circumstances require the Government to expend additional funds to accommodate citizens’ religious beliefs.” 92 Id. The Court did acknowledge, however, that “cost may be an important factor in the least-restrictive-means analysis” and emphasized the fact that, in comparison with the ACA as a whole, the cost burden to the government of directly providing contraception would likely be fairly low when compared with the overall cost of the ACA (though, notably, the Court did not give any specific estimates or analysis of this cost burden). 93 Id. (“ACA’s insurance-coverage provisions will cost the Federal Government more than $1.3 trillion through the next decade. If [access to contraception is a compelling interest], it is hard to understand HHS’s argument that it cannot be required under RFRA to pay anything in order to achieve this important goal.” (footnote omitted) (citation omitted)). The Court went on to point out that such a measure was not necessary, however, because the government had already established a program by which religious nonprofits could apply for a religious exemption from the contraception provision and the burden would then shift to the insurance company itself. 94 Id. at 2782. Given this, there was no reason why this same “less restrictive means” could not be applied to for-profit corporations as well. 95 Id. Notably, the Court did not decide whether that less restrictive mechanism would pass the RFRA test. This will be a question before the Court in the immediate future. See Geneva Coll. v. Sec’y U.S. Dep’t of Health & Human Servs., 778 F.3d 422, 435 (3d Cir. 2015) (granting certiorari on question of whether filling out forms for religious exemptions from contraception requirement constitutes “substantial burden”), cert. granted sub nom. Zubik v. Burwell, 136 S. Ct. 529 (mem.) (Nov. 17, 2015).

E. The Problem of Anti-LGBT Housing Discrimination

As in many other areas, LGBT individuals face frequent discrimination in the housing market as a result of their actual or perceived sexual orientation or gender identity. LGBT individuals may face higher rents and application fees, denial of housing applications, homophobic comments or behavior by rental agents, and even sexual harassment. 96 See Fair Hous. Ctr. of Se. Mich., Sexual Orientation and Housing Discrimination in Michigan: A Report of Michigan’s Fair Housing Centers 9 (2007), http:// www.fhcmichigan.org/images/Arcus_web1.pdf [http://perma.cc/P4T9-VD2G] (finding significant discriminatory treatment of LGBT individuals in Michigan’s housing market); Jaime M. Grant et al., Nat’l Ctr. for Transgender Equality & Nat’l Gay & Lesbian Task Force, Injustice at Every Turn: A Report of the National Transgender Discrimination Survey 106 (2011), http://www.thetaskforce.org/static_html/downloads/reports/reports/ ntds_full.pdf [http://perma.cc/SH8A-SU7B] (finding nineteen percent of transgender national survey respondents reported being denied home or apartment and eleven percent being evicted because they were transgender or gender nonconforming). This discrimination sometimes forces LGBT individuals to resort to homeless shelters, where they may be denied access or face a hostile and dangerous environment because of their orientation or gender expression. 97 Grant, et al., supra note 96, at 106 (finding twenty-nine percent of transgender and gender-nonconforming people who tried to access homeless shelters were turned away, forty-two percent were forced to stay in facilities designated for the wrong gender, fifty-five percent reported being harassed, twenty-five percent were physically assaulted, and twenty-two percent were sexually assaulted).

There have been efforts to address this discrimination at both the state and federal levels. While the federal Fair Housing Act does not explicitly prohibit discrimination based on sexual orientation or gender identity, 98 Fair Housing Act LGBT Page, U.S. Dep’t of Hous. & Urb. Dev., http://portal. hud.gov/hudportal/HUD?src=/program_offices/fair_housing_equal_opp/LGBT_Housing_Discrimination [http://perma.cc/E32H-LS8S] [hereinafter Fair Housing Act LGBT Page] (last visited Jan. 28, 2016). the U.S. Department of Housing and Urban Development (HUD) has promulgated rules designed to promote equal access to housing for LGBT individuals in federally funded housing programs. 99 These rules prohibit the denial of housing based on sexual orientation or gender identity, forbid inquiries related to these characteristics by HUD housing administrators or mortgage lenders, and expand the definition of “family” and “household” to include LGBT individuals. 24 C.F.R. §§ 5.105(a)(2), 5.403, 200.300 (2012). In addition, twenty-two states ban housing discrimination based on sexual orientation, and seventeen of those states also ban discrimination based on gender identity. 100 Fair Housing Act LGBT Page, supra note 98. Over 200 cities and counties have also joined the effort against discrimination by passing ordinances forbidding housing discrimination against LGBT people. 101 Know Your Rights: Housing and Homeless Shelters, Nat’l Ctr. for Gender Equality, http://www.transequality.org/know-your-rights/housing-and-homeless-shelters [http:// perma.cc/UC5X-KFQF] (last visited Jan. 28, 2016).

The interaction of these antidiscrimination efforts with RFRA laws at both the state and federal level raises the very real possibility the landlords could use Hobby Lobby to try to seek an exemption from anti-discrimination laws. Currently, several states with antidiscrimination laws also have heightened religious freedom protection via a state RFRA or a RFRA-like interpretation of the state free exercise clause. 102 Compare Fair Housing Act LGBT Page, supra note 98 (showing states with laws banning housing discrimination on basis of sexual orientation or gender identity), with supra note 7 (providing sources showing states with heightened religious freedom protection). Addition­ally, in most states with RFRAs that do not have state-level anti-discrimination protections, at least some cities or other municipalities provide such protections on the local level; a landlord in such a munici­pality may use the state RFRA to challenge a local antidiscrimination ordinance. 103 Cf. Local Employment Non-Discrimination Ordinances, Movement Advancement Project, http://www.lgbtmap.org/equality-maps/non_discrimination_ordinances [http:// perma.cc/5Y3Q-JN9X] (last updated Jan. 27, 2016) (showing most states have municipali­ties with local antidiscrimination laws, even states that do not have them on state level). Furthermore, landlords operating federally funded housing subject to HUD’s antidiscrimination regulations under the Fair Housing Act might use the federal RFRA to seek exemptions from these regulations. Finally, provisions preventing sexual orientation and gender-identity discrimination are becoming more popular on both the state and local level as public support for LGBT equality increases. 104 See Justin McCarthy, Same-Sex Marriage Support Reaches New High at 55%, Gallup (May 21, 2014), http://www.gallup.com/poll/169640/sex-marriage-support-reaches-new-high.aspx (on file with the Columbia Law Review) (describing increasing public support for LGBT equality). At the same time, there has been a recent resurgence in state-level RFRAs and similar legislation as traditional religious believers seek to ensure their conscience rights are protected in the face of this new trend. 105 Mark A. Kellner, State Lawmakers Eye Expanding Religious Freedom Protections, Wash. Times (Jan. 13, 2015), http://www.washingtontimes.com/news/2015/ jan/13/state-lawmakers-eye-expanding-religious-freedom-pr/?page=all [http://perma.cc/ 6BB8-RVJS] (describing recent efforts to adopt RFRAs in states that did not previously have them). Therefore, questions about the validity of RFRA-based challenges to housing antidiscrimination laws are vital and will become increasingly important as the conflict between religious liberty and LGBT equality grows. 106 See Emma Green, Can States Protect LGBT Rights Without Compromising Religious Freedom?, Atlantic (Jan. 6, 2016), http://www.theatlantic.com/politics/archive/ 2016/01/lgbt-discrimination-protection-states-religion/422730/ [http://perma.cc/4NH7-ZE77] (discussing states where “battles are . . . looming” over proposed sexual-orientation antidiscrimination laws, religious exemptions from those laws, and state-level RFRAs).

II. “Substantial Burden” Analysis Under Burwell v. Hobby Lobby

The remainder of this Note addresses the realistic possibility that landlords might employ the logic of Hobby Lobby to seek religious exemptions from laws banning housing discrimination on the basis of sexual orientation. This Part addresses the first prong of the RFRA test: whether a landlord can realistically claim that an antidiscrimination law imposes a “substantial burden” on his or her religious beliefs. Section II.A describes the possible arguments a landlord could make, with a particular focus on how Hobby Lobby has likely changed the way a court would analyze this question in favor of the landlord. Section II.B discusses possible responses to these arguments.

A. Arguments in Favor of Finding a “Substantial Burden”

A religious landlord’s first task under Hobby Lobby would be to show that being forced to rent to LGBT individuals constitutes a “substantial burden” on the landlord’s sincerely held religious beliefs. 107 See Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751, 2775–79 (2014) (determining “whether the HHS contraceptive mandate ‘substantially burden[s]’ the exercise of religion”). Such an argument would run similarly to the arguments employed by landlords objecting to laws requiring them to rent to unmarried couples: By renting to a gay or lesbian couple, or an LGBT individual, the landlord would be giving them a place in which to engage in “sinful” relations with each other (in the case of a couple) or other individuals of the same sex (in the case of an individual), thereby facilitating the sinful behavior. Therefore, a law forcing landlords to do this imposes a substantial burden on their free exercise of religion. See supra notes 49–51 and accompanying text (describing similarity between unmarried-couples cases and LGBT discrimination cases). One argument a landlord could make is that the Hobby Lobby Court rejected the principle that a burden arising from one’s voluntary participation in the commercial real estate market is not “substantial” for RFRA purposes. Hobby Lobby involved closely held, for-profit corporations, and the Court made clear that the corporation owners’ voluntary choice to pursue a profit in the secular realm did not require them to sacrifice their religious convictions. 108 Hobby Lobby, 134 S. Ct. at 2767 (“HHS would put these merchants to a difficult choice: either give up the right to seek judicial protection of their religious liberty or forgo the benefits, available to their competitors, of operating as corporations.”). Of course, Hobby Lobby’s holding applies only to closely held corporations as opposed to large, publicly traded corporations. The case left open the question of whether such large corporations could bring RFRA claims. However, since the vast majority of landlords are individuals or closely held corporations as opposed to publicly traded corporations and it is unlikely that large publicly traded conglomerates would bring free exercise claims against antidiscrimination laws in the housing context, this potential limitation on Hobby Lobby’s holding is largely irrelevant to this Note’s analysis. See id. at 2774 (“These cases, however, do not involve publicly traded corporations, and it seems unlikely that the sort of corporate giants to which HHS refers will often assert RFRA claims.”). In light of this holding, landlords most likely cannot be compelled to choose between their religious convictions and obeying the law when the “burden” on their beliefs results from voluntary participation in the rental market, despite several state courts’ previous holdings that they could. 109 See supra note 53 and accompanying text (describing California Supreme Court’s holding that burden resulting from voluntary commercial activity cannot be “substantial”); supra note 58 and accompanying text (discussing Alaska Supreme Court’s similar holding); supra note 71 and accompanying text (presenting scholarly support for this proposition).

Second, a landlord could argue that the Hobby Lobby Court seemed to do away with the principle, repeated often in previous free exercise cases, that simply making the exercise of one’s beliefs more expensive does not constitute a substantial burden. 110 Hobby Lobby, 134 S. Ct. at 2770 (“Thus, a law that ‘operates so as to make the practice of . . . religious beliefs more expensive’ in the context of business activities imposes a burden on the exercise of religion.” (quoting Braunfeld v. Brown, 366 U.S. 599, 605)). In contrast to this statement, free exercise cases traditionally held that simply making a religious belief more expensive was not, itself, a substantial burden. See supra note 55 and accompanying text (discussing Braunfield v. Brown’s holding that making practice of religious beliefs more expensive is not substantial burden). Indeed, the potential cost the contraception mandate would impose on the plaintiffs for abiding by their religious convictions was a major motivating factor in the Court’s decision. 111 See Hobby Lobby, 134 S. Ct. at 2775 (basing determination that contraception mandate constituted “substantial burden” partly on fact that “if the [plaintiffs] do not yield . . . the economic consequences will be severe”). Since landlords who fail to abide by antidiscrimination laws will generally face some form of sanction, ranging from civil penalties to imprisonment, 112 See Smith v. Fair Emp’t & Hous. Comm’n, 913 P.2d 909, 946 (Cal. 1996) (Kennard, J., dissenting) (describing penalties faced by landlords who refuse to comply with antidiscrimination provisions, including “civil penalties, cease-and-desist order dictating her future conduct, and imprisonment”). a landlord would argue that such a law makes the exercise of the landlord’s religion significantly more expensive, and under the Court’s logic in Hobby Lobby, this should be enough to establish a substantial burden.

The third limiting principle found in state RFRA housing cases—that a substantial burden will not be found when a religious exemption would interfere with the rights of third parties—is similarly called into question by the Court’s decision. One of the major arguments in the government’s favor in the Hobby Lobby case was that granting a religious exemption would allow the plaintiff corporations to impermissibly burden their employees by denying them insurance coverage for certain contraceptives, coverage which is guaranteed to employees of nonreligious organizations by law. 113 See Hobby Lobby, 134 S. Ct. at 2790 (Ginsburg, J., dissenting) (“The exemption sought by Hobby Lobby and Conestoga would override significant interests of the corporations’ employees and covered dependents. It would deny legions of women who do not hold their employers’ beliefs access to contraceptive coverage that the ACA would otherwise secure.”); id. at 2801 (Ginsburg, J., dissenting) (“No tradition, and no prior decision under RFRA, allows a religion-based exemption when the accommodation would be harmful to others—here, the very persons the contraceptive coverage requirement was designed to protect.”). Landlords asserting religious objections may argue that the Court’s rejection of this “third-party-harm” argument in Hobby Lobby signals that possible harm to tenants from a landlord’s religious exemption, such as denial of access to housing and injuries to the tenant’s dignity or well-being, 114 See infra notes 128–135 and accompanying text (discussing various harms to tenants resulting from landlord’s religious exemption). should not automatically prevent the landlord from claiming a substantial burden on their beliefs. 115 Cf. Eugene Volokh, Intermediate Questions of Religious Exemptions—A Research Agenda with Test Suites, 21 Cardozo L. Rev. 595, 608–09 (1999) (arguing burden imposed on third parties should not limit religious exemptions, as “laws are generally allowed to impose all sorts of burdens on people” and thus a burden resulting from someone’s free exercise should not be given same weight as a burden on someone’s free exercise).

Finally, Hobby Lobby’s deferential approach to the “substantial burden” prong means that it may be fairly easy for landlords to meet this portion of the test. In his majority opinion, Justice Alito implied that a plaintiff’s sincere belief that his or her religious exercise is substantially burdened will be given deference, regardless of how attenuated the conduct mandated by the law is from the practice the plaintiffs believe is immoral or how reasonable the plaintiff’s belief appears to the ordinary person. 116 Compare Hobby Lobby, 134 S. Ct. at 2778 (majority opinion) (refusing to chal­lenge reasonableness or sincerity of plaintiffs’ belief that paying for employees’ contraception was “substantial burden,” regardless of how attenuated law’s requirements may be from plaintiffs’ stated religious beliefs), with id. at 2798 (Ginsburg, J., dissenting) (criticizing majority for failing to distinguish between “‘factual allegations that plaintiffs’ beliefs are sincere and of a religious nature,’ which a court must accept as true, and the ‘legal conclusion . . . that [plaintiffs’] religious exercise is substantially burdened,’ an inquiry the court must undertake” (quoting Kaemmerling v. Lappin, 553 F.3d 669, 679 (D.C. Cir. 2008))). Therefore, while there may have once been a potential argument that simply renting to an individual who identifies as LGBT cannot reasonably be seen as condoning their behavior, 117 Cf. Smith, 913 P.2d at 923 (suggesting, in some contexts, testimony by other members of plaintiff’s church establishing church did not believe renting to unmarried couples was facilitating immoral behavior could be helpful in evaluating plaintiff’s sincerity). a landlord can assert that such an argument should not prevail in a post–Hobby Lobby world. It is true that Justice Alito also mentioned the magnitude of the potential fines the plaintiff faced in justifying the Court’s finding of “substantial burden.” This might imply that a court may still use other facts beyond the plaintiff’s own assessment to make an independent judgment as to whether a substantial burden exists. However, even read this way, it is likely that Hobby Lobby supports a finding of a substantial burden in a case like this, since violating antidiscrimination laws will generally carry some penalty ranging from fines to imprisonment, meaning these “other facts” will almost always be present. 118 See, e.g., N.Y. Exec. Law § 299 (McKinney 2013) (providing penalties of up to one year of prison time or up to $500 in fines for violation of New York’s antidiscrimination law).

B. An LGBT Individual’s Response to a “Substantial Burden” Claim

Despite Hobby Lobby’s deferential substantial burden standard, an LGBT tenant or the government might argue that it is unclear, and doubtful, that its holding was meant to imply that virtually any claim of a substantial burden will pass muster under the RFRA test. First, as discussed above, Justice Alito’s discussion of the economic cost to the Hobby Lobby plaintiffs for refusal to comply with the ACA may imply that courts can still consider external facts about the burden on plaintiffs’ beliefs, so long as the analysis does not turn into a constitutionally dubious examination of the sincerity or legitimacy of an individual’s claim of religious burden. 119 See supra notes 78–87 and accompanying text (describing subjective and objective components of Justice Alito’s “substantial burden” analysis). As discussed above, at present, it is unclear exactly how much deference a court must give the plaintiff’s own assessment that there is a “substantial burden” on his or her beliefs. Perhaps in a case where the court determines that the burden is fairly minor based on facts beyond the plaintiff’s own assessment of the situation, a court may have room to conclude for itself that no substantial burden exists despite the plaintiff’s protests. Of course, this doctrinal point in itself is unlikely to help LGBT advocates much in the housing-discrimination context, as any serious antidiscrimination law will include penalties and sanctions for violations. However, it does give reason to consider possible limits a court might place on a plaintiff’s ability to claim a “substantial burden.”

Second, LGBT individuals might argue that the courts should draw a distinction between a religious individual being forced to provide something he or she finds morally objectionable per se, such as abortifa­cients, and being forced to provide something morally neutral, such as housing, that is tangentially related to the morally objectionable behavior at issue (here, same-sex sexual activity). Arguably, Hobby Lobby’s holding only applies to the former (albeit even “attenuated” actions involving the former), while it does not reach the latter. After all, it seems almost ridiculous to assert that providing someone with housing indicates complicity in everything the tenant does in that housing, while it is much less strange to say that paying for someone’s abortifacients indicates complicity in their use of said abortifacients. Such an argument, however, is essentially no different than saying that providing housing in which couples will engage in morally objectionable sexual relations is “too attenuated” to that behavior to be considered a burden—an argument that seems to have been soundly rejected in Hobby Lobby. Therefore, it is unlikely that this particular argument will be salient in a post–Hobby Lobby world.

More compellingly, while Hobby Lobby has clearly rejected the primary ways lower courts have found “substantial burdens” in the past, 120 See supra section II.A (discussing impact of Hobby Lobby holding on validity of various “substantial burden” arguments in housing discrimination context). one key argument employed by lower courts in the housing context remains viable: Burdens on third parties should lessen a landlord’s ability to claim a “substantial burden.” 121 See supra note 71 and accompanying text (describing impact of third parties as factor in “substantial burden” analysis); see also Smith, 913 P.2d at 925 (“[T]he landlord’s request for an accommodation in the case before us has a serious impact on the rights and interests of third parties.”). In his opinion, Justice Alito made clear that Hobby Lobby’s holding rested on the fact that the effect of a religious exemption on the plaintiffs’ female employees would be “precisely zero.” 122 Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751, 2760 (2014). In the Court’s view, granting a religious exemption to the plaintiffs would not deny their female employees the right to access and use contraception because the government would see to it that these women would have access to contraception without cost-sharing, either by providing it directly or requiring insurance companies to do so. 123 Id. at 2759–60. Since, in the Court’s determination, there is no extra cost to women for receiving birth control from the government or an insurance company rather than an employer, the Court held that there were no third-party harms to granting a religious exemption that might offset the employer’s legitimate substantial-burden claim. 124 Id. at 2780–82 (describing alternative ways for government to achieve goal of providing access to contraception without cost-sharing).

Regardless of what one thinks of the Court’s analysis in the birth-control context, it is clear that, in the housing discrimination context, allowing a religious exemption imposes real, tangible costs on third parties. First, even if we assume that a potential tenant denied housing based on sexual orientation or gender identity could find decent housing elsewhere, the act of discrimination prevented that tenant from obtain­ing their housing of choice. The housing they ultimately find may be in a worse location, of poorer quality, or less desirable to the tenant in some other way. 125 See, e.g., Donahue v. Fair Emp’t & Hous. Comm’n, 2 Cal. Rptr. 2d 32, 34 (Ct. App. 1991) (relating how, as result of discrimination by preferred apartment’s landlord, unmarried couple had to settle for apartment with higher rent and of poorer quality). Thus, while such discrimination may not deny the tenants access to housing per se, it does prevent them from having access to the full range of options that heterosexual or cisgender tenants would have, imposing a special cost that is not imposed on every tenant. 126 A landlord might argue that the government can address this “special cost” by granting subsidies to discriminated-against individuals to eliminate the added rent cost of more expensive alternative housing, or by setting minimum-quality requirements to ensure that discriminated-against individuals have access to at least suitable housing. These possibilities, however, do not solve the fundamental issue: The LGBT individual is denied his or her idiosyncratic choice of housing, which he or she would otherwise qualify for, due to the irrelevant characteristic of sexual orientation or gender identity. Nothing the government can do, short of forcing landlords to accept otherwise qualified LGBT tenants, can correct this fundamental inequality between LGBT individuals and others who do not face this same problem (i.e. make the cost on tenants “precisely zero”). For further explanation of why government-funded alternatives cannot provide a less restrictive means to an antidiscrimination law, see infra section III.B.3 (arguing government-subsidized housing is not appropriate “less restrictive means” to fulfill antidiscrimination interest).

Second, denying potential tenants housing due to their sexual orientation or gender identity imposes an emotional cost on the tenants by forcing them to suffer the indignity of being discriminated against for a core aspect of their identity. 127 See Kara Loewentheil, When Free Exercise Is a Burden: Protecting “Third Parties” in Religious Accommodation Law, 62 Drake L. Rev. 433, 485–86 (2014) (describing “dignitary and expressive harm” of housing discrimination as equal or more important than practical harms such as denial of housing). While a woman is arguably unlikely to suffer a major indignity or emotional cost by obtaining her birth control from a source other than her employer, 128 There is a reasonable argument, however, that a woman’s dignity, mental health, or emotional well-being may suffer from her employer’s refusal to cover contraception, as she is forced to suffer the indignity of working for an employer who looks down on her for her reproductive choices. Neither the Hobby Lobby majority nor the dissent considered this argument. Thus, regardless of whether this argument has merit, the Court’s failure to either affirm or reject it leaves room to utilize a similar argument in the discrimination context, where the harm to an individual’s emotional and mental health from discrimination is well established. See infra note 129 and accompanying text (discussing these harms from discrimination). subjecting a potential tenant to overt discrimination, even when it does not ultimately affect the tenant’s ability to find housing, singles out and marginalizes the tenant as an “other” unworthy of the landlord’s business, which can be harmful to the individual’s well-being and mental health. 129 See, e.g., Vickie M. Mays & Susan D. Cochran, Mental Health Correlates of Perceived Discrimination Among Lesbian, Gay, and Bisexual Adults in the United States, 91 Am. J. Pub. Health 1869, 1869 (2001) (discussing research findings suggesting perceived sexual-orientation discrimination is “positively associated with both harmful effects on quality of life and indicators of psychiatric morbidity”).

Third, allowing landlords to refuse housing to LGBT individuals on religious grounds by exempting them from civil rights laws may burden the LGBT individuals’ own constitutionally protected rights to privacy and sexual autonomy. 130 See Lawrence v. Texas, 539 U.S. 558, 564–65, 578–79 (2003) (declaring statute criminalizing homosexual acts in defendant’s own home violates defendant’s constitutional “right to privacy”); Roe v. Wade, 410 U.S. 113, 154 (1973) (affirming right to privacy extends to abortion rights); Griswold v. Connecticut, 381 U.S. 479, 485–86 (1965) (striking down state contraception ban as violation of individuals’ right to privacy). In the Hobby Lobby context, granting a religious exemption would not burden a woman’s constitutional right to use birth control because (in the Court’s view) there were clearly other ways by which the government could ensure this right, meaning an infringement on the employer’s religious rights was unnecessary. 131 Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751, 2760 (2014) (holding religious exemption had “precisely zero” effect on plaintiff’s female employees because of ways in which government can ensure women’s access to contraception). By contrast, allowing a landlord to exercise his religious rights to discriminate against LGBT individuals would essentially allow the landlord to tell a tenant that in order to lease the landlord’s apartment, she must refrain from constitutionally protected sexual acts 132 See Lawrence, 539 U.S. at 578–79 (declaring statute prohibiting sexual acts between consenting adults unconstitutional); supra note 127 and accompanying text (noting potential burden of religious exemptions on LGBT individuals’ constitutional rights). while in the privacy of her living space. Again, while the LGBT individual may be able to find other housing in which she can exercise these rights, the landlord’s actions are limiting the LGBT person’s ability to exercise privacy and sexual autonomy rights to only those living spaces run by landlords who are willing to accept them. Therefore, unlike in the birth-control context, allowing a religious exemption for a landlord in this context can be viewed as an impermissible imposition by the government on tenants’ constitutionally protected rights. 133 Cf. Wistner, supra note 71, at 1104–06 (describing antidiscrimination housing laws as means of protecting individuals’ privacy and intimate-association interests); Ruhlin, supra note 73, at 635 (arguing religious exemptions in housing cases unduly burden tenants’ rights to sexual and decisional privacy and freedom of association).

III.  “Compelling Interest” and “Least Restrictive Means” Analysis After Hobby Lobby

Assuming a landlord asserting a RFRA-type claim or defense can demonstrate that antidiscrimination laws create a substantial burden, she would next have to show that the government does not have a compelling interest behind the law, or, alternatively, that there is a less restrictive means for achieving this interest. 134 See supra note 40 and accompanying text (noting requirements for RFRA claim). This Part addresses this aspect of the RFRA test. Section III.A discusses arguments a landlord might make. Section III.B provides responses to these arguments and concludes that protecting LGBT individuals from housing discrimination is a compelling interest and an antidiscrimination law is the least restric­tive means of furthering that interest.

A. Arguments in Favor of a Landlord

On the compelling interest question, one argument a landlord could make is that, like unmarried couples, LGBT people are not a protected class; therefore, the government does not have a compelling interest in protecting them from discrimination. 135 Cf. Stephanie Hammond Knutson, Note, The Religious Landlord and the Conflict Between Free Exercise Rights and Housing Discrimination Laws—Which Interest Prevails?, 47 Hastings L.J. 1669, 1719–24 (1996) (discussing “hierarchy of discrimination categories” argument used by state courts to determine marital-status discrimination is not as compelling as racial or gender discrimination). Under the Supreme Court’s Equal Protection jurisprudence, laws or other state actions that discriminate based on certain “suspect classifications,” such as race or gender, are subject to heightened scrutiny. 136 See Korematsu v. United States, 323 U.S. 214, 216 (1944) (“[A]ll legal restrictions which curtail the civil rights of a single racial group are immediately suspect . . . . [C]ourts must subject them to the most rigid scrutiny.”). For a law to survive such heightened scrutiny, the government must show that it has a compelling interest justifying the discriminatory law, as opposed to merely showing that the law has a “rational basis.” See Johnson v. California, 543 U.S. 499, 505 (2005) (“Under strict scrutiny, the government has the burden of proving that racial classifications ‘are narrowly tailored measures that further compelling governmental interests.’” (quoting Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 227 (1995))). For RFRA purposes, if a party is seeking an exemption to discriminate against a group the Court considers a protected class, this exemption will be less likely to be granted, as the government’s interest in protecting this class will be viewed as very compelling. 137 See Knutson, supra note 135, at 1719–24 (arguing laws designed to prevent discrimination against certain protected classes like race and gender will almost certainly serve as sufficient “compelling interest” to override burden on landlord’s beliefs, while those designed to prevent discrimination against nonprotected classes like unmarried couples may or may not). Despite the major advances in LGBT rights over the past several decades, including the Court’s most recent decision holding state laws forbidding same-sex marriage are unconstitutional as a violation of the fundamental right to marry, 138 See Obergefell v. Hodges, 135 S. Ct. 2584, 2607–08 (2015) (declaring constitu­tional right to same-sex marriage). the Court has been hesitant to declare sexual orientation or gender identity protected classes that would merit this heightened scrutiny. 139 See id. (holding anti-same-sex-marriage laws unconstitutional without addressing protected class issue); United States v. Windsor, 133 S. Ct. 2675, 2706 (2013) (Scalia, J., dissenting) (commenting majority opinion striking down Defense of Marriage Act “does not resolve and indeed does not even mention what had been the central question in this litigation: whether, under the Equal Protection Clause, laws restricting marriage to a man and a woman are reviewed for more than mere rationality”); Romer v. Evans, 517 U.S. 620, 631 (1996) (applying rational basis review because amendment targeting LGBT individuals did not “target[] a suspect class” but nevertheless striking law down). It is true that “sex,” defined in the binary sense of being male or female, is a protected class subject to a form of heightened scrutiny. See, e.g., Craig v. Boren, 429 U.S. 190, 197 (1976) (“[C]lassifications by gender must serve important governmental objectives and must be substantially related to achievement of those objectives.”); Frontiero v. Richardson, 411 U.S. 677, 688 (1973) (“[C]lassifications based upon sex, like classifications based upon race, alienage, or national origin, are inherently suspect, and must therefore be subjected to strict judicial scrutiny.”). However, despite several lower court opinions suggesting that discriminating against someone for their expressed gender identity is a form of sex-discrimination, the Supreme Court has yet to hold that gender nonconforming individuals are either themselves a protected class or protected by the Court’s sex discrimination jurisprudence. See, e.g., Morales v. ATP Health & Beauty Care, Inc., No. 3:06-CV-01430, 2008 WL 3845294, at *8 (D. Conn. Aug. 18, 2008) (holding transgender plaintiff properly alleged she was member of protected class under Title VII and Connecticut nondiscrim­ination law because of her perceived failure to conform to gender stereotypes); Dep’t of Fair Emp’t & Hous. v. Marion’s Place, FEHC Dec. No. 06-01, 2006 WL 1130912, at *13 (Cal. Fair Emp’t & Hous. Comm’n Feb. 1, 2006) (holding nightclub violated state’s law prohibiting sex discrimination in public accommodations by excluding transgender patrons). For example, in cases involving anti-same-sex-marriage laws, the Court has avoided stating that these laws constitute discrimination against a protected class, instead striking them down as deprivations of the “fundamental right” to marry belonging to all citizens. 140 See Obergefell, 135 S. Ct. at 2608 (declaring state same-sex marriage bans uncon­stitutional as violation of fundamental right to marry). A landlord might make the argument that, since the Court has refused to find that LGBT individuals are deserving of heightened protection despite having ample opportunity to do so, preventing discrimination against them is not an interest sufficiently compelling to justify burdening a religious individual’s deeply held, constitutionally protected beliefs. 141 Cf., e.g., Swanner v. Anchorage Equal Rights Comm’n, 874 P.2d 274, 287–89 (Alaska 1994) (Moore, J., dissenting) (stating “not . . . every form of discrimination is equally invidious [nor does the] state’s interest in preventing it necessarily outweigh[] fundamental constitutional rights” and holding marital status discrimination does not fall into most invidious category because unmarried couples do not constitute protected class); Attorney General v. Desilets, 636 N.E.2d 233, 247 (Mass. 1994) (O’Connor, J., dissenting) (arguing state’s interest in protecting free exercise of religion is greater than preventing discrimination based on marital status because “[m]arital status discrimination is not as intense a State concern as is discrimination based on certain other classifications”).

Assuming the suspect class argument fails, a landlord’s second argument might be that the government interest at issue is in ensuring that all individuals have access to decent and affordable housing, which is not served by refusing to grant religious exemptions to particular landlords with sincere religious objections. 142 Indeed, all but one of the courts that have examined the compelling interest question in this context have framed the interest as one in ensuring that all members of the discriminated-against group have access to decent and affordable housing. See, e.g., Donahue v. Fair Emp’t and Hous. Comm’n, 2 Cal. Rptr. 2d 32, 45 (Ct. App. 1993) (“There is no state housing interest in providing prospective tenants with one rental unit as opposed to any other unit when both units are . . . decent and not unsanitary, unsafe, overcrowded, or congested.”); Desilets, 636 N.E.2d at 239–41 (examining interest in terms of ensuring all individuals have access to housing rather than simply preventing invidious discrimination). But see Swanner, 874 P.2d at 282–83 (asserting state has transactional interest in protecting unmarried couples from discrimination). A landlord would assert that, to justify burdening the landlord’s belief, the government has to demonstrate evidence of substantial housing discrimination against LGBT individuals in the particular state (or city, or town) that has adopted the antidiscrimination law and show that this discrimination is so widespread that allowing even a select few landlords to discriminate would severely reduce the supply of housing available to LGBT people. 143 See Volokh, RFRA Strict Scrutiny, supra note 59 (“[U]nder the particularized analysis that RFRA calls for . . . this would only justify denying the exemption in those areas, rather than throughout the whole state. So in deciding on the exemptions, courts would have to hear factual evidence about . . . housing discrimination based on marital status in the particular community.”). Consistent with this analysis, state courts that have examined the issue have held or implied that, to show that this interest justifies a burden on a landlord’s religious belief, the government must bring forth evidence showing that housing discrimination against the discriminated-against group is such a pervasive problem that allowing an individual landlord to obtain a religious exemption would severely hinder this interest. See, e.g., Desilets, 636 N.E. 2d at 239–41 (requiring specific factual findings showing discrimination against unmarried couples in state is so prevalent as to justify burdening landlord’s belief). This may be a difficult task, as even if the government can show prevalent, above-average discrimination in a particular area, it may be unlikely that an LGBT individual denied housing will not be able to find some other suitable housing. 144 See Wistner, supra note 71, at 1107–08 (“[A] few exemptions probably will not result in much harm to the individual tenants. It may slightly inconvenience them, but if they find alternative housing, the governmental interest is still served. Moreover, for economic reasons, it is unlikely that droves of landlords will seek religious exemption from antidiscrimination laws.”); Kelly Catherine Chapman, Note, Gay Rights, the Bible, and Public Accommodations: An Empirical Approach to Religious Exemptions for Holdout States, 100 Geo. L.J. 1783, 1820–21 (2012) (positing market forces generally discourage discrimination because discrimination based on characteristics like sexual orientation is not generally profit-maximizing for landlords or employers); Volokh, RFRA Strict Scrutiny, supra note 59 (positing “in most communities, unmarried couples will find many places to live, even if a few landlords refuse to rent to them” because few landlords believe renting to such couples is sinful and to claim exemption goes against landlord’s economic interest). Even in extreme cases, such as extremely small and homogeneous communities where housing discrimination is truly so widespread that an LGBT person really cannot find nondiscriminatory housing, a landlord would claim that this at most justifies a compelling interest in preventing discrimination in that particular community, but not on the state or even city level. See id. (“[U]nder the particularized analysis that RFRA calls for . . . [heavy discrimination in particular areas] would only justify denying the exemption in those areas, rather than throughout the whole state.”).

If the above arguments fail, a landlord might argue that government provision of housing for discriminated-against people can constitute a “less restrictive means” under Hobby Lobby, even if this increases costs for the government. 145 See Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751, 2781 (2014) (“[B]oth RFRA and its sister statute, RLUIPA, may in some circumstances require the Government to expend additional funds to accommodate citizens’ religious beliefs.”); see also id. at 2787 (Ginsburg, J., dissenting) (arguing “commercial enterprises . . . can opt out of any law . . . they judge incompatible with their sincerely held religious beliefs” so long as there is a less restrictive alternative, which “there always will be whenever . . . the government, i.e., the general public, can pick up the tab”). States and the federal government already provide government-subsidized housing for low-income tenants who cannot afford to rent on the private market. In some cases, nondiscrimination on the basis of sexual orientation is a condition of these subsidies; for example, federal HUD regulations prohibit discrimination based on sexual orientation for properties receiving HUD subsidies. 146 See supra note 98 and accompanying text (describing HUD regulations). Therefore, such a housing program that explicitly bars discrimination may serve as a “less restrictive alternative” to burdening the religious belief of a private landlord. 147 Of course, this argument would only be possible for a private landlord; a landlord already participating in a subsidized housing program would obviously not have such an argument available to them. Of course, in some places, this may require an expansion of public housing or the addition of a provision that bars discrimination in housing covered by a subsidized housing program. However, expanding publicly subsidized housing to shelter victims of discrimination who are unable to find housing on the private market would not require the creation of a new government program but merely the expansion of an existing one. The Hobby Lobby Court held that such an expansion can be forced upon the government as a “less restrictive means” if the alternative is burdening someone’s religious beliefs. 148 Hobby Lobby, 134 S. Ct. at 2781 (majority opinion) (holding RFRA “surely allows” government to be required to modify existing programs to avoid burdening someone’s religious beliefs). Further, such a public housing expansion would not need to have the capacity to house every single LGBT individual in a state but need only be large enough to serve as a last resort to house those individuals who could not find housing on the private market due to discrimination. Therefore, a landlord might argue that the alternative of public housing would fit nicely with Hobby Lobby’s articulation of a less restrictive means. 149 Cf. Loewentheil, supra note 127, at 485 (“Theoretically, the state could establish a national system of subsidized public housing that ensures that, practically speaking, no one would be deprived of access to housing on the basis of race or sex.”).

B. Antidiscrimination Laws Are the “Least Restrictive Means” of Ensuring LGBT Individuals Have Equal Access to Housing

This section argues that antidiscrimination laws are the least restrictive means of furthering the compelling interest of protecting LGBT individuals from housing discrimination, even under the logic of Hobby Lobby. Section III.B.1 argues that protecting LGBT people from discrimination is a compelling interest because, as Supreme Court jurispru­dence has established, protecting these individuals from discrimina­tion is a goal worthy of heightened concern, whether or not they are technically members of a “suspect class.” Section III.B.2 argues that, whether this interest is framed as “ensuring access to affordable housing” or “protecting individuals from discrimination,” an antidiscrimination law is the least restrictive means of fulfilling it. Section III.B.3 argues that expanding public housing to include victims of discrimina­tion, a possible alternative to an antidiscrimination law, is not a realistic or appropriate means of fulfilling the government’s interest.

1. The Government Has a Compelling Interest in Protecting LGBT Individuals from Discrimination. — On the “compelling interest” question, the first argument the government must respond to is that sexual orientation and gender identity are not “suspect classifications,” and thus preventing discrimination based on those characteristics cannot be seen as a “compelling interest” on the level of racial or gender discrimina­tion. 150 See supra notes 135–138 and accompanying text (outlining argument that LGBT people are not “protected class” and thus protecting them from discrimination is not “compelling interest”). First, the government might argue that, despite courts’ avoidance of the question in previous cases, LGBT individuals are properly viewed as members of a suspect class worthy of special protection. Suspect classification applies when those comprising the class have suffered a history of discrimination; exhibit obvious, immutable, or distinguishing characteristics that define them as members of a discrete group; and are politically powerless or a minority. 151 Cf. Bowen v. Gilliard, 483 U.S. 587, 602–03 (1987) (finding group not suspect or quasi-suspect class where they have experienced none of these factors). LGBT individuals likely meet these criteria, as they are a minority group that has historically suffered dis­crimination as a result of their expressed sexual or gender identities. 152 See, e.g., Gary W. Harper & Margaret Schneider, Oppression and Discrimination Among Lesbian, Gay, Bisexual, and Transgendered People and Communities: A Challenge for Community Psychology, 31 Am. J. Community Psychol. 243, 243–44 (2003) (describing historical oppression and discrimination against LGBT people).

One argument against suspect classification is that LGBT individuals, while certainly a minority group, can no longer be considered “politically powerless” given their recent success at political organization and activism. 153 See Romer v. Evans, 517 U.S. 620, 636 (1996) (Scalia, J., dissenting) (referring to LGBT people as “politically powerful minority”). However, by this logic, women and racial minorities also should no longer be considered politically powerless, since both of these groups have made great strides in gaining greater societal power and acceptance. Furthermore, the continued lack of protections for LGBT individuals in many places on both the state and federal level, as well as the fact that much of the recent progress in LGBT rights has come through court decisions rather than legislative enact­ments, undermines the notion that LGBT individuals are so powerful that they need no protection in the courts.

Another argument is that housing discrimination, and anti-LGBT discrimination more generally, is not based on individuals’ LGBT identity per se, but on their behavior, i.e. having sexual relations with people of the same sex or expressing a gender different from one’s biological sex. Therefore, LGBT individuals are not facing discrimination due to an “immutable characteristic,” 154 See Frontiero v. Richardson, 411 U.S. 677, 686 (1973) (establishing heightened scrutiny should apply to classifications based on “immutable characteristic[s] determined solely by the accident of birth,” since discriminating on basis of such characteristics would “violate ‘the basic concept of our system that legal burdens should bear some relationship to individual responsibility’” (quoting Weber v. Aetna Cas. & Sur. Co., 406 U.S. 164, 175 (1972))). but due to actions that others find morally repulsive. However, such an argument ignores the centrality of sexual behaviors and gender expression to an LGBT individual’s identity. There is a growing consensus in the scientific community that sexual orientation and gender identity are not products of personal choice but inherent components of one’s identity, much like race or biological sex, and that the freedom to behave in ways congruous with these characteristics is a fundamental component of these identities. 155 Am. Psychology Ass’n, Answers to Your Questions: For a Better Understanding of Sexual Orientation and Homosexuality 1–2 (2008), http://www.apa.org/topics/lgbt/ orientation.pdf [http://perma.cc/B3N3-UWTM] (stating scientific consensus that “most people experience little or no sense of choice about their sexual orientation” and sexual orientation “refers to a person’s sense of identity based on those attractions, related behaviors, and membership in a community of others who share those attractions”). Therefore, sexual orientation and gender identity can reasonably be seen as akin to gender or racial discrimination. 156 See Robert Wintemute, Sexual Orientation and Gender Identity, in Human Rights in the Community: Rights as Agents for Change 175, 185 (Colin Harvey ed., 2005) (describing analogy between sexual-orientation discrimination and other forms of discrimination).

If LGBT individuals are not members of a suspect class, however, this does not automatically mean that their protection from discrimination is not a compelling interest. Even while avoiding the “suspect class” question, the Court has implicitly (if not explicitly) applied heightened scrutiny, rather than mere “rationality review,” to questions involving LGBT discrimination. 157 See, e.g., Jerald W. Rogers, Note, Romer v. Evans: Heightened Scrutiny Has Found a Rational Basis—Is the Court Tacitly Recognizing Quasi-Suspect Status for Gays, Lesbians and Bisexuals?, 45 U. Kan. L. Rev. 953, 969 (1996) (arguing Romer implicitly made sexual orientation quasi-suspect class); Jeremy B. Smith, Note, The Flaws of Rational Basis with Bite: Why the Supreme Court Should Acknowledge Its Application of Heightened Scrutiny to Classifications Based on Sexual Orientation, 73 Fordham L. Rev. 2769, 2781–83 (2005) (describing language in Romer indicating application of heightened scrutiny beyond rational basis to sexual orientation classifications). Ordinary rationality review is extremely deferential; a law will withstand scrutiny so long as the court can find “any reasonably conceivable state of facts that could provide a rational basis,” even if this state of facts did not actually motivate the legislature or is based only on “rational speculation unsupported by evidence or data.” 158 FCC v. Beach Commc’ns Inc., 508 U.S. 307, 313–15 (1993). Despite this, “[w]hen a law exhibits such a desire to harm a politically unpopular group, [the Court has] applied a more searching form of rational basis review.” 159 Lawrence v. Texas, 539 U.S. 558, 580 (2003) (O’Connor, J., concurring). Therefore, despite purporting to apply rational basis review to laws seemingly motivated by animosity or dis­crimina­tory attitudes toward LGBT people, the Court has consistently struck these laws down, even in the presence of justifications that would likely pass ordinary rationality review. 160 See, e.g., Obergefell v. Hodges, 135 S. Ct. 2584, 2608 (2015) (declaring constitutional right to same-sex marriage based on “fundamental right” to marriage that cannot be denied to LGBT people); United States v. Windsor, 133 S. Ct. 2675, 2693 (2013) (striking down federal same-sex marriage ban as motivated by “bare congressional desire to harm a politically unpopular group” despite various stated justifications based on desire for uniformity and stability in federal government’s definition of marriage (internal quotations omitted)); Romer v. Evans, 517 U.S. 620, 632–35 (1995) (striking down state constitutional amendment forbidding discrimination protections for LGBT individuals on grounds that it was “born of animosity toward” them, despite stated rationales such as respect for other citizens’ freedom of association and state need to conserve antidiscrimination resources). This special attention to antidiscrimination interests indicates that, even if LGBT people do not meet the criteria for full protected class status, the Court considers discrimination against LGBT people a sufficiently important interest to warrant heightened scrutiny even under apparent “rationality” review. In the context of a housing-related RFRA claim, this jurisprudence implies that preventing discrimination toward even a non-protected class can be justified as a compelling interest worthy of the state’s intervention.

2. No Matter How the “Compelling Interest” Is Defined, an Antidiscrimination Law Is the Least Restrictive Means of Meeting It. — Once it is determined that the government has a compelling interest in protecting LGBT individuals from discrimination, an antidiscrimination law is the least restrictive means of furthering that interest, no matter whether that interest is defined as an interest in preventing discrimina­tion per se or simply in ensuring everyone has access to housing. The government should first argue that the state has a “transactional interest” in protecting individuals from discrimination based on irrelevant characteristics such as sexual orientation or gender identity. 161 See Swanner v. Anchorage Equal Rights Comm’n, 874 P.2d 274, 282–83 (Alaska 1994) (describing state’s transactional interest in preventing discrimination per se as opposed to merely practical harms resulting from discrimination). Under this view, it does not matter whether the discrimination prevents an LGBT individual or couple from accessing housing. The simple act of discrimination alone damages individuals’ well-being, causes isolation and marginalization from society, and can harm an individuals’ mental health, and therefore is itself an evil the state legitimately seeks to eradicate. 162 See id. (identifying “independent social evil” of discrimination as affront to “human dignity”); see also Loewentheil, supra note 127, at 485–86 (“[T]here are symbolic and expressive elements of discrimination that rise above its practical effects, and even the most outlandish solutions would not eradicate these elements.”); Mays & Cochran, supra note 129, at 1869–72 (describing effects of perceived sexual orientation discrimination on LGBT individuals’ quality of life and mental health). Since there is no less restrictive means of preventing acts of discrimination than passing an antidiscrimination law, the antidis-crimination law would survive the compelling interest test.

However, if the court decides that the compelling interest is not preventing discrimination per se but ensuring all individuals have access to decent and affordable housing, an antidiscrimination law is still the least restrictive means of furthering this interest. Again, the key to the Hobby Lobby Court’s finding of a less restrictive means was that there were alternative ways in which the government could accomplish the exact same goal of providing access to birth control without burdening an employer’s religious exercise. 163 Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751, 2760 (2014) (justifying holding based on fact that effect on women’s access to contraceptives—the compelling interest in question—is “precisely zero”). By contrast, if certain landlords are able to obtain religious license to discriminate against LGBT tenants, these tenants may find other suitable housing, but it may be in a worse location, have higher rent, be of poorer quality, or otherwise have less desirable specifications than the place they would have chosen but for the discrimination. 164 See, e.g., Donahue v. Fair Emp’t & Hous. Comm’n, 2 Cal. Rptr. 2d 32, 34 (Ct. App. 1991) (describing how unmarried couple was forced to settle for higher-cost, lower-quality housing as result of discrimination). Therefore, such tenants are still denied the range and choice of housing options that non-LGBT individuals have access to. Given this reality, an interest in ensuring access to housing is not truly served in the presence of religious exemptions, since religious exemp­tions limit the housing options available to LGBT individuals and place these tenants’ housing access at the mercy of landlords’ good (or ill) will. The fact that the government cannot achieve the exact same result if religious exemptions are allowed takes this situation outside the scope of Hobby Lobby’s holding. 165 See supra note 126 (explaining why other government alternatives to correct this problem, such as rent subsidies or quality requirements, do not achieve same result as anti-discrimination law).

3. Public Housing Is Not a Realistic or Acceptable Alternative to an Antidiscrimination Law. — There are also several arguments the government and LGBT activists can use to demonstrate that expanding public housing, or making already existing housing subsidies conditional on agreeing not to discriminate, cannot be a real “less restrictive alternative” to fulfill the interest of ensuring LGBT individuals have access to housing. First, public housing is not analogous to the feasible and (in the Court’s view) inexpensive “less restrictive means” available in Hobby Lobby. Even while holding that RFRA may “in some circumstances require the Government to expend additional funds to accommodate citizens’ religious beliefs,” Justice Alito’s opinion made clear that “cost may be an important factor in the least-restrictive-means analysis.” 166 Hobby Lobby, 134 S. Ct. at 2781. It also was careful to emphasize that the cost to the government of directly providing contraception would likely be “minor when compared with the overall cost of [the Affordable Care Act].” 167 Id. By contrast, an expansion of public housing beyond its original intention—providing housing for those individuals who cannot afford housing on the private market—to also shelter victims of discrimination would be unprecedented and carries an unknown cost and feasibility. Such an expansion would not only involve increased spending on maintenance and public housing projects 168 How much increased spending this would involve is unclear. For example, it is possible that, if landlords knew they did not have to comply with an antidiscrimination law because public housing was available, many of them would seek religious exemptions, leading to a major increase in individuals who need public housing and consequently a major increase in costs to the government. On the other hand, it is possible that only a few truly devout landlords would seek religious exemptions, meaning that only a small expansion in public housing would be needed as a last resort for the relatively few individuals who could not find housing on the private market due to discrimination. See Wistner, supra note 71, at 1107–08 (arguing most landlords will not seek religious exemptions to discriminate simply because it is not financially desirable to discriminate against tenants who are willing to pay rent). but also additional administrative costs resulting from the need for additional employees to process, analyze, and approve claims of discrimination. 169 Cf. N.Y.C. Hous. Auth., Guide to Applying for Public Housing, NYC.gov (Mar. 2014), http://www.nyc.gov/html/nycha/downloads/pdf/070008.pdf [http://perma.cc/ 3UQS-TZNZ] (describing various eligibility criteria, necessary documentation, and bureaucratic processes for determining eligibility for public housing).

These costs would not simply be financial but would include efficiency costs from additional bureaucracy; faced with constrained resources, government agents would have to decide which of those individuals petitioning for public housing claiming discrimination truly needed the housing and which were likely able to find housing on the private market. Done poorly, the system could lead to many meritorious claims being turned away and many discriminated-against individuals left without any recourse. Cf. Tonia Bui, Kathryn Kliff (12) Protects Homeless Families Rights to Shelters, Equal Justice Works (July 11, 2014,12:58 pm), http://www.equaljusticeworks.org/news/blog/Kathryn-Kliff [http://perma.cc/KC3J-8C97] (relating anecdotes of families denied emergency housing under New York City’s Prevention Assistance and Temporary Housing program, despite being eligible). These costs could constrain resources and lead to the rejection of many meritorious claims of discrimination. In short, such a plan is far more impractical than the feasible and relatively inexpensive plan approved in Hobby Lobby.

Even ignoring the practical difficulties of such a system, public housing is not a feasible “less restrictive means” because it does not allow the government to achieve the same result as an antidiscrimination law. 170 See supra note 163 and accompanying text (explaining Hobby Lobby’s emphasis that “less restrictive means” approved by Court would allow government to achieve exact same goal as ACA contraception requirement). An LGBT individual who has the capacity to afford her preferred housing on the private market but has to settle for public housing (which is often of lower quality) due to discrimination in the private market, is not afforded the same range of options that his or her heterosexual counterparts have. 171 See supra notes 162–163 and accompanying text (describing how limiting range of housing options LGBT individuals have by allowing religious exemptions for landlords fails RFRA’s “compelling interest” test). It also imposes harm to the potential tenant’s dignity and emotional well-being, as the tenant is denied his or her choice of housing and must settle for whatever housing the government sees fit to provide, purely as a result of the person’s sexual orientation or gender identity. 172 See Loewentheil, supra note 127, at 485–86 (describing problem with public housing as alternative to antidiscrimination laws as not “only—or even primarily—the expense and logistical difficulty (or impossibility) of doing so, but also the dignitary and expressive harm of such a ‘separate but equal’ regime”). Therefore, the availability of public housing does not make the effect of discrimination on this individual “precisely zero” 173 Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751, 2760 (2014). but in fact still imposes a great cost on LGBT tenants.

These reasons perhaps explain why public housing has never been taken seriously as an alternative to antidiscrimination laws in the racial and gender context. 174 See Loewentheil, supra note 127, at 485–86 (describing idea of providing public housing as alternative to antidiscrimination laws and concluding “even to state th[is] hypothetical[] is to dismiss [it] as absurd”). A final argument against public housing as a less restrictive means might be that allowing it in the LGBT context would also raise the possibility of using it in the racial- and gender-discrimination contexts as well. 175 After all, if public housing is an acceptable alternative to an antidiscrimination law in the LGBT context, there is no clear reason why it would not be acceptable in the racial and gender contexts as well, assuming a landlord could assert a sincerely held belief that renting to individuals of different races or genders violates his or her religious principles. This could open the door to religious exemptions from housing antidiscrimination laws beyond the LGBT context, threatening to make the cost of such a regime even higher and possibly leading to greater racial and gender segregation.

Conclusion

RFRA began as a bipartisan effort to protect religious exercise from oppression and discrimination in a legal environment that was widely seen as subordinating individual religious freedom to secular goals. 176 See supra section I.B (describing motivation behind and enactment of RFRA). In the wake of Hobby Lobby, many legitimately fear that it has become a tool for individuals to use religion as an excuse to thwart even vitally important social goals, such as access to health care or discrimination prevention. 177 See supra notes 5–8 and accompanying text (describing concerns that Hobby Lobby will lead to socially harmful religious exemptions in wide range of contexts). Such concerns have even led some to call for RFRA’s repeal and have made such provisions at the state level extremely controversial. 178 See, e.g., Katha Pollitt, Why It’s Time to Repeal the Religious Freedom Restoration Act, Nation (June 30, 2014), http://www.thenation.com/article/180832/why-its-time-repeal-religious-freedom-restoration-act# [http://perma.cc/59XJ-UP87] (arguing RFRA grants too much freedom from generally applicable laws and should be repealed); Action Alert: Ask Congress to Counter Supreme Court’s Hobby Lobby Ruling, Freedom from Religion Found. (June 30, 2014), http://ffrf.org/news/action/item/20865-ask-congress-to-counter-supreme-court-s-hobby-lobby-ruling [http://perma.cc/2CQS-VG65] (calling for members of public to call congressional representatives and demand RFRA’s repeal); Emma Margolin, “Religious Freedom” Measure Moves Forward in Michigan, MSNBC (Dec. 6, 2014 1:47 pm,), http://www.msnbc.com/msnbc/religious-freedom-measure-moves-forward-michigan [http://perma.cc/7W2L-RSW8] (describing controversy over state RFRAs in Michigan and other states in wake of Hobby Lobby). By using the example of housing discrimination, this Note has aimed to assuage these concerns by showing that the holding of Hobby Lobby truly is limited by its facts, and its logic should not be successfully used to allow religious exemptions in one key area of concern: housing discrimination against LGBT individuals. Of course, housing discrimination is only one of many areas in which RFRA might be applicable; one can envision a wide variety of challenges to generally applicable laws by both conservative and liberal 179 While the most prominent RFRA cases come from politically conservative religious practitioners, it should be noted that progressive and liberal religious individuals may also attempt to use RFRA to avoid or challenge laws they find burdensome. See, e.g., Nicola Menzie, Church Leaders: Alabama Anti-Immigration Law ‘Merciless,’ Christian Post (Aug. 2, 2011), http://www.christianpost.com/news/church-leaders-alabama-immigration-law-tramples-religious-freedom-53216/ [http://perma.cc/Q6KL-PUYZ] (describ­ing chal­lenge to state law making it crime to transport, conceal, harbor, or shield illegal immigrants on grounds that it would inhibit Christians’ ability to follow religious mandate to “welcome and care for all people”); Eyder Peralta, United Church of Christ Challenges North Carolina Ban on Gay Marriage, Nat’l Pub. Radio (Apr. 28, 2014, 7:13 pm), http://www.npr.org/blogs/thetwo-way/2014/04/28/307793118/united-church-of-christ-challenges-north-carolina-ban-on-gay-marriage [http://perma.cc/KJ8F-K384] (describing challenge to state same-sex marriage ban by LGBT-friendly church asserting ban violates church’s religious practice of marrying same-sex couples). religious objectors, and these will have to be evaluated on a case-by-case basis. However, many of the arguments in this Note related to Hobby Lobby’s limitations can be applied in other contexts as well. With principled and thoughtful applications of the principles laid out in Hobby Lobby and prior case law, RFRA can, and should, remain an important tool for protecting religious exercise while continuing to uphold important social goals.