Grace Schools v. Sylvia Mathews Burwell

• Court: Court of Appeals for the Seventh Circuit
• Decided: September 4, 2015
• Docket: 14-1430
• Status: Published

Opinion

In the

United States Court of Appeals For the Seventh Circuit Nos. 14-1430 & 14-1431

GRACE SCHOOLS, et al., AND DIOCESE OF FORT WAYNE-SOUTH BEND, INC., et al., Plaintiffs-Appellees,

v.

SYLVIA MATHEWS BURWELL, et al., Defendants-Appellants.

Appeals from the United States District Court for the Northern District of Indiana. Nos. 3:12-cv-00459-JD-CAN and 1:12-cv-00159-JD-RBC — Jon E. DeGuilio, Judge.

ARGUED DECEMBER 3, 2014 — DECIDED SEPTEMBER 4, 2015

Before MANION, ROVNER, and HAMILTON, Circuit Judges.

ROVNER, Circuit Judge. The district court entered a prelimi- nary injunction in favor of the plaintiffs, a number of religious, not-for-profit organizations, preventing the defendants from applying or enforcing the so-called “contraceptive mandate” of the Patient Protection and Affordable Care Act of 2010 (“ACA”) to the plaintiffs. See 42 U.S.C. § 300gg-13(a)(4); Pub. 2 Nos. 14-1430 & 14-1431

L. No. 111-148, 124 Stat. 119 (2010). The plaintiffs contend that the ACA’s accommodations for religious organizations impose a substantial burden on their free exercise of religion, and that the ACA and accompanying regulations are not the least restrictive means of furthering a compelling government interest, in violation of the plaintiffs’ rights under the Religious Freedom Restoration Act of 1993 (“RFRA”). See 42 U.S.C. § 2000bb et seq. The defendants, several agencies of the United States government, appeal. We conclude that ACA does not impose a substantial burden on the plaintiffs’ free exercise rights and so we reverse and remand. However, we will maintain the injunction for a period of sixty days in order to allow the district court adequate time to address additional arguments made by the parties but not addressed prior to this appeal. I. The ACA requires group health plans and third-party administrators of self-insured plans to cover preventive care for women under guidelines supported by the Health Re- sources and Services Administration (“HRSA”), a component of the Department of Health and Human Services (“HHS”). 42 U.S.C. § 300gg-13(a)(4); 45 C.F.R. § 147.130(a)(1)(iv); University of Notre Dame v. Burwell, 786 F.3d 606, 607 (7th Cir. 2015) (hereafter “Notre Dame II”); University of Notre Dame v. Sebelius, 743 F.3d 547, 548 (7th Cir. 2014), vacated by 135 S. Ct. 1528 (2015) (hereafter “Notre Dame I”). The relevant guidelines include “all Food and Drug Administration approved contra- ceptive methods, sterilization procedures, and patient educa- tion and counseling for all women with reproductive capac- ity.” 77 Fed. Reg. 8725-26. The regulations adopted by the three Nos. 14-1430 & 14-1431 3

Departments implementing this part of the ACA require coverage of, among other things, all of the contraceptive methods described in the guidelines. See 45 C.F.R. § 147.130(a)(1)(iv) (HHS); 29 C.F.R. § 2590.715-2713(a)(1)(iv) (Labor); 26 C.F.R. § 54.9815-2713(a)(1)(iv) (Treasury).1 In anticipation of objections from religious organizations to these requirements, the Departments provided an exemption from the contraception coverage provision for religious employers. 45 C.F.R. § 147.131(a). A religious employer is defined as “an organization that is organized and operates as a nonprofit entity and is referred to in section 6033(a)(3)(A)(i) or (iii) of the Internal Revenue Code of 1986, as amended.” 45 C.F.R. § 147.131(a); 26 U.S.C. § 6033(a)(3)(A). That provision of the Internal Revenue Code, in turn, refers to “churches, their integrated auxiliaries, and conventions or associations of churches,” and “the exclusively religious activities of any religious order.” 26 U.S.C. § 6033(a)(3)(A)(i) and (iii). But the exemption did not cover religiously-affiliated non-profit corporations such as schools and hospitals that did not meet the IRS guidelines for religious employers. The Departments therefore adopted additional regulations providing accommo- dations for group health plans provided by these non-profit

1 All three of these regulations have been amended since this suit was filed. The most recent amendments, which are scheduled to take effect Sept. 14, 2015, address accommodations for closely-held for-profit corporations whose owners have religious objections to some or all of the contraceptive coverage requirements of the ACA. See Burwell v. Hobby Lobby, 134 S. Ct. 2751 (2014). Because these most recent amendments are not relevant to the issues raised here, we will be referring to the version of the regulations in effect at the time this suit was filed, unless we state otherwise. 4 Nos. 14-1430 & 14-1431

religious corporations, called “eligible organizations” in the regulations: (b) Eligible organizations. An eligible organization is an organization that satisfies all of the following requirements: (1) The organization opposes providing coverage for some or all of any contraceptive services required to be covered under § 147.130(a)(1)(iv) on account of religious objections. (2) The organization is organized and operates as a nonprofit entity. (3) The organization holds itself out as a religious organization. (4) The organization self-certifies, in a form and manner specified by the Secretary, that it satisfies the criteria in paragraphs (b)(1) through (3) of this section, and makes such self-certification available for examination upon request by the first day of the first plan year to which the accommodation in paragraph (c) of this section applies. The self-certification must be executed by a person authorized to make the certification on behalf of the organization, and must be maintained in a manner consistent with the record retention requirements under section 107 of the Employee Retirement Income Security Act of 1974. Nos. 14-1430 & 14-1431 5

45 C.F.R. § 147.131(b).2 See also 78 Fed. Reg. 39,874-75. Eligible organizations are not required “to contract, arrange, pay, or refer for contraceptive coverage” to which they have religious objections. 78 Fed. Reg. 39,874. The government developed a two-page form for eligible organiza- tions to use to comply with this accommodation, the “EBSA Form 700 – Certification.”3 The short form requires the eligible organization to supply its name, the name and title of the individual authorized to make the certification on behalf of the organization, and a mailing address and telephone number for that individual. The form also requires a signature verifying the statement, “I certify the organization is an eligible organi- zation (as described in 26 CFR 54.9815-2713A(a), 29 CFR 2590.715-2713A(a); 45 CFR 147.131(b)) that has a religious objection to providing coverage for some or all of any contra- ceptive services that would otherwise be required to be covered.” The organization must then provide a copy of the certification to the organization’s health insurance issuer or, for self-insured plans, to its third-party administrator. The insurer or administrator receiving the certification is obligated to provide (or arrange for the provision of) contraception cover- age for the health plan’s participants without cost sharing through alternate mechanisms established by the regulations.

2 This regulation will also be updated as of Sept. 14, 2015. Again, we cite to the earlier version.

3 The form can be found at http://www.dol.gov/ebsa/pdf/preventive serviceseligibleorganizationcertificationform.pdf, last visited September 3, 2015. 6 Nos. 14-1430 & 14-1431

45 C.F.R. § 147.131(c). The insurer4 may not impose a charge of any variety, either directly or indirectly, on the eligible organization for the provision of contraception services.5 The insurer must also inform plan participants that the eligible organization will not provide or fund any contraception coverage. 45 C.F.R. § 147.131(d). As we will discuss below, since the filing of this suit, these regulations have been amended to allow a second method of objecting to contracep- tive coverage, by notifying HHS directly of any religiously- based objection. The plaintiffs are various religiously-based non-profit organizations including the Diocese of Fort Wayne-South Bend, Inc. (“Diocese”); Catholic Charities of the Diocese of Fort Wayne-South Bend, Inc. (“Catholic Charities”); Saint Anne Home & Retirement Community of the Diocese of Fort Wayne-

4 From this point forward, when we use the term “insurer,” we mean to include third-party administrators in those instances where the plan is self- insured unless we state otherwise.

5 Insurers are expected to recoup the costs of contraceptive coverage from savings on pregnancy medical care as well as from other regulatory offsets. See Notre Dame II, 786 F.3d at 609–10; 78 Fed. Reg. 38977-78 (“Issuers are prohibited from charging any premium, fee, or other charge to eligible organizations or their plans, or to plan participants or beneficiaries, for making payments for contraceptive services, and must segregate the premium revenue collected from eligible organizations from the monies they use to make such payments. In making such payments, the issuer must ensure that it does not use any premiums collected from eligible organiza- tions.”). Third-party administrators may seek reimbursement of up to 110% of their costs from the government. Notre Dame II, 786 F.3d at 609; 45 C.F.R. § 156.50(d)(3). Nos. 14-1430 & 14-1431 7

South Bend, Inc. (“St. Anne Home”); Franciscan Alliance, Inc.; Specialty Physicians of Illinois LLC (“Specialty Physicians”); University of Saint Francis (“St. Francis”); Our Sunday Visitor, Inc. (“Sunday Visitor”); Biola University, Inc. (“Biola”) and Grace Schools. The plaintiffs objected below to the regulatory scheme, which they characterize as a “contraceptive services mandate,” on numerous grounds. Primarily, they asserted that the regulations force them to participate in a system that contravenes their religious beliefs in violation of the RFRA. 42 U.S.C. § 2000bb et seq.6 In particular, they are forced to contract with insurers or third-party administrators that will provide their employees (and, in some cases, their students) with coverage for contraceptives, sterilization, and abortion- inducing products, all in violation of their deeply held reli- gious beliefs. The accommodation provides them no relief, they contended below, because it causes them to trigger and facilitate the same objectionable services for their employees and students. A non-complying employer7 who does not meet

6 The plaintiffs also allege that the challenged statute and regulations violate their rights under the First Amendment and under the Administra- tive Procedures Act, 5 U.S.C. § 500 et seq. Because the district court issued the injunction after considering only the RFRA, and because neither side has briefed the other issues, we will confine our discussion to the RFRA. On remand, the plaintiffs are free to pursue their other theories for relief and, in fact, we will leave the injunction in place for a limited time in order to allow the court to consider those additional claims.

7 The disputed regulations apply equally to employers providing insurance to employees and to institutions of higher education providing student health insurance. See 45 C.F.R. § 147.131(f). Some of the plaintiffs provide (continued...) 8 Nos. 14-1430 & 14-1431

an exemption faces fines of $2000 per year per full time employee8 for not providing insurance that meets coverage requirements, 26 U.S.C. § 4980H(c), or $100 per day per employee for providing insurance that excludes the required contraceptive coverage, 26 U.S.C. § 4980D, and will face the risk of other enforcement actions. The Diocese itself is exempted from challenged require- ments under the religious employer exemption,9 and the remaining plaintiffs are subject to the accommodation for non- profit, religiously-affiliated employers. The government does not contest the sincerity of the plaintiffs’ religious objections to the required contraceptive coverage. Moreover, all of the plaintiffs consider the provision of health insurance for their employees and students to be part of their religious mission.

7 (...continued) both employee and student health coverage.

8 When calculating the number of employees for the purpose of assessing this penalty, the statute directs that thirty employees be subtracted from the total number of employees, essentially reducing the penalty by $60,000 per year for affected employers. 26 U.S.C. § 4980H(c)(2)(D)(i).

9 Although the Diocese is itself exempt, the Diocesan Health Plan insures employees of the non-exempt Catholic Charities. In order to protect Catholic Charities from having to comply with either the contraceptive mandate or the accommodation, the Diocese has forgone almost $200,000 annually in increased premiums in order to maintain its grandfathered status under the ACA. See 42 U.S.C. § 18011. Grandfathered plans are those health plans that need not comply with the coverage requirements of the ACA because they were in existence when the ACA was adopted and have not made certain changes to the terms of their plans. Nos. 14-1430 & 14-1431 9

Although the plaintiffs concede that they are not required to pay for the objectionable services, they contended in the district court that being forced to contract with insurers or third-party administrators who must then provide those services makes them a facilitator of objectionable conduct, complicit in activity that violates their core religious beliefs. The plaintiffs also asserted below that the government’s interest in providing contraceptive services is not compelling and that the means the government employed are not the least restrictive available to achieve the government’s goals. On those bases, the plaintiffs sought and received a preliminary injunction in the district court. The district court noted that the RFRA prohibits the federal government from placing substantial burdens on a person’s exercise of religion unless it can demonstrate that applying the burden is “in furtherance of a compelling governmental interest,” and is the “least restrictive means of furthering that compelling governmental interest.” 42 U.S.C. § 2000bb-1(a) and (b). The court first considered whether the contraception regulations create a substantial burden on eligible employers in light of the accommodation provided by the regulations. Citing our opinion in Korte v. Sebelius, 735 F.3d 654 (7th Cir. 2013), cert. denied, 134 S. Ct. 2903 (2014), the court noted that “the pertinent inquiry for the substantial burden test under RFRA is whether the claimant has an honest conviction that what the government is requiring or pressuring him to do conflicts with his religious beliefs and whether the governmen- tal pressure exerts a sufficiently coercive influence on the plaintiffs’ religious practice.” Grace Schools v. Sebelius, 988 F. Supp. 2d 935, 950 (N.D. Ind. 2013); Diocese of Fort Wayne- 10 Nos. 14-1430 & 14-1431

South Bend, Inc. v. Sebelius, 988 F. Supp. 2d 958, 972 (N.D. Ind. 2013). The court found that the plaintiffs sincerely believe that the accommodation compels them to facilitate and serve as a conduit for objectionable contraceptive services for their employees and students. If the plaintiffs want to provide health insurance for their students and employees as part of their religious mission (and in order to avoid the fines imposed by the ACA on employers who fail to meet coverage require- ments), the court reasoned, then they must either provide the objectionable coverage themselves or comply with the accom- modation. And the plaintiffs sincerely believe that invoking the accommodation facilitates and enables the provision of contraceptive services to their employees and students; the accommodation, in short, makes them complicit in the provi- sion of services to which they possess a religious objection. That they need not pay for the services provides no relief from their religious dilemma, the district court reasoned, because they must violate their religious beliefs by either forgoing providing health insurance to their employees and students, or they must take critical steps (i.e. comply with the accommoda- tion) to facilitate a third party’s provision of the objectionable coverage. Because failure to take either of these equally objectionable routes would result in the imposition of large financial penalties, the district court found that the plaintiffs demonstrated that the ACA imposes a substantial burden on their free exercise rights in contravention of the RFRA. The court then assumed that the government possessed a compel- ling interest in providing seamless contraceptive services to women in group health plans, but found that the accommoda- Nos. 14-1430 & 14-1431 11

tion was not the least restrictive means of accomplishing that goal. The court therefore enjoined the defendants from enforcing against the plaintiffs the requirements “to provide, pay for, or otherwise facilitate access to coverage for FDA approved contraceptive methods, abortion-inducing drugs, sterilization procedures, and related patient education and counseling.” Grace Schools, 988 F. Supp. 2d at 958; Diocese of Fort-Wayne-South Bend, 988 F. Supp. 2d at 980. The government appeals. II. Several months after the district court entered the injunc- tions for the plaintiffs here, we issued our opinion in Notre Dame I, where we affirmed the denial of a motion for a prelimi- nary injunction under strikingly similar circumstances to those presented by these appeals. The government asserts that our decision in Notre Dame I controls the result here and requires that we reverse the preliminary injunctions granted by the district court. The plaintiffs argue that Notre Dame I is distin- guishable and that application of the substantial burden test from Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751 (2014), and Korte requires that we affirm the preliminary injunctions here. After this appeal was fully briefed and argued, the Supreme Court vacated and remanded our opinion in Notre Dame I “for further consideration in light of Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751 (2014).” University of Notre Dame v. Burwell, 135 S. Ct. 1528 (2015). We recently issued a new opinion addressing the effect of Hobby Lobby on Notre Dame’s appeal. See Notre Dame II, 786 F.3d at 615–19. We will begin our analysis with our original Notre Dame I opinion, which continues to apply to some of the questions raised here, 12 Nos. 14-1430 & 14-1431

before we turn to Notre Dame II. “We review the district court's findings of fact for clear error, its balancing of the factors for a preliminary injunction under the abuse of discretion standard, and its legal conclusions de novo.” United Air Lines, Inc. v. Air Line Pilots Ass’n, Int’l, 563 F.3d 257, 269 (7th Cir. 2009). To obtain a preliminary injunction, a party must establish that it is likely to succeed on the merits, that it is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in its favor, and that issuing an injunc- tion is in the public interest. Smith v. Executive Dir. of Ind. War Mem’ls Comm’n, 742 F.3d 282, 286 (7th Cir. 2014). A. In Notre Dame I, a non-profit Catholic university moved to enjoin the enforcement of the ACA’s contraception provisions against it. 743 F.3d at 551. Notre Dame provides health benefits to its employees and students. The university self-insures the employees and utilizes a third-party administrator to manage the plan. It contracts directly with an insurance provider for the student health plan. 743 F.3d at 549. The ACA requires the university, as an eligible organization, either to provide contraceptive coverage for its employees or to comply with the accommodation by opting out through the use of the EBSA Form 700 certification (“Form 700"), which we described above. 743 F.3d at 550. The relevant regulations required Notre Dame to provide the completed Form 700 to its third-party adminis- trator and to the insurer of the student plan. Notre Dame filed suit shortly before the deadline for complying with the accommodation and moved for a preliminary injunction. The district court denied the motion and Notre Dame appealed, with fewer than two weeks left to meet the deadline for Nos. 14-1430 & 14-1431 13

compliance. We denied the university’s motion for an injunc- tion pending the appeal but ordered expedited briefing. On the last day to comply with the regulations, Notre Dame signed the Form 700 and supplied it to its insurer and third-party administrator. 743 F.3d at 551. The appeal proceeded. We noted that Notre Dame’s primary objection was to the regulations surrounding the Form 700 certification. One regulation provides that: the copy of the self-certification [EBSA Form 700] provided by the eligible [to opt out] organization [Notre Dame] to a third party administrator [Meritain] (including notice of the eligible organiza- tion's refusal to administer or fund contraceptive benefits) … shall be an instrument under which the plan is operated, [and] shall be treated as a designa- tion of the third party administrator as the plan administrator under section 3(16) of ERISA for any contraceptive services required to be covered under § 2590.715–2713(a)(1)(iv) of this chapter to which the eligible organization objects on religious grounds. Notre Dame I, 743 F.3d at 552–53 (quoting 29 C.F.R. § 2510.3-16). Notre Dame interpreted that regulation as if its mailing of the Form 700 to its insurer and its third-party administrator were the cause of the provision of contraceptive services to its employees and students, in violation of its religious beliefs. We noted that was not the case. Instead, the Form 700 allows the university to opt out of the provision of objectionable services entirely and the law then places the burden of providing the 14 Nos. 14-1430 & 14-1431

services on the insurer and the third-party administrator. 743 F.3d at 553. In assessing the likelihood of Notre Dame’s success on the merits, we considered and rejected the school’s claim that filling out and mailing the Form 700 is a “substantial burden” on the university’s exercise of religion. 743 F.3d at 554. Notre Dame complained that completing the form and distributing it to the insurer and third-party administrator triggered contraceptive coverage for employees and students, making the university complicit in the provision of objectionable services and burdening the university’s religious exercise. We found that the Form 700 self-certification does not trigger, cause or otherwise enable the provision of contraceptive services: Federal law, not the religious organization's signing and mailing the form, requires health-care insurers, along with third-party administrators of self-insured health plans, to cover contraceptive services. By refusing to fill out the form Notre Dame would subject itself to penalties, but Aetna [the insurer] and Meritain [the third-party administrator] would still be required by federal law to provide the services to the university’s students and employees unless and until their contractual relation with Notre Dame terminated. Notre Dame I, 743 F.3d at 554. We also rejected Notre Dame’s argument that its insurer and third-party administrator would not have been authorized as plan fiduciaries to provide the contraceptive services until the school executed Form 700. Nos. 14-1430 & 14-1431 15

743 F.3d at 554–55. The law and the regulations (and not the Form 700) designate the insurer and third-party administrator as plan fiduciaries who are then obligated by federal law to provide the contraceptive services. 743 F.3d at 555. We also concluded that the contraception regulations do not impose a substantial burden simply because the university must contract with a third party willing to provide (at the behest of the government) the services that Notre Dame finds objectionable. Because that third party did not object to providing the services, we called any such claim speculative and not a ground for equitable relief. We emphasized, in the end, that it was not the Form 700 or anything that Notre Dame was required to do by the regulatory accommodation that caused the university’s employees and students to receive the objec- tionable coverage; rather it was federal law that authorized, indeed required, insurers and third-party administrators to provide coverage. 743 F.3d at 559. Because the true objection was not to actions that the school itself was required to take but rather to the government’s independent actions in mandat- ing contraceptive coverage, we concluded that there was no substantial burden on the university’s religious exercise. 743 F.3d at 559. B. As litigation on the ACA’s contraception requirements has progressed in other cases and other circuits, new regulations have been issued in response to interim orders from the Supreme Court. In Little Sisters of the Poor Home for the Aged, Denver, Colo. v. Sebelius, 134 S. Ct. 1022 (2014), after a district court declined to enjoin the operation of the ACA against a religious organization that did not wish to file the Form 700, 16 Nos. 14-1430 & 14-1431

the Court entered an injunction pending the appeal of that decision: If the employer applicants inform the Secretary of Health and Human Services in writing that they are non-profit organizations that hold themselves out as religious and have religious objections to providing coverage for contraceptive services, the respondents are enjoined from enforcing against the applicants the challenged provisions of the Patient Protection and Affordable Care Act and related regulations pending final disposition of the appeal by the United States Court of Appeals for the Tenth Circuit. To meet the condition for injunction pending appeal, applicants need not use the form prescribed by the Government and need not send copies to third-party administrators. The Court issues this order based on all the circumstances of the case, and this order should not be construed as an expression of the Court's views on the merits. Little Sisters, 134 S. Ct. at 1022. The order, in short, relieved the Little Sisters of their obligation to file the Form 700 so long as they directly notified the government of their objection. Subsequently, the Court entered a similar injunction in a case within our circuit. See Wheaton College v. Burwell, 134 S. Ct. 2806 (2014). After essentially repeating the language from the very short order in Little Sisters, the Court clarified: Nothing in this interim order affects the ability of the applicant's employees and students to obtain, without cost, the full range of FDA approved contra- Nos. 14-1430 & 14-1431 17

ceptives. The Government contends that the appli- cant's health insurance issuer and third-party ad- ministrator are required by federal law to provide full contraceptive coverage regardless whether the applicant completes EBSA Form 700. The applicant contends, by contrast, that the obligations of its health insurance issuer and third-party administra- tor are dependent on their receipt of notice that the applicant objects to the contraceptive coverage requirement. But the applicant has already notified the Government—without using EBSA Form 700—that it meets the requirements for exemption from the contraceptive coverage requirement on religious grounds. Nothing in this order precludes the Government from relying on this notice, to the extent it considers it necessary, to facilitate the provision of full contraceptive coverage under the Act. Wheaton College, 134 S. Ct. at 2807. As with Little Sisters, the order relieved Wheaton College of its obligation to file Form 700 as long as it notified the government directly of its objec- tion. But the government was permitted to use this direct notice to facilitate the coverage required by the ACA. And finally, after the Third Circuit reversed a temporary injunction sought by a religious employer and granted by a district court, the Court again intervened: The application for an order recalling and staying the issuance of the mandate of the Court of Appeals pending the filing and disposition of a petition for a 18 Nos. 14-1430 & 14-1431

writ of certiorari, having been submitted to Justice Alito and by him referred to the Court, the applica- tion as presented is denied. The Court furthermore orders: If the applicants ensure that the Secretary of Health and Human Services is in possession of all information necessary to verify applicants’ eligibility under 26 CFR § 54.9815-2713A(a) or 29 CFR § 2590.715-2713A(a) or 45 CFR § 147.131(b) (as appli- cable), the respondents are enjoined from enforcing against the applicants the challenged provisions of the Patient Protection and Affordable Care Act and related regulations pending final disposition of their petition for certiorari. Nothing in this interim order affects the ability of the applicants’ or their organiza- tions’ employees to obtain, without cost, the full range of FDA approved contraceptives. Nor does this order preclude the Government from relying on the information provided by the applicants, to the extent it considers it necessary, to facilitate the provision of full contraceptive coverage under the Act. See Wheaton College v. Burwell, 573 U. S. ___ (2014). This order should not be construed as an expression of the Court’s views on the merits. Ibid. Justice Sotomayor would deny the application. Zubik v. Burwell, 2015 WL 3947586 (June 29, 2015) (full text found at http://www.supremecourt.gov/search.aspx? filename=/docketfiles/14a1065.htm, last visited September 3, 2015). As a result of these interim orders from the Supreme Court, the regulations have been amended so that objectors may now notify HHS directly rather than filing the Form 700. Nos. 14-1430 & 14-1431 19

And the government may, in turn, facilitate the required contraceptive coverage based on such notice. C. We turn to our recent decisions in Notre Dame II and Wheaton College v. Burwell, 791 F.3d 792 (7th Cir. 2015). In Notre Dame II, we noted that, shortly after filing its suit and immedi- ately before the regulatory deadline, the university signed the Form 700 and sent it to the insurer of its students and the third- party administrator of its employee plan. That action left us wondering what relief Notre Dame sought. Ultimately, we determined, Notre Dame wanted us to enjoin the government from forbidding Notre Dame to bar Aetna and Meritain from providing contraceptive coverage to any of the university's students or employees. Because of its contractual relations with the two companies, which continue to provide health insurance coverage and administra- tion for medical services apart from contraception as a method of preventing pregnancy, Notre Dame claims to be complicit in the sin of contraception. It wants to dissolve that complicity by forbidding Aetna and Meritain—with both of which, to repeat, it continues to have contractual relations—to pro- vide any contraceptive coverage to Notre Dame students or staff. The result would be that the students and staff currently lacking coverage other than from Aetna or Meritain would have to fend for themselves, seeking contraceptive coverage else- where in the health insurance market. 20 Nos. 14-1430 & 14-1431

Notre Dame II, 786 F.3d at 611. The university’s primary objection to the ACA was that its contractual relationship with its insurer and third-party administrator made the school a conduit for the provision of objectionable services. According to Notre Dame, the contraception regulations imposed a substantial burden on it by forcing the university to identify and contract with a third party willing to provide objectionable contraceptive services. 786 F.3d at 611–12. We noted that, although Notre Dame is the final arbiter of its religious beliefs, only the courts may determine whether the law actually forces the university to act in a way that would violate those beliefs. 786 F.3d at 612. The record contained no evidence to support a conduit theory. Nor is it within our usual practice to enjoin non-parties such as Notre Dame’s insurer and third-party administrator. We also rejected Notre Dame’s claim that the regulation requiring employers to provide Form 700 to its insurers was the cause of the provision of contracep- tive services; rather the services are provided because federal law requires the insurers to provide them. Notre Dame II, 786 F.3d at 613–14 (“It is federal law, rather than the religious organization's signing and mailing the form, that requires health-care insurers, along with third-party administrators of self-insured health plans, to cover contraceptive services.”). Because the insurer must provide the services no matter what the employer does, we noted that “signing the form simply shifts the financial burden from the university to the govern- ment, as desired by the university.” 786 F.3d at 615. See supra note 5. We thus re-asserted the core reasoning of our earlier opinion before turning to any effect that Hobby Lobby had on the case. Nos. 14-1430 & 14-1431 21

Hobby Lobby, we noted, involved closely-held for-profit corporations whose owners objected on religious grounds to the contraceptive mandate. The Supreme Court held that the RFRA applied to nonreligious institutions owned by persons with sincerely held religious objections to the ACA’s contra- ception regulations. Hobby Lobby, 134 S. Ct. at 2776–78; Notre Dame II, 786 F.3d at 615. The Court noted that the companies’ objections could be addressed by allowing them to invoke the same accommodation that the government created for religious non-profit employers, namely signing and filing the Form 700. 134 S. Ct. at 2782. The Court left open the issue of whether the accommodation that was adequate for nonreligious, for-profit corporations would be sufficient to protect the rights of religious non-profit employers. As to that issue, we examined various alternative schemes that Notre Dame proposed as possible accommodations and found each of them lacking. We also noted that the Supreme Court had created an alternative to Form 700 by allowing employers to notify the government directly of its objection to the mandate. Notre Dame II, 786 F.3d at 617–18; Wheaton College, 134 S. Ct. at 2806. We rejected Notre Dame’s objections to the Wheaton College alternative notice, citing Bowen v. Roy, 476 U.S. 693 (1986). We noted that the Roy Court rejected Roy's religious objection to the government's use of his daughter's Social Security number for its purposes. The Court held “Roy may no more prevail on his religious objection to the Government's use of a Social Security number for his daughter than he could on a sincere religious objection to the size or color of the Government's filing cabinets.” Roy, 476 U.S. at 700. Notre Dame's objection to the government designating insurers as substitutes to provide contraceptive 22 Nos. 14-1430 & 14-1431

coverage was an analogous challenge to the government's management of its affairs and, accordingly, Notre Dame had not demonstrated a substantial burden to its religious exercise. Notre Dame II, 786 F.3d at 618. In Wheaton College, we similarly rejected a religious school’s objections to the contraception regulations under the RFRA, the First Amendment and the Administrative Procedures Act. 791 F.3d at 801. The college asserted that the government was using the school’s insurance plan and putting additional terms into its contracts with insurers in order to provide the objec- tionable coverage. The college sought an injunction prohibiting the government’s effort to use Wheaton’s plans as the vehicle for making contraceptive coverage available to its employees and students. It objected to notifying its insurers or the government that it claimed a religious exemption, and also to providing the government with the names of its insurers so that the government could then implement the coverage separate from the college. We noted that the ACA and accom- panying regulations do not alter any employer’s insurance plans or contracts. 791 F.3d at 794. Nor is the college being forced to allow the use of its plan to provide objectionable services. The ACA and regulations require only that the college notify either its insurers or the government that it objects, which takes the school out of the loop. 791 F.3d at 795. As in Notre Dame II, we rejected the claim that the provision of notice to insurers or the government somehow triggers or facilitates the provision of objectionable coverage. 791 F.3d at 796. As was the case with Notre Dame, Wheaton also objected to being forced to contract with insurers which, in turn, provided objectionable services, contending that this made the college Nos. 14-1430 & 14-1431 23

complicit in the provision of those services. We saw no complicity in the operation of the law, which makes every effort to separate religious employers from the provision of any objectionable services. We again noted that courts generally do not enjoin non- parties, and Wheaton had not made its insurers parties to the suit. Wheaton also expressed a reluctance to identify its insurers to the government, instead preferring that the govern- ment discover through its own research the names of the insurers. But Wheaton made no connection between the means for identifying the insurers and its religious commitments. We also noted Wheaton’s assertion that its students and employees sign a covenant agreeing to abide by the school’s moral standards, indicating perhaps that Wheaton’s concerns about the ACA are largely academic because the employees and students are unlikely to actually use the services offered. Finally, we rejected Wheaton’s claims under the First Amend- ment, ERISA and the Administrative Procedures Act, all issues which were not argued in the instant appeal, and so we will not address them further. 791 F.3d at 797–800. Before we move on to the plaintiffs’ objections in this case, we note that the case law analyzing the contraceptive mandate is rapidly evolving. Recently, the six other circuits to consider these same issues have all come to the same conclusion as our opinions in Notre Dame and Wheaton College, namely, that the contraceptive mandate, as modified by the accommodation, does not impose a substantial burden on religious organiza- tions under the RFRA. See Catholic Health Care System v. Burwell, — F.3d —, 2015 WL 4665049, *7-*16 (2d Cir. Aug. 7, 2015) (concluding that the accommodation does not impose a 24 Nos. 14-1430 & 14-1431

substantial burden); Little Sisters of the Poor Home for the Aged v. Burwell, 794 F.3d 1151 , 2015 WL 4232096, *16 (10th Cir. 2015), petition for cert. filed, 84 USLW 3056 (U.S. July 24, 2015) (No. 15- 105) (concluding that the mandate does not impose a substan- tial burden on religious exercise under RFRA and affirming the denial of a preliminary injunction in one instance and revers- ing the grant of preliminary injunctions in two others); East Texas Baptist University v. Burwell, 793 F.3d 449, 459 (5th Cir. 2015), petition for cert. filed, 84 USLW 3050 (U.S. July 8, 2015) (No. 15-35) (holding that the ACA does not impose a substan- tial burden under the RFRA and reversing the grant of a preliminary injunction); Geneva College v. Secretary United States Department of Health & Human Servs., 778 F.3d 422, 442 (3d Cir. 2015), petition for cert. filed, 83 USLW 3894 (U.S. May 29, 2015) (Nos. 14-1418 & 14A1065), and stay denied by Zubik v. Burwell, 135 S.Ct. 2924, 2015 WL 3947586 (June 29, 2015) (reversing grant of preliminary injunction and concluding that the accommodation procedures do not impose a substantial burden on religious exercise); Priests for Life v. United States Dep’t of Health & Human Servs., 772 F.3d 229, 256 (D.C. Cir. 2014), petition for cert. filed, 83 USLW 3918 (U.S. June 9, 2015) (No. 14-1453) (affirming denial of injunctive relief and conclud- ing that the ACA’s mandate does not impose a substantial burden on religious exercise); Michigan Catholic Conference v. Burwell, 755 F.3d 372, 390 (6th Cir. 2014), cert. granted, judgment vacated and remanded, 135 S. Ct. 1914 (2015), reissued, — F.3d — 2015 WL 4979692 (6th Cir. Aug. 21, 2015) (because objectors may obtain the accommodation from the contraceptive-coverage requirement without providing, paying for, and/or facilitating access to contraception, the contracep- Nos. 14-1430 & 14-1431 25

tive-coverage requirement does not impose a substantial burden on their exercise of religion).10 No court of appeals has concluded that the contraceptive mandate imposes a substan- tial burden under the RFRA. D. After this court issued its opinion in Notre Dame II, we asked the parties to file position statements addressing the effect of that opinion on this appeal. We turn now to the parties’ position statements as well as the arguments raised in their original briefs. The government, in its original brief, contended that Notre Dame I was controlling. It argued that the plaintiffs are permitted to opt out of providing contraceptive coverage, and that the plaintiffs improperly object to require- ments imposed by the accommodation on third parties rather than on themselves. The government also asserted that it is the province of the court rather than the plaintiffs to determine whether a particular regulation or law “substantially” burdens the plaintiffs’ free exercise of religion under the RFRA. Finally, the government maintained that, even if we were to determine that the regulations impose a substantial burden on the plaintiffs under the RFRA, the government’s interest in

10 The Sixth Circuit released its opinion a few weeks prior to the issuance of Hobby Lobby, but denied rehearing en banc several months later. The Supreme Court subsequently granted the petition for certiorari, vacated the opinion and remanded for further consideration in light of Hobby Lobby. The Sixth Circuit recently reissued and reaffirmed its earlier opinion and filed a supplemental opinion addressing Hobby Lobby. The Sixth Circuit continues to hold that the ACA’s contraception provisions do not impose a substantial burden under RFRA. Michigan Catholic, 2015 WL 4979692, *6- *15. 26 Nos. 14-1430 & 14-1431

providing the coverage is compelling and the regulations are narrowly tailored to meet that interest. In its position statement, the government adds that Notre Dame II rejected all of the arguments raised by the plaintiffs here. Specifically, the government again notes that the regula- tions allow the plaintiffs to opt out of providing the mandated contraceptive services, making them effectively exempt. After objectors opt out, the government tasks third parties with providing the coverage. Moreover, the opt-out does not operate as a trigger or cause for the coverage; rather federal law imposes on third parties the obligation to provide the coverage. Nothing in the ACA or regulations makes the plaintiffs complicit or allows their contracts with insurers or third party administrators to act as conduits for the provision of contraceptive services. The government repeats that, even if the regulations impose a substantial burden on the plaintiffs’ free exercise of religion under the RFRA, the regulations serve a compelling government interest and are the least restrictive means of achieving those interests. According to the govern- ment, our opinion in Notre Dame II demonstrates that the current regulations are narrowly tailored to achieve the compelling interest, and that none of the plaintiffs’ suggested alternatives would be effective. In their opening brief, the plaintiffs argued, as they did below, that the contraception regulations impose a substantial burden on their exercise of religion. The plaintiffs asserted that they exercise their religion “by refusing to take actions in furtherance of a regulatory scheme to provide their employees with access to abortion-inducing products, contraceptives, sterilization, and related education and counseling.” Brief at Nos. 14-1430 & 14-1431 27

29. The plaintiffs maintained that submitting Form 700 renders them complicit in a grave moral wrong because the form has certain legal effects that facilitate the provision of the objection- able services. The accommodation, the plaintiffs added, requires them to amend the documents governing their health plans to provide the very coverage to which they object. The plaintiffs also objected to contracting with and paying premi- ums to insurance companies or third party administrators that are authorized to provide their employees with contraceptive coverage. Moreover, the plaintiffs pointed out that if they fail to comply with the regulations, they will face onerous fines. The plaintiffs asserted that Notre Dame I is distinguishable on the facts, and that Notre Dame I did not address the arguments of the Catholic appellees here that (1) the Diocese is being forced to forgo $200,000 annually in increased premiums in order to maintain its grandfathered status11 to avoid its plan becoming a conduit for objectionable coverage for the employ- ees of Catholic Charities that are enrolled in the Diocese’s health plan; and (2) the mandate has the additional effect of artificially dividing the Catholic Church into a “worship arm” and a “good works arm.” The plaintiffs also maintained in their opening brief that the government’s “substantial burden” analysis incorrectly focuses on the nature of the actions that the regulations require the

11 “Grandfathered plans” are plans that were in existence when the ACA was adopted and that have not made certain changes to the terms of the plans. Grandfathered plans need not comply with the ACA’s coverage requirements. See 42 U.S.C. § 18011; 26 C.F.R. § 54.9815-1251T. Certain increases in premiums could cause a plan to lose its grandfathered status and thus become subject to the ACA’s coverage requirements. 28 Nos. 14-1430 & 14-1431

plaintiffs to take rather than the pressure the government has placed on the plaintiffs to take those actions. They contended that the focus of the analysis should be on the intensity of the coercion applied by the government to act contrary to their religious beliefs. Finally, they asserted that they object to actions they themselves must take under the regulations, not to the actions of third parties. In their position statement, the plaintiffs contend that Notre Dame II is distinguishable on the facts, that it is not binding here, and that it is based on legal errors. Finally, the plaintiffs argue that the strict scrutiny analysis in Notre Dame II is both foreclosed by the government’s concession in this case and inconsistent with circuit precedent. E. We turn to the specific objections raised by the plaintiffs here. They contend that the accommodation does not operate as a true “opt-out” because it requires them to engage in numerous religiously objectionable actions. The actions to which the plaintiffs object fall under two categories: first, the mandate requires them to contract with insurance companies or third-party administrators that are authorized to provide the objectionable coverage and which will provide that coverage once the plaintiffs communicate their objections. Second, they must submit the Form 700 or notify the government directly of their objection. They sincerely believe that the required actions render them complicit in a grave moral wrong because their insurance contracts serve as conduits for the provision of the objectionable services, and the notification triggers or facilitates the provision of objectionable services. They practice their Nos. 14-1430 & 14-1431 29

religion, they assert, by refusing to take actions “in furtherance of” a scheme to provide the objectionable services. And if they decline to engage in these actions, the mandate subjects them to onerous fines. The core of the disagreement between the plaintiffs and the government lies in how we apply the substantial burden test. The plaintiffs cite our decision in Korte for the proposition that the substantial burden test under the RFRA focuses primarily on the intensity of the coercion applied by the government to act contrary to religious beliefs. Korte, 735 F.3d at 683. Citing Hobby Lobby, they also assert that the RFRA allows private religious believers to decide for themselves whether taking a particular action (such as filing the Form 700 or contracting with an insurance company) is connected to objectionable conduct in a way that is sufficient to make it immoral. Hobby Lobby, 134 S. Ct. at 2778. In Korte, we noted that “exercise of religion” means “any exercise of religion, whether or not compelled by, or central to, a system of religious belief.” 735 F.3d at 682; 42 U.S.C. § 2000cc–5(7)(A). A substantial burden on free exercise may arise when the government compels a religious person to perform acts undeniably at odds with fundamental tenets of that person’s religious beliefs, and also when the government places substantial pressure on a person to modify his or her behavior in a way that violates religious beliefs. Korte, 735 F.3d at 682. “Put another way, the substantial-burden inquiry evaluates the coercive effect of the governmental pressure on the adherent's religious practice and steers well clear of deciding religious questions.” Korte, 735 F.3d at 683. Relying on Korte and Hobby Lobby, the plaintiffs urge us to engage in a 30 Nos. 14-1430 & 14-1431

two-step analysis of first identifying the religious belief at issue, and second, determining whether the government has placed substantial pressure on the plaintiffs to violate that belief. The plaintiffs are correct that it is not our province to decide religious questions. Hobby Lobby, 134 S. Ct. at 2778 (the RFRA presents the question of whether the mandate imposes a substantial burden on the ability of the objecting parties to conduct business in accordance with their religious beliefs, but courts have no business addressing whether the religious belief asserted is reasonable); Notre Dame II, 786 F.3d at 612 (an objector is the final arbiter of its religious beliefs); Little Sisters of the Poor, 794 F.3d at —, 2015 WL 4232096, at *19 (substantial- ity does not permit a court to scrutinize the theological merit of a plaintiff's religious beliefs); Geneva College, 778 F.3d at 436 (courts should defer to the reasonableness of a plaintiff’s religious beliefs). The plaintiffs in Hobby Lobby were closely- held, for-profit corporations that were required by the ACA to provide and pay for health insurance which included coverage of certain emergency contraceptives that they believed oper- ated as abortifacients. Similar to the plaintiffs here, they believed that providing the required coverage 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 commis- sion of an immoral act by another.” Hobby Lobby, 134 S. Ct at 2778. Nos. 14-1430 & 14-1431 31

So we defer to the plaintiffs’ sincerely held beliefs regard- ing questions of religion and moral philosophy. But whether the government has imposed a substantial burden on their religious exercise is a legal determination. Notre Dame II, 786 F.3d at 612; Little Sisters of the Poor, 794 F.3d at —, 2015 WL 4232096, at *18; East Texas Baptist University, 793 F.3d at 456–59 & n.33; Geneva College, 778 F.3d at 436; Priests for Life, 772 F.3d at 247–49; Michigan Catholic, 755 F.3d at 385. And we are not required to defer to the plaintiffs’ beliefs about the operation of the law. Notre Dame II, 786 F.3d at 612 (although an objector is the final arbiter of its religious beliefs, it is for the courts to determine whether the law actually forces the objector to act in a way that would violate those beliefs); Little Sisters of the Poor, 794 F.3d at —, 2015 WL 4232096, at *18 (courts need not accept the legal conclusion, cast as a factual allegation, that a plain- tiff’s religious exercise is substantially burdened); Geneva College, 778 F.3d at 436 (courts need not accept an objector’s characterization of a regulatory scheme on its face but may consider the nature of the action required of the objector, the connection between that action and the objector’s beliefs, and the extent to which that action interferes with or otherwise affects the objector’s exercise of religion, all without delving into the objector’s beliefs); Michigan Catholic, 755 F.3d at 385 (although a court may acknowledge that the objectors believe that the regulatory framework makes them complicit in the provision of contraception, the court will independently determine what the regulatory provisions require and whether they impose a substantial burden on the objector's exercise of religion). In this instance, and as was the case in Notre Dame I and II, the plaintiffs misapprehend the operation of federal 32 Nos. 14-1430 & 14-1431

law. As many courts have noted, contraceptive coverage under the ACA results from federal law, not from any actions required by objectors under the accommodations. Notre Dame II, 786 F.3d at 614; and 786 F.3d at 623 (Hamilton, J., concur- ring); Little Sisters of the Poor, 794 F.3d at —, 2015 WL 4232096, at *16; East Texas Baptist, 793 F.3d at 459; Geneva College, 778 F.3d at 437; Michigan Catholic, 755 F.3d at 387. The first action to which the plaintiffs object is filing the Form 700. They assert that the Form 700 is far more than a simple notification or objection, that it instead (1) designates the third party administrator as plan administrator and claims administrator for contraceptive benefits; (2) serves as an instrument under which the plans are operated vis-à-vis contraceptive services; and (3) apprises the third party admin- istrator of its obligations to provide contraceptive coverage. We rejected this very argument in Notre Dame II, holding that treating the mailing of the Form 700 as the cause of the provision of contraceptive services is legally incorrect. 786 F.3d at 613. The Form 700, we noted, has the effect of throwing the entire administrative and financial burden of providing contraception on the insurer and the third party administrator. 786 F.3d at 613–14. As a result, the burden is lifted from the objector’s shoulders. 786 F.3d at 614. “It is federal law, rather than the religious organization's signing and mailing the form, that requires health-care insurers, along with third-party administrators of self-insured health plans, to cover contracep- tive services.” 786 F.3d at 614. See also Little Sisters of the Poor, 794 F.3d at —, 2015 WL 4232096 at *16 & *22-24 (finding that plaintiffs do not “trigger” or otherwise cause contraceptive coverage because federal law, not the act of opting out, entitles Nos. 14-1430 & 14-1431 33

plan participants and beneficiaries to coverage); Geneva College, 778 F.3d at 437–38 (same); Michigan Catholic, 755 F.3d at 387 (same). Moreover, the regulations have been amended during this litigation, and now employers need not file the Form 700. Instead, consistent with the Supreme Court’s interim orders in Little Sisters of the Poor and Wheaton College, the plaintiffs may contact the Department of Health and Human Services directly, alerting the government that they have a religious objection to providing contraceptive coverage, and providing only the name and contact information for their insurers or third party administrators. 80 Fed. Reg. 41342-47 (July 14, 2015). The burden then falls on the government to make appropriate arrangements with the insurer or third-party administrator to provide coverage for contraceptive services. The plaintiffs object to that action as well, asserting that it also makes them complicit in the provision of coverage. But that notification does nothing more than completely remove an objector from the provision of the objectionable services. See Geneva College, 778 F.3d at 436 (“While the Supreme Court reinforced in Hobby Lobby that we should defer to the reason- ableness of the appellees' religious beliefs, this does not bar our objective evaluation of the nature of the claimed burden and the substantiality of that burden on the appellees' religious exercise.”). As we noted in our Notre Dame opinions, the plaintiffs are in the strange position of objecting not to the contraceptive mandate itself but to the accommodation that relieves them of any involvement in the implementation of the contraceptive mandate. Notre Dame I, 743 F.3d at 557–58; Notre Dame II, 786 F.3d at 621 (Hamilton, J., concurring). See also Little 34 Nos. 14-1430 & 14-1431

Sisters of the Poor Home for the Aged v. Burwell, 794 F.3d —, 2015 WL 4232096, *14-15 (10th Cir. 2015) (noting the unusual nature of a claim that attacks the government's attempt to accommo- date religious exercise by providing a means to opt out of compliance with a generally applicable law). [T]he arrangements the government makes to find substitutes for those given the benefit of a religious exemption are imposed as a matter of federal law, not as a result of the exemption itself. The party claiming the exemption is not entitled to raise a religious objection to the arrangements the govern- ment makes for a substitute. Notre Dame II, 786 F.3d at 623 (Hamilton, J., concurring). In short, requiring an employer to notify the government of its objection to the mandate is no more burdensome than the government’s use of a girl’s Social Security number in a benefits program even though her father sincerely believed that the use of the number would harm his daughter’s spirit. See Notre Dame II, 786 F.3d at 618–19 (discussing Bowen v. Roy, 476 U.S. 693 (1986)). So too with the plaintiffs here. The second action to which the plaintiffs object is contract- ing with insurers and third-party administrators who will then provide the objectionable coverage, albeit at no cost to, and without further involvement of, the plaintiffs. The plaintiffs admittedly want to provide their employees and students with health insurance; indeed they have said that it is part of their religious mission to do so. But they wish to provide health insurance without the objectionable coverage. Yet this is exactly what the accommodation allows them to do. Notre Nos. 14-1430 & 14-1431 35

Dame II, 786 F.3d at 621–22 (Hamilton, J., concurring) (once an employer files the Form 700 or notifies the government directly of its religious objection, it can avoid contracting, paying, arranging, or referring for the objectionable contraceptive care); Wheaton College, 791 F.3d at 795–96 (once the college notifies its insurer or the government of its religious objections, the college and its health plans are bypassed). As with the notification requirement, the plaintiffs believe that their contracts further the provision of objectionable services. They assert that the government is using their health plans or altering the terms of their health plans to provide contraceptive coverage. But once they have objected, the government does not use the health plans or contracts at all, much less alter any terms. See Wheaton College, 791 F.3d at 796 (“Call this ‘using’ the health plans? We call it refusing to use the health plans.”). As we noted in Wheaton College: The upshot is that the college contracts with health insurers for contraceptive coverage exclusive of coverage for emergency contraceptives, and the Department of Health and Human Services con- t r a c t s w i t h t h o s e in s u r e r s t o c o ve r emergency-contraceptive benefits. The latter con- tracts are not part of the college's health plans, and so the college is mistaken when it tells us that the government is “interfering” with the college’s contracts with its insurers. The contracts, which do not require coverage of emergency contraception, are unchanged. New contracts are created, to which the college is not a party, between the government and the insurers. 36 Nos. 14-1430 & 14-1431

Wheaton College, 791 F.3d at 796. We rejected any notion of complicity in the provision of contraceptive services arising from the mere existence of a contract to provide health insur- ance without any contraceptive coverage. 791 F.3d at 797. See also Little Sisters of the Poor, 794 F.3d at —, 2015 WL 4232096 at *16 (the de minimis administrative tasks required to opt out of the mandate relieves objectors from complicity); East Texas Baptist, 793 F.3d at 461 (“Under the accommodation, the contracts are solely for services to which the plaintiffs do not object; the contracts do not provide for the insurers and third-party administrators to cover contraceptives, do not make it easier for those entities to pay for contraceptives, and do not imply endorsement of contraceptives.”). To the extent that the act of opting out causes the legal responsibility to provide contraceptive coverage to shift from the plaintiffs to their insurers or third-party administrators, this feature relieves rather than burdens their religious exercise. Little Sisters of the Poor, 794 F.3d at —, 2015 WL 4232096 at *16. As our colleagues in the Tenth Circuit noted, “An opt out religious accommodation typically contemplates that a non-objector will replace the religious objector and take over any legal responsibilities.” Little Sisters of the Poor, 794 F.3d at —, 2015 WL 4232096, *16 n.21; East Texas Baptist, 793 F.3d at 461–62 (RFRA does not entitle plaintiffs to block third parties such as the government or insurers from engaging in conduct with which the plaintiffs disagree); Geneva College, 778 F.3d at 438 n.13 (the provision of contraceptive coverage is not dependent upon the objector’s contract with its insurance company); Michigan Catholic, 755 F.3d at 388 (the government's imposition of an independent obligation on a third party does Nos. 14-1430 & 14-1431 37

not impose a substantial burden on an objector’s exercise of religion). Finally, the Catholic plaintiffs here (namely, the Diocese, Catholic Charities, St. Anne Home, Franciscan Alliance, Specialty Physicians, St. Francis and Sunday Visitor) assert what they characterize as unique RFRA claims that were not presented in the Notre Dame appeal and therefore are not resolved by the Notre Dame opinions. In particular, they argue that the mandate substantially burdens the Diocese’s religious exercise by forcing it to forgo almost $200,000 annually in increased premiums to maintain its grandfathered status so that it may avoid its health plan becoming a conduit for objectionable coverage for Catholic Charities’ employees who are enrolled in its health plan. See note 11 supra. But if the Diocese were to lose its grandfathered status, it would become exempt from the ACA’s contraceptive mandate, and Catholic Charities would be able to opt out of the mandate by employ- ing the accommodation. As we just concluded, that scenario would not impose a substantial burden on the free exercise rights of either the Diocese or Catholic Charities. The Catholic plaintiffs also contend that the mandate has the effect of artificially dividing the Catholic Church into a “worship” arm (the Diocese) and a “good works” arm (the remaining Catholic plaintiffs). Again, though, groups affiliated with the Diocese may opt out of providing contraceptive coverage using the accommodation and thus continue to provide health coverage under the Diocese’s health plan. Both arms of the Church are therefore extricated from the provision of objectionable contraceptive services, albeit through different means. Moreover, any division is created not by the ACA but 38 Nos. 14-1430 & 14-1431

by the Internal Revenue Code that excepts “churches, their integrated auxiliaries, and conventions or associations of churches” from certain requirements. See 26 U.S.C. § 6033(a)(3)(A)(i). It is difficult to see how laws and regulations that grant tax advantages to churches and their integrated auxiliaries somehow impose a substantial burden on affiliates. III. The accommodation does not serve as a trigger or a conduit for the provision of contraceptive services. Notre Dame II, 786 F.3d at 612–15; Wheaton College, 791 F.3d at 795–97. It is the operation of federal law, not any actions that the plaintiffs must take, that causes the provisions of services that the plaintiffs find morally objectionable. The accommodation has the legal effect of removing from objectors any connection to the provision of contraceptive services. As we noted above, every other circuit court to consider the issue of whether the mandate imposes a substantial burden on religious exercise has come to the same conclusion. As a result, the plaintiffs are not entitled to a preliminary injunction against the enforcement of the ACA regulations. If they wish to object, they may either employ the Form 700 or they may notify the Department of Health and Human Services directly. We extend the injunc- tions here for 60 days so that the district court may consider in the first instance the additional arguments that plaintiffs raised in support of injunctive relief. We reverse the district court’s judgments and remand for proceedings consistent with this opinion. REVERSED AND REMANDED. Nos. 14‐1430 & 14‐1431 39

MANION, Circuit Judge, dissenting. The HHS accommodation is the long and winding exten‐ sion cord the government uses to power its contraceptive mandate. It winds through regulations and additions and revisions. The court, through a perfunctory examination, in‐ terprets the accommodation’s twisted framework and holds that it frees the religious nonprofits from having to power the mandate themselves and, thus, does not violate the RFRA. The court is wrong: A thorough examination reveals that the accommodation’s tangled mess is hiding the fact that the extension cord gets its power from the nonprofits’ health plans and must be plugged in before it will work. It also ex‐ poses the fact that the government is forcing the nonprofits to plug in the accommodation themselves by signing the self‐certification or providing the alternative notice. This dissent, as long and detailed as it is, reveals that the accommodation never relieves the religious nonprofits or their health plans from the provision of contraceptive ser‐ vices which burdens their religious exercise. Section I ex‐ plains how the court, as many others have before it, uses a caricature of the HHS accommodation to avoid accepting the nonprofits’ sincerely held religious belief as required by the Supreme Court in Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751 (2014). Section II shows that the nonprofits correctly understand the accommodation’s operation, so that the court must accept their sincerely held religious belief and hold that the accommodation imposes a substantial burden on their religious exercise. Section III demonstrates that the government has utterly failed to prove that the HHS ac‐ 40 Nos. 14‐1430 & 14‐1431

commodation furthers a compelling governmental interest: The government has failed to establish any of the causal links necessary to prove that increasing the availability of contraceptive services will improve the health of women generally, let alone that of the nonprofits’ employees. Fur‐ thermore, the government’s stated interest is overbroad, un‐ derinclusive, and marginal at best. Section IV demonstrates that, even if the government had a compelling interest, the accommodation is not the least restrictive means. For these reasons, Section V concludes that the HHS accommodation violates RFRA, which means the nonprofits have a signifi‐ cant likelihood of success on the merits of their claim and the district court’s preliminary injunction should be affirmed. For the many reasons that follow, I dissent. I. The court refuses to apply RFRA. RFRA prevents the government from substantially bur‐ dening a person’s religious exercise unless the burden on the person is the least restrictive means of furthering a compel‐ ling governmental interest. 42 U.S.C. § 2000bb‐1. The Su‐ preme Court has made it abundantly clear that courts are wholly incompetent to decide whether a governmental ac‐ tion burdens a person’s religious exercise. Rather, courts must accept a person’s sincere belief that it is a burden. Hob‐ by Lobby, 134 S. Ct. at 2778–79. Courts determine whether the burden is substantial, but they do so by examining the level of coercion applied to compel compliance, not what is re‐ quired by that compliance and to what extent it violates the person’s religion. Id. at 2779; Korte v. Sebelius, 735 F.3d 654, 683 (7th Cir. 2013). Thus, the proper analysis is to determine Nos. 14‐1430 & 14‐1431 41

first, that the nonprofits have a sincere belief that compliance with the law would violate their religion, and second, that the pressure applied by the government to coerce compli‐ ance with the law is substantial. The outcome of a thorough and proper analysis is ultimately simple and straightfor‐ ward: As in Hobby Lobby, the government concedes the sin‐ cerely held religious belief and the fines imposed for non‐ compliance are “enormous.” Hobby Lobby, 134 S. Ct. at 2779. So, following Hobby Lobby, the accommodation imposes a substantial burden. That the government labels it an ac‐ commodation makes no difference to the burden it imposes on the nonprofits. The analysis remains the same. The court does not apply these straightforward steps be‐ cause it balks at the idea that we must accept a person’s as‐ sertion that a law burdens their religion. The court fears that such a rule will allow a person to escape any number of reg‐ ulations, including this brave new world of free and univer‐ sal contraceptives, unless the government can meet the strict scrutiny test laid down by RFRA.1 This was the same con‐ cern that prompted the Supreme Court’s decision to limit free exercise protections in Employment Div., Dept. of Human Resources of Ore. v. Smith, 494 U.S. 872 (1990). Hobby Lobby, 134 S. Ct. at 2761–62. Nevertheless, when it enacted RFRA, Congress meant to restore exactly the level of protection to religious exercise that now so concerns the court. Id. at 2761–

1 What goes unsaid by this critique is the conclusion that the non‐

profits’ religious beliefs are less deserving of protection than the gov‐ ernment’s scheme to marginally increase the availability of contraceptive services for certain employees. 42 Nos. 14‐1430 & 14‐1431

62; Korte, 735 F.3d at 671–72. So, foreclosed by the Supreme Court, the court rules as it and many others have before: The court rejects the nonprofits’ sincere belief that compliance with the HHS accommodation is prohibited by their religion by holding that the nonprofits misunderstand the manner in which the accommodation operates. Then, acting as an ex‐ pert theologian, the court holds that the accommodation’s operation as understood by the court is not a substantial burden to the nonprofits’ religious exercise. Ante, at 38; see also Wheaton Coll. v. Burwell, 791 F.3d 792 (7th Cir. 2015); Univ. of Notre Dame v. Burwell, 786 F.3d 606 (7th Cir. 2015) (Notre Dame II); Michigan Catholic Conference v. Burwell, 2015 WL 4979692 (6th Cir. Aug. 21, 2015); Little Sisters of the Poor Home for the Aged v. Burwell, 794 F.3d 1151 (10th Cir. Jul. 14, 2015); East Texas Baptist Univ. v. Burwell, 793 F.3d 449 (5th Cir. Jun. 22, 2015); Geneva College v. Secretary United States Dep’t of Health & Human Servs., 778 F.3d 422 (3d Cir. 2015); Priests for Life v. United States Dep’t of Health & Human Servs., 772 F.3d 229 (D.C. Cir. 2014). But cf. Notre Dame II, 786 F.3d at 626 (Flaum, J., dissenting); Little Sisters, 794 F.3d 1151, 2015 WL 4232096, *41 (Baldock, J., dissenting); Priests for Life v. United States HHS, 2015 U.S. App. LEXIS 8326, *16 (D.C. Cir. May 20, 2015) (en banc denied) (Brown, J. and Kavanaugh, J., dis‐ senting); Eternal Word Television Network, Inc. v. Sec’y, U.S. Dep’t of Health & Human Servs., 756 F.3d 1339, 1340 (11th Cir. 2014) (Pryor, J., concurring). The court does this by improperly judging the nonprofits’ religious beliefs and ignoring the penalties used for compli‐ ance. Had the nonprofits said that they sincerely believe that the HHS accommodation violates their religion and left it at Nos. 14‐1430 & 14‐1431 43

that, perhaps the injunction would remain in place because there would be nothing for the court to attack. But since the nonprofits said that they sincerely believe that the accom‐ modation violates their religion because the accommodation makes them complicit in the provision of contraceptive ser‐ vices, the court has attacked their claim that the law makes them complicit. The court is right that it is “not required to defer to the plaintiffs’ beliefs about the operation of the law.” Ante, at 31. Nevertheless, it is the nonprofits that are right about the operation of the law, not the court. II. The accommodation imposes a substantial burden on the nonprofits’ religious exercise. The court denies that the self‐certification and alternative notice process trigger the provision of contraceptive cover‐ age. According to the court, it is federal law, not the self‐ certification form or alternative notice, which results in the contraceptive coverage. The court says that self‐certification throws the burden of contraceptive coverage on to the non‐ profits’ health insurance issuers (insurers) and third party administrators (TPAs). Ante, at 32. But how does this lift the burden off the nonprofits when the accommodation imposes the “free” contraceptive coverage requirement only on the insurers and TPAs that the nonprofits have hired? In spite of that imposition, the court also denies that the accommoda‐ tion uses the nonprofits’ health plans to provide the contra‐ ceptive coverage. Instead, it says that the government con‐ tracts with the insurers and TPAs to provide the coverage to only the beneficiaries on the nonprofits’ health plans. Ante, at 35. But, given that connection with the nonprofits’ health 44 Nos. 14‐1430 & 14‐1431

plans, how can the provision of coverage be completely in‐ dependent of those same plans? The court can only make such sweeping claims by ignor‐ ing the true operation of the accommodation and the legal consequences the government has attached to the self‐ certification and alternative notice. The court may think that the nonprofits “throw” their burden onto their insurers and TPAs, but it ignores who is forced to do the throwing and who ultimately carries the burden once thrown. A close ex‐ amination of the manner in which the regulations actually operate reveals that the government’s promise of accommo‐ dation is illusory. The nonprofits’ claim that the HHS ac‐ commodation makes them complicit in the provision of con‐ traceptive coverage becomes obvious. A. The self‐certification form and alternative notice trigger the coverage of contraceptive services. The court holds that the self‐certification and alternative notice are simply signs that the nonprofits have opted out of providing contraceptive coverage, and once the sign is made known the law obliges the nonprofits’ insurers and TPAs to provide the unwanted coverage. Ante, at 32–33. In reality, once the nonprofits formally object, they are opted in. The self‐certification and alternative notice do more than give notice of the nonprofits’ objections. And they are much more than de minimis paperwork necessary to effectuate the non‐ profits’ objection. They create the insurers’ and TPAs’ obliga‐ Nos. 14‐1430 & 14‐1431 45

tion to provide the contraceptive coverage.2 For a nonprofit with a self‐insured plan, the effect of the self‐certification and alternative notice is abundantly clear: the government makes them legal instruments under which the nonprofit’s health plan is operated. This then allows the regulations to treat them as legal designations of the TPA as plan adminis‐ trator and claims administrator for coverage of contraceptive services under the nonprofit’s health plan.3 Only a nonprofit can designate its plan administrator.4 The government’s abil‐ ity to define how a plan administrator is designated does not give it the power to designate who will be a plan administra‐

2 45 C.F.R. § 147.131(c)(2)(i) (“A group health insurance issuer that re‐

ceives a copy of the self‐certification or notification … must … [p]rovide sepa‐ rate payments for any contraceptive services for plan participants and beneficiaries for so long as they remain enrolled in the plan.” (emphasis added)); 29 C.F.R. § 2590.715–2713A(c)(2)(i) (identical requirement for TPAs); 78 Fed. Reg. 39878 (listing among the key elements of the accom‐ modation the need for eligible organizations with insured group health plans to self‐certify and that it is the “issuer that receives a self‐ certification” that must comply with the accommodation’s require‐ ments); 78 Fed. Reg. 39880 (“A third party administrator that receives a copy of the self‐certification … must provide or arrange separate pay‐ ments for contraceptive services … .”). 3 78 Fed. Reg. 39879 (“The self‐certification … will be treated as a

designation of the third party administrator(s) as plan administrator and claims administrator for contraceptive benefits pursuant to section 3(16) of ERISA.”); 29 C.F.R. § 2510.3–16 (defining the term plan administrator to include the regulatory treatment of the self‐certification and alterna‐ tive notice as acts of designation and declaring that each “shall be an in‐ strument under which the plan is operated”). 4 29 U.S.C. § 1102(a)(2). 46 Nos. 14‐1430 & 14‐1431

tor.5 For the TPA to have the necessary authority to provide coverage for contraceptive services, the nonprofit must desig‐ nate the TPA as a plan administrator.6 Such an act would ob‐ viously violate the nonprofit’s religion. So the government has loaded the self‐certification and alternative notice with the legal significance of designating the TPA. It is not the op‐ eration of law. It is the acts of self‐certification and alterna‐ tive notice that designate the TPA and facilitate the provision of the unwanted contraception coverage. Without possession of the self‐certification or alternative notice, the TPA cannot receive reimbursement for the provision of contraceptives.7 In sum, the government can only require the nonprofits’ TPAs to cover contraceptive services if the nonprofits give the government the legal authority to do so. The government has hidden that legal authority in self‐certification and alter‐ native notice. For insurers the situation is not as clear, but it is not the less burdensome. The law requires insurers to include con‐ traceptive coverage in every health plan they offer.8 (Insurers will not, however, provide something for which they are not

5 29 U.S.C. § 1002(16)(A).

6 78 Fed. Reg. 39880 (“The third party administrator serving as the

plan administrator for contraceptive benefits ensures that there is a party with legal authority to arrange for payments for contraceptive services and administer claims in accordance with ERISA’s protections for plan participants and beneficiaries.”). 7 See infra note 18.

8 45 C.F.R. § 147.131(a)(1). Nos. 14‐1430 & 14‐1431 47

paid.9) The self‐certification and alternative notice permit an insurer to offer a health plan that appears not to include con‐ traceptive coverage. But this is on the condition that the in‐ surer still provides the coverage itself in the form of direct payments.10 The assertion that it is the operation of law that designates an objecting nonprofit’s insurer as the replace‐ ment is misleading. It ignores the fact that but for the non‐ profit’s hiring of the insurer, and the nonprofit’s continuing contractual relationship with it, the government (or the op‐ eration of law) would not make any designation. Without the objection and designation by the nonprofits, the insurer is not required to act at all, despite the court’s claim to the contrary. The government has turned the act of objecting into the act of designating and it cannot escape the consequences of that conflation by calling it an act of law. This is not like the case of a conscientious objector who objects and the government finds a replacement. Under the regulations, the government does not find the replacement, the nonprofit does. The designation does not take place un‐ less the nonprofit either delivers the self‐certification form to its insurer or TPA, or uses the alternative notice to inform the government who its insurer or TPA is and which health plan is at issue. By insisting that the nonprofit deliver the form or supply the plan information for the government’s use, the government uses the objecting nonprofit to do its

9 See infra note 21.

10 See supra note 2. 48 Nos. 14‐1430 & 14‐1431

dirty work. The government has not provided an exit—it of‐ fers a revolving door with only one opening.11 Furthermore, this is not like the case of a conscientious objector who refuses to object and goes to jail, and the gov‐ ernment still finds a replacement. If the nonprofits refuse to self‐certify or provide the alternative notice and instead pay the huge fines, their insurers and TPAs will not automatical‐ ly provide the contraception coverage. To comply with the law, the insurers would refuse to sell plans without the cov‐ erage, while the TPA would refuse to provide their services. In spite of the huge monetary penalties, the nonprofits would still be prevented from providing health plans for their employees, which they have asserted is also a violation of their religious beliefs. So the no‐win substantial burden would hit them on both sides. The court has implied that requiring the nonprofit to identify its insurer (or TPA) is merely the most efficient means for the government to achieve its objective, Wheaton, 791 F.3d at 798, but efficiency does not excuse the substantial burden imposed by the requirement. Identifying its insurer so that the government can instruct that insurer to provide

11 This is not the case of a conscientious objector walking into the

draft board, voicing his objection, being excused, and walking out. For the analogy to fit the HHS accommodation, the draft board must decide that every objector will be replaced by the objector’s friend, and the ob‐ jector’s objection is only effective if the objector delivers written notice of his objection to his friend or tells the draft board who his friend is and where the board can find him. Then, the objector must send his friend money so that that his friend will remain his friend for the purpose of being his replacement. Nos. 14‐1430 & 14‐1431 49

contraceptive coverage is just as burdensome to the nonprof‐ it as if it had to pick its own replacement, because it has done just that by hiring its insurer and then objecting to the coverage requirement. That the nonprofits could not object if the government, on its own, were to find a replacement in‐ surer and discover to which employees it had to provide the coverage is not relevant. Of course the nonprofits would not have an objection to the government contracting with a third party to provide the contraceptive coverage to other third parties. They believe the provision of objectionable contra‐ ceptives is always immoral, but they know they have no le‐ gal means to stop the government from contracting with third parties. That is not what is happening here. The gov‐ ernment is using the nonprofits, their health plans, and their contractual relationships with their insurers and TPAs, to provide the contraception coverage to which they object. B. The accommodation uses the nonprofits’ health plans. The HHS accommodation requires significantly more in‐ volvement on the part of the nonprofits and their health plans than the court relates. For starters, the accommodation does not create independent policies or contracts. In fact, as the nonprofits assert, the accommodation relies wholly on the existing contract between the nonprofits and the insurers and TPAs to provide separate payments directly to the non‐ profits’ health plan beneficiaries.12 The accommodation must

12 78 Fed. Reg. 39874 (“[T]he accommodations established under these final regulations do not require the issuance of a separate excepted bene‐ fits individual health insurance policy covering contraceptive services … but 50 Nos. 14‐1430 & 14‐1431

use the existing insurance contracts to issue payments be‐ cause separate policies would violate insurance laws.13 The separate payments are only provided so long as an employ‐ ee is enrolled in the nonprofit’s health plan, thus requiring the nonprofits’ health plans to determine eligibility.14 The ac‐ commodation also relies on the nonprofits’ health plans’ en‐ rollment procedures. The insurers and TPAs must provide notice of the separate payments when they provide notice of the other benefits under the nonprofits’ health plans.15 The separate payments can be limited to the same provider net‐ work as the other plan benefits.16 The end result is that the

instead require a simpler method of providing direct payments for con‐ traceptive services.” (emphasis added)). 13 78 Fed. Reg. 39876 (“As the payments at issue derive solely from a

federal regulatory requirement, not a health insurance policy, they do not implicate issues such as issuer licensing and product approval re‐ quirements under state law … ”). 14 45 C.F.R. § 147.131(c)(2)(i)(B) (insurers must “[p]rovide separate

payments for any contraceptive services … for plan participants and beneficiaries for so long as they remain enrolled in the plan.”); 29 C.F.R. § 2590.715–2713A(c)(2)(i)(B) (identical regulatory requirements for TPAs). 15 78 Fed. Reg. 39881 (“The notice [regarding the provision of contra‐

ceptive services] must be provided contemporaneous with (to the extent possible), but separate from, any application materials distributed in connec‐ tion with enrollment (or re‐enrollment) in coverage … .” (emphasis add‐ ed)). 16 78 Fed. Reg. 39877 (“[A]n issuer … may require that contraceptive

services be obtained in‐network (if an issuer has a network of providers) in order for plan participants and beneficiaries to obtain such services without cost sharing.” (emphasis added)). Nos. 14‐1430 & 14‐1431 51

contraceptive services become a de facto benefit under the nonprofits’ health plans.17 The government admitted as much when it stated that it was by design that the accom‐ modation makes the provision of contraceptive coverage “seamless[]” with the other plan benefits. Gov’t Supp and Reply Br., 14. These circumstances sharply conflict with the court’s conclusion that the accommodation does not use the nonprofits’ health plans and “makes every effort to separate religious employers from the provision of any objectionable services.” Ante, at 23. “[E]very effort” does not disguise the fact that the offensive provision is inseparably imbedded in the nonprofits’ health plan. Id. A further indication that the accommodation uses the nonprofits’ health plans is the fact that the only way an em‐ ployee receives coverage for contraceptive services under the accommodation is to enroll in the objecting nonprofit’s health plan. An employee cannot reject coverage under the nonprofit’s plan and still receive coverage under the accom‐ modation. The coverage under the accommodation is not separate from the coverage under the nonprofit’s health plan. It is the employee’s status as a beneficiary of the non‐ profit’s health plan, not as an employee, that entitles the em‐ ployee to coverage under the accommodation. Simply being hired as an employee is not enough to receive coverage; an

17 78 Fed. Reg. 39880 (“[T]he self‐certification … identifies the limited

set of plan benefits (that is, contraceptive coverage) that the employer re‐ fuses to provide and that the third party administrator must therefore provide or arrange for an issuer or another entity to provide.” (emphasis added)). 52 Nos. 14‐1430 & 14‐1431

employee must enroll in the nonprofit’s health plan. Cf. Notre Dame II, 786 F.3d at 617. C. The nonprofits are involved in the payment for con‐ traceptive services. Finally, there is the matter of payment. For TPAs, the self‐ certification and alternative notice act as authorizations for payment, without which the TPAs cannot receive reim‐ bursement from the government for payments made under the accommodation.18 The government assumed that the re‐ imbursements for TPAs would not be passed on to the non‐

18 Payments for contraceptive services provided by TPAs under the

HHS accommodation are funded through an adjustment (i.e., discount) to the federally‐facilitated exchange (FFE) user fee. See 78 Fed. Reg. 39882. The FFE user fee is paid by insurance issuers that participate in a federal health care exchange to support the operations of the exchange. See 78 Fed. Reg. 15412; 45 C.F.R. § 156.50(c). The amount of the adjust‐ ment is equal to the total amount of the payments made for contracep‐ tive services provided by the TPA plus an allowance of at least 10 percent for administrative costs. 45 C.F.R. § 156.50(d)(3). To receive the FFE user fee adjustment, a TPA must submit to HHS “[a]n attestation that the payments for contraceptive services were made in compliance with 26 CFR 54.9815‐2713A(b)(2) or 29 CFR 2590.715‐2713A(b)(2).” 45 CFR § 156.50(d)(2)(iii)(E). Both the provisions cited by § 156.50 provide that the TPA will provide the separate payments for contraceptive services “[i]f a third party administrator receives a copy of the self‐certification from an eligible organization or a notification.” 26 CFR § 54.9815‐2713AT(b)(2) and 29 CFR § 2590.715‐2713A(b)(2). Moreover, § 156.50 requires a TPA which receives an adjustment to maintain for 10 years and make availa‐ ble upon request “[a] copy of the self‐certification referenced in 26 CFR 54.9815‐2713A(a)(4) or 29 CFR 2590.715‐2713A(a)(4) for each self‐insured plan with respect to which an adjustment is received.” 45 CFR § 156.50(d)(7)(i). Nos. 14‐1430 & 14‐1431 53

profits but, as more nonprofits are forced to use the accom‐ modation and more contraceptive services are provided un‐ der the accommodation, that assumption is unlikely to prove true.19 For insurers there is ostensibly no reimbursement be‐ cause the government claims the cost of contraceptive ser‐ vices will be offset by the reduction in unintended pregnan‐ cies.20 Whether this claim is true will be hard to determine because the regulations allow insurers to recapture costs for contraceptive services provided under the accommodation through what is called the “risk corridor program.”21 The

19 The government assumed that the adjustments granted under the

accommodation for 2014 would be small enough to have no impact on the fee. 78 Fed. Reg. 39882. However, the FFE user fee will have to be increased to cover 1) more adjustments as more nonprofits are forced to take advantage of this accommodation, and 2) greater adjustments be‐ cause the HHS mandate incentivizes more expensive forms of contracep‐ tion. An increase in the FFE user fee will eventually be recouped through an increase in premiums, albeit an increase across the insurance issuer’s entire portfolio, but the nonprofits may be in that same portfolio. 20 78 Fed. Reg. 39877 (“The Departments continue to believe, and

have evidence to support, that, with respect to the accommodation for insured coverage established under these final regulations, providing payments for contraceptive services is cost neutral for issuers.”). 21 78 Fed. Reg. 39878 (“[A]n issuer of group health insurance cover‐

age that makes payments for contraceptive services under these final regulations may treat those payments as an adjustment to claims costs for purposes of medical loss ratio and risk corridor program calculations. This adjustment compensates for any increase in incurred claims associ‐ ated with making payments for contraceptive services.”). The “risk cor‐ ridor program” is a complex cost‐sharing program in which insurers with healthier beneficiaries cover the costs of insurers which either failed to raise premiums or could not raise premiums enough to cover more 54 Nos. 14‐1430 & 14‐1431

program is temporary, but since the HHS accommodation was enacted during the program’s implementation, it will be difficult to determine how the accommodation’s separate payments affect premiums. Nevertheless, if it were true that payments for contraceptive services are cost‐neutral, then the premiums that would otherwise go toward childbirths are instead used for contraceptive services in order to reduce

costly beneficiaries, including those that received separate payments for contraceptive services. See 45 C.F.R. § 153.500 et seq.; 78 Fed. Reg. 7235 (“Section 1342 of the Affordable Care Act directs the Secretary to estab‐ lish a temporary risk corridors program that provides for the sharing in gains or losses resulting from inaccurate rate setting from 2014 through 2016 between the Federal government and certain participating plans.”); see also 78 Fed. Reg. 72323 (“In 2014, HHS will also operationalize the premium stabilization programs established by the Affordable Care Act—the risk adjustment, reinsurance, and risk corridors programs— which are intended to mitigate the impact of possible adverse selection and stabilize the price of health insurance in the individual and small group markets.”). The program is supposed to pay for itself, but the reg‐ ulations allow the government to use appropriated funds to cover insur‐ er loses. See 79 Fed. Reg. 30260 (“In the unlikely event of a shortfall for the 2015 program year, HHS recognizes that the Affordable Care Act requires the Secretary to make full payments to issuers. In that event, HHS will use other sources of funding for the risk corridors payments, subject to the availability of appropriations.”). Perhaps this is why insur‐ ers do not object to the HHS accommodation. Insurers know that the federal government will ultimately bear the burden of covering the costs for contraceptive services they are unable to recoup. The risk corridor program has been criticized as a health insurer bailout program. See No‐ am N. Levey, Critics call Obama funding plan for health insurer losses a ‘bailout’, L.A. TIMES, May 21, 2014, http://www.latimes.com/nation/la‐ na‐insurance‐bailout‐20140521‐story.html (last visited Sept. 3, 2015). Nos. 14‐1430 & 14‐1431 55

childbirths because the nonprofits’ premiums are the only source of funding. This is also an objectionable outcome. D. The proper substantial burden analysis: the court must accept the nonprofits’ sincere belief that the accommo‐ dation violates their religion because the nonprofits under‐ stand its operation. The HHS accommodation is a purposely complicated act of bureaucratic legalese and accounting tricks that enables the government to claim that the objecting nonprofits have nothing to do with the provision of contraceptive services. Yet, as shown in much detail above, the accommodation in‐ fects the nonprofits’ health plans with an offensive provision that eradicates their purpose, which is the exercise of the nonprofits’ religion. It is the nonprofits which understand the operation of the HHS accommodation, not the court, and we must accept their sincere belief that it violates their reli‐ gion. The accommodation imposes a substantial burden be‐ cause the nonprofits have a sincere belief that compliance with the law violates their religion and the penalties applied by the government to coerce compliance are enormous. III. The accommodation does not further a compelling gov‐ ernmental interest. The government must grant the nonprofits an exemption from the accommodation unless “it demonstrates that appli‐ cation of the burden to the person—(1) is in furtherance of a compelling governmental interest; and (2) is the least restric‐ tive means of furthering that compelling governmental in‐ terest.” 42 U.S.C. § 2000bb‐1 (emphasis added). “This re‐ quires us to look beyond broadly formulated interests and to 56 Nos. 14‐1430 & 14‐1431

scrutinize the asserted harm of granting specific exemptions to particular religious claimants—in other words, to look to the marginal interest in enforcing the contraceptive mandate in these cases.” Hobby Lobby, 134 S. Ct. at 2779 (internal quo‐ tation and alteration marks omitted). “RFRA creates a broad statutory right to case‐specific exemptions from laws that substantially burden religious exercise even if the law is neu‐ tral and generally applicable, unless the government can sat‐ isfy the compelling‐interest test.” Korte, 735 F.3d at 671. “In short, RFRA operates as a kind of utility remedy for the inev‐ itable clashes between religious freedom and the realities of the modern welfare state, which regulates pervasively and touches nearly every aspect of social and economic life.” Id. at 673. “Congress’s express decision to legislate the compelling interest test indicates that RFRA challenges should be adju‐ dicated in the same manner as constitutionally mandated applications of the test … .” Gonzales v. O Centro Espirita Be‐ neficente Uniao do Vegetal, 546 U.S. 418, 430 (2006). Thus, the government “must specifically identify an ‘actual problem’ in need of solving, and the curtailment of [the right] must be actually necessary to the solution.” Brown v. Entmʹt Merchs. Assʹn, 131 S. Ct. 2729, 2738 (2011) (citation omitted). This re‐ quires a “high degree of necessity.” Id. at 2741. The govern‐ ment must show a “direct causal link.” Id. at 2738. The gov‐ ernment’s “predictive judgment” is insufficient, “and be‐ cause it bears the burden of uncertainty, ambiguous proof will not suffice.” Id. at 2738–39. (citation omitted). The gov‐ ernment must prove that what it seeks to regulate actually causes the harm it wishes to prevent; evidence of a correlation Nos. 14‐1430 & 14‐1431 57

is insufficient. Id. at 2739. “[S]tudies [that] suffer from signif‐ icant, admitted flaws in methodology” fail to provide this proof. Id. If the regulation is underinclusive it is a sign that the governmental interest is not compelling. Id. at 2740. Put another way, “only those interests of the highest order and those not otherwise served” can be considered compelling. Wisconsin v. Yoder, 406 U.S. 205, 215 (1972). But, “a law can‐ not be regarded as protecting an interest of the highest order when it leaves appreciable damage to that supposedly vital interest unprohibited.” Church of Lukumi Babalu Aye v. City of Hialeah, 508 U.S. 520, 547 (1993) (internal quotation and al‐ teration marks omitted). Finally, the government does not have a compelling interest in “[f]illing the remaining modest gap” or in “each marginal percentage point by which its goals are advanced.” Brown, 131 S. Ct. at 2741, n.9. The government asserts the same interest in furtherance of the HHS accommodation that it asserts in furtherance of the HHS contraceptive mandate, namely, the increased availability of contraceptive services to improve the health of women. The government also says that it wishes to increase the availability of contraceptive services to equalize the pro‐ vision of preventive care for women and men so that women can participate in the workforce and society on an “equal playing field with men.” The latter interest boils down to a concern for women’s health. The government claims that the inequality stems from the additional cost of contraception and that the additional cost can deter women from using contraceptives, thus allowing the negative health outcomes that prevent women from achieving equal economic status. 77 Fed. Reg. 8728. 58 Nos. 14‐1430 & 14‐1431

To justify increasing the availability of contraception to improve the health of women, the government relies exclu‐ sively on the Institute of Medicine’s 2011 study, Clinical Pre‐ ventive Services for Women: Closing the Gaps (IOM Study). The IOM Study is a 235‐page study of the current preventative services available for women. Only eight pages of the study deal with the issue of contraceptive services. IOM Study, 102–09. The study does not claim that contraceptives im‐ prove women’s health generally, but that they prevent cer‐ tain negative health outcomes associated with unintended pregnancies. See Priests for Life, 772 F.3d at 261 (“A core rea‐ son the government sought under the ACA to expand access to contraception is that use of contraceptives reduces unin‐ tended pregnancies.”). The government’s interest advanced by the accommodation, then, is best identified as increasing the availability of contraceptive services in order to prevent the negative health outcomes caused with unintended preg‐ nancies. When put to the test, the government’s interest fails to prove compelling.22 A. A lack of available contraception and unintended pregnancies are not actual problems in need of solving. The HHS accommodation relies on a lengthy chain of causality: 1) the accommodation will make contraceptives more available by removing administrative and cost bur‐ dens; 2) if contraceptives are more available, then more

22 For a comprehensive explanation of how the government’s inter‐

est thoroughly fails the compelling interest test, see generally Helen Al‐ varé, No Compelling Interest: The “Birth Control” Mandate and Religious Freedom, 58 VILL. L. REV. 379 (2013). Nos. 14‐1430 & 14‐1431 59

women will use them; 3) if more women use contraceptives, then there will be fewer unintended pregnancies; and 4) if there are fewer unintended pregnancies, then there will be fewer of the negative health outcomes associated with them. The government, therefore, must prove more than the exist‐ ence of negative health outcomes. It must prove first, that unintended pregnancies cause the negative outcomes; sec‐ ond, that contraceptive use will cause fewer unintended pregnancies; and third a higher availability of contraceptives will cause more women to use them. The IOM Study fails to prove these “direct casual links.” Brown, 131 S. Ct. at 2738. Instead, the study shows merely a correlation. First, the study admits that “for some outcomes [of unin‐ tended pregnancy], research is limited.” Id. at 103. It then proceeds to describe the outcomes correlated with unintend‐ ed pregnancies: outcomes that “may” or “may not” happen, are “more” or “less likely,” have been “reported,” and have “increased odds,” or are “associated with.” IOM Study 103. Second, the study discusses “evidence of [contraceptive] method effectiveness,” but does not prove that increasing the use of even an effective contraceptive causes fewer unin‐ tended pregnancies. This is because such a simple correla‐ tion does not take into account the factors that inhibit perfect use of contraception or the societal changes that result from increased reliance on contraception.23 Rather than prove that

23 See Alvaré, supra note 22, at 408–411 for a discussion of the “grow‐

ing body of scholarship … indicating that the persistence or worsening of high rates of unintended pregnancy, abortion, and sexually transmit‐ ted diseases, and also our nation’s high rates of nonmarital births (the 60 Nos. 14‐1430 & 14‐1431

greater contraceptive use causes fewer unintended pregnan‐ cies, the study only states that “evidence exists” that it does. Id. at 105. The IOM study bases this statement on two other studies, but they are insufficient to provide the necessary ev‐ idence.24 According to the study, “[i]t is thought that greater use of long‐acting, reversible contraceptive methods— including intrauterine devices and contraceptive implants that require less action by the woman and therefore have lower use failure rates—might help further reduce unintend‐ ed pregnancy rates.” Id. at 108 (emphasis added; citation omitted). Third, the study fails to prove that increasing the availa‐ bility of contraceptives will cause an increase in their use, but concludes that “[t]he elimination of cost sharing for con‐ traception therefore could greatly increase its use, including use of the more effective and longer‐acting methods, espe‐ cially among poor and low‐income women most at risk for unintended pregnancy.” IOM Study, 109 (emphasis added). However, the conclusion that eliminating cost sharing “could” increase its use is based on two studies, neither of which concerned contraceptive services specifically. The first concerned preventative and primary care services generally, and the second concerned mammograms. Id. The final claim the study makes is that “when out‐of‐pocket costs for con‐

chief predictor of female poverty), are the ‘logical’ result—in economic and psychological terms—of the new marketplace for sex and marriage made possible by increasingly available contraception (in some cases, combined with available abortion).” 24 Alvaré, supra note 22, at 399‐405. Nos. 14‐1430 & 14‐1431 61

traceptives were eliminated or reduced, women were more likely to rely on more effective long‐acting contraceptive methods.” Id. But, a review of the study underlying that con‐ clusion reveals that “[w]e cannot be certain that the changes in procurement were solely due to the removal of cost to the patient, but there was a shift toward prescribing the most effective methods ([intrauterine contraceptives] and injecta‐ ble contraceptives) and a substantial increase in prescribing of [emergency contraceptive pills].”25 So, not only was the study inconclusive, it is ambiguous regarding the IOM Study’s intended purpose because a substantial increase in emergency contraceptive pills would seem to follow from a decrease in regular contraceptive use. On the whole, the IOM study’s lack of causality renders the government’s claim that it must increase the availability of contraceptives noth‐ ing more than a “predictive judgment.” Brown, 131 S. Ct. at 2738. Another reason the IOM Study fails to prove “an ‘actual problem’ in need of solving” is because it is overbroad. Brown, 131 S. Ct. at 2738. The study starts with the estimation that “[i]n 2001, … 49 percent of all pregnancies in the United States were unintended,” but the study defines an unintend‐ ed pregnancy as one that is “unwanted or mistimed at the time of conception.” IOM Study, 102. This definition includes pregnancies that were unwanted at the time of conception, but still wanted when the mother discovered she was preg‐ nant, and mothers who intended to become pregnant, but

25 Debbie Postlethwaite, et al., A comparison of contraceptive procure‐

ment pre‐ and post‐benefit change, 76 CONTRACEPTION 360, 364 (2007). 62 Nos. 14‐1430 & 14‐1431

did not intend to become pregnant by the specific conjugal act that resulted in conception. The government has zero in‐ terest in preventing these pregnancies. Under the study’s overbroad definition, “all sexually active women with re‐ productive capacity are at risk for unintended pregnancy.” Id. at 103. Aside from the study’s problems with its own def‐ inition, unintended pregnancies are an extremely difficult thing to quantify.26 Overall, the IOM Study lacks the necessary quality and rigor. It heavily relies on studies from biased organizations, such as the Guttmacher Institute and the journal CONTRACEPTION, and offers no consideration of competing studies. Id. at 102–109. The study’s own dissenting opinion says it best: Readers of the Report should be clear on the fact that the recommendations were made without high quality, systematic evidence of the preventive nature of the services consid‐ ered. Put differently, evidence that use of the services in question leads to lower rates of dis‐ ability or disease and increased rates of well‐ being is generally absent. The view of this dissent is that the commit‐ tee process for evaluation of the evidence lacked transparency and was largely subject to the preferences of the committee’s composition. Troublingly, the process tended to result in a

26 Alvaré, supra note 22, at 396–97. Nos. 14‐1430 & 14‐1431 63

mix of objective and subjective determinations filtered through a lens of advocacy. An abiding principle in the evaluation of the evidence and the recommendations put forth as a conse‐ quence should be transparency and strict ob‐ jectivity, but the committee failed to demon‐ strate these principles in the Report. This dis‐ sent views the evidence evaluation process as a fatal flaw of the Report particularly in light of the importance of the recommendations for public policy and the number of individuals, both men and women, that will be affected. IOM Study, 232–33. The study itself shows that the lack of available contra‐ ceptive services is not a problem in need of solving. Accord‐ ing to the IOM Study, “[c]ontraceptive coverage has become standard practice for most private insurance and federally funded insurance programs.” Id. at 108. Further, “[s]ince 1972, Medicaid, the state‐federal program for certain low‐ income individuals, has required coverage for family plan‐ ning in all state programs and has exempted family planning services and supplies from cost‐sharing requirements.” Id. Finally, [C]omprehensive coverage of contraceptive services and supplies [is] “the current insur‐ ance industry standard,” with more than 89 percent of insurance plans covering contracep‐ tive methods in 2002. A more recent 2010 sur‐ vey of employers found that 85 percent of large 64 Nos. 14‐1430 & 14‐1431

employers and 62 of small employers offered coverage of FDA‐approved contraceptives. Id. at 109 (citations omitted). Not only are contraceptive ser‐ vices already widely available, but they are also already widely used: “More than 99 percent of U.S. women aged 15 to 44 years who have ever had sexual intercourse with a male have used at least one contraceptive method.” IOM Study, 103 (citation omitted). According to the Centers for Disease Control and Prevention, contraceptive use is “virtu‐ ally universal among women of reproductive age.”27 The study similarly fails to prove that there is a need to increase the availability of contraceptives to alleviate “the increased risk of adverse pregnancy outcomes for pregnan‐ cies that are too closely spaced” or for “women with certain chronic medical conditions” who “may need to postpone pregnancy” and “women with serious medical conditions” for whom “pregnancy may be contraindicated.” IOM Study, 103. Amazingly, the study does not even pretend to demon‐ strate a causal link in these circumstances, relying instead on the reader to make the inference mistakenly. The study hopes the reader ignores the common sense fact that women in these circumstances have a higher incentive to use contra‐ ceptives if that is their chosen method to prevent these out‐ comes.

27 CDC, “Advance Data No. 350, Dec. 10, 2004: Use of Contraception

and Use of Family Planning Services in the United States: 1982‐2002”, http://www.cdc.gov/nchs/data/ad/ad350.pdf (last visited Sept. 3, 2015). Nos. 14‐1430 & 14‐1431 65

The study offers no evidence regarding the effects that extra paperwork or other administrative and logistical ob‐ stacles would have on contraceptive availability or use. Such a finding is absolutely necessary for the government to as‐ sert that it has a compelling interest in using the nonprofits’ health plans so that the coverage for contraceptive services will be “seamless.” Instead, the IOM Study’s conclusions are limited to the elimination of cost‐sharing and provide no reason why a government‐run option would not work equal‐ ly as well as the HHS accommodation. Finally, the IOM Study does not concern the employees of the nonprofits who are less likely to use contraception given their own religious beliefs. Instead, its conclusions mostly concern the “poor and low‐income women most at risk for unintended pregnancy.” Id. at 109. The study’s hope is that the elimination of cost sharing for contraception will induce the poor to use more effective, long‐acting methods, such as IUDs, implants, and sterilization. Id. at 108‐109. However, “the government must establish a compelling and specific justification for burdening these claimants.” Korte, 735 F.3d at 685; see also Hobby Lobby, 134 S. Ct. at 2761. The IOM Study fails to prove any connection whatsoever with the nonprofits’ employees. In fact, there are already a high level of access to contraception, a higher rate of use, and an increased use of more effective methods among the women with more income and education.28 Simply put, the IOM study fails to “specifically identify an ‘actual problem’ in need of solving,” and, consequently, the government has

28 Avaré, supra note 22, at 426. 66 Nos. 14‐1430 & 14‐1431

failed to demonstrate a compelling interest. Brown, 131 S. Ct. at 2738. B. The accommodation is underinclusive. The HHS accommodation’s underinclusiveness is another sign that the governmental interest is not compelling. Id. at 2740. The government “leaves appreciable damage to that supposedly vital interest unprohibited” by allowing reli‐ gious employers, grandfathered plans, and employers with fewer than 50 employees to avoid providing contraceptive coverage. Lukumi, 508 U.S. at 547 (internal quotation marks omitted). Although more health plans will lose their grand‐ fathered status the longer the ACA is in place, the number of persons employed by religious employers and organizations with fewer than 50 employees will remain considerable in light of the less than 2,000 covered employees concerned here. The accommodation is also underinclusive because it does not account for the other causes of the negative health outcomes the IOM Study correlates with unintended preg‐ nancies. According to the study, “women with unintended pregnancies are more likely than those with intended preg‐ nancies to receive later or no prenatal care, to smoke and consume alcohol during pregnancy, to be depressed during pregnancy, and to experience domestic violence during pregnancy.” IOM Study, 103. The study implies that unin‐ tended pregnancies cause these conditions, but there could just as well be another cause that causes not only these con‐ ditions, but the unintended pregnancy as well: poverty, lack of education, abuse, or other causes of risk taking behaviors. Nos. 14‐1430 & 14‐1431 67

The HHS accommodation addresses none of these alterna‐ tive causes, focusing solely on unintended pregnancies. Most notably, the study does not acknowledge the fact that preg‐ nancies resulting from failed contraceptives are also consid‐ ered unintended. Most damaging to the government’s asserted interest in the contraceptive mandate is the fact that those women most at risk for an unintended pregnancy are “women who are aged 18 to 24 years and unmarried, who have a low income, who are not high school graduates, and who are members of a racial or ethnic minority group.” IOM Study, 102 (citation omitted). These women—let alone the nonprofits’ employ‐ ees—are less likely to be served by the HHS accommodation, or the ACA’s contraception mandate generally, because they are less likely to have the type of employment that qualifies them for the health insurance under the ACA. These women would not obtain contraceptive services through the HHS accommodation, but through a number of government pro‐ grams such as Medicaid, 42 U.S.C. § 1396 et seq. (2010), and the Title X Family Planning Program, 42 U.S.C. § 300 (2006). “The consequence is that [the HHS accommodation] is wild‐ ly underinclusive when judged against its asserted justifica‐ tion, which … is alone enough to defeat it.” Brown, 131 S. Ct. at 2740. C. Forcing nonprofits to use the accommodation can on‐ ly provide a marginal increase in contraception. Contraceptive services are already widely available and their use is virtually universal. The HHS accommodation on‐ ly fills the “remaining modest gap” by making already prev‐ 68 Nos. 14‐1430 & 14‐1431

alent contraceptive services free for employees of religious nonprofits. Id. at 2741. This “can hardly be a compelling state interest.” Id. Further, the “more focused inquiry” of RFRA requires the government to demonstrate that it has a compel‐ ling interest in filling the gap made by the less than 2,000 employees of the nonprofits here. Hobby Lobby, 131 S. Ct. at 2779. This is even less of a compelling interest. Further still, the accommodation fills in even less of the gap when viewed from the perspective of unintended pregnancies. This is be‐ cause the accommodation seeks to treat unintended preg‐ nancies through contraceptive services, but contraceptives are not always effective for a variety of reasons. Even if this gap could be decreased by improving the effectiveness of contraceptives, “the government does not have a compelling interest in each marginal percentage point by which its goals are advanced.” Brown, 131 S. Ct. at 2741, n.9. D. A primary concern underlying the accommodation is cost. Cost appears to be a primary concern underlying the HHS accommodation. After all, babies are expensive. Of the IOM Study’s eight‐page discussion of contraceptives, a sig‐ nificant portion is spent on the cost savings to be expected from their use despite the study’s acknowledgement that cost considerations are out of scope: Although it is beyond the scope of the committee’s consideration, it should be noted that contraception is highly cost‐ effective. The direct medical cost of unin‐ tended pregnancy in the United States was Nos. 14‐1430 & 14‐1431 69

estimated to be nearly $5 billion in 2002, with the cost savings due to contraceptive use esti‐ mated to be $19.3 billion. … It is thought that greater use of long‐acting, reversible contra‐ ceptive methods—including intrauterine de‐ vices and contraceptive implants that require less action by the woman and therefore have lower use failure rates—might help further re‐ duce unintended pregnancy rates. Cost barri‐ ers to use of the most effective contraceptive methods are important because long‐acting, reversible contraceptive methods and steriliza‐ tion have high up‐front costs. IOM Study, 107–08 (citations omitted). The study’s primary conclusion is that the use of contraceptive services— particularly longer‐acting methods like IUDs—will greatly increase if they are free, “especially among poor and low‐income women.” Id. at 109 (emphasis added). The appearance is that the government desires to use contraceptives that “require less action by the woman” to prevent poor, unmarried, mi‐ nority women from having babies, as if babies were a costly disease. IOM Study, 108. Of course, this appearance is less‐ ened by the fact that the government is vigorously enforcing the HHS contraception mandate on even religious nonprofits through the accommodation. Because the government has failed to prove that the HHS accommodation furthers a compelling governmental inter‐ est, it is not allowed to burden the nonprofits’ religious exer‐ cise with the accommodation. 42 U.S.C. § 2000bb‐1. Thus, the 70 Nos. 14‐1430 & 14‐1431

government must grant the nonprofits the same exemption that it grants to religious employers. 45 C.F.R. § 147.131(a). IV. The accommodation is not the least restrictive means. Even if the government had proved that the HHS ac‐ commodation was in furtherance of a compelling interest, it would still have to grant the nonprofits’ an exemption from the accommodation because the accommodation is not the least restrictive means. 42 U.S.C. § 2000bb‐1(b)(2). The Su‐ preme Court in Hobby Lobby spoke of an obvious means that would be less restrictive than the HHS accommodation: The most straightforward way of doing this would be for the Government to assume the cost of providing the [objectionable] contracep‐ tives at issue to any women who are unable to obtain them under their health‐insurance poli‐ cies due to their employers’ religious objec‐ tions. This would certainly be less restrictive of the plaintiffs’ religious liberty, and HHS has not shown … that this is not a viable alterna‐ tive. 131 S. Ct. at 2780. The government argues that RFRA does not require the government to create entirely new programs to accommodate religious objections, but the government provides no authority for its position. The Court did not hold that it was so in Hobby Lobby. Id. at 2786. Rather, the Court stated that Congress understood that by passing RFRA it might cost the government extra to avoid burdening religion. Id. at 2781. Besides, the government already main‐ tains programs, such as Medicaid and the Title X Family Nos. 14‐1430 & 14‐1431 71

Planning Program mentioned earlier, which could be opened up to the employees of the nonprofits. The government also argues that a government‐run pro‐ gram is not a valid means because it would create additional burdens for the nonprofits’ employees and RFRA does not protect religious exercise that “unduly restrict[s] other per‐ sons, such as employees, in protecting their own interests, interests the law deems compelling.” Id. at 2786–87 (Kenne‐ dy, J., concurring). This requirement is not found in RFRA. What the government fails to acknowledge is that the pur‐ pose of an inquiry into the burdens on others is to determine whether a particular religious accommodation violates the Establishment Clause. See Cutter v. Wilkinson, 544 U.S. 709, 719–20 (2005) (Ginsburg, J.). To determine whether a reli‐ gious accommodation under RFRA is compatible with the Establishment Clause “courts must take adequate account of the burdens a requested accommodation may impose on nonbeneficiaries, and they must be satisfied that the Act’s prescriptions are and will be administered neutrally among different faiths.” Id. at 712 (citation omitted). A religious ac‐ commodation’s effect on third parties must be examined be‐ cause “[a]t some point, accommodation may devolve into ‘an unlawful fostering of religion.’” Id. at 714 (quoting Corpo‐ ration of Presiding Bishop of Church of Jesus Christ of Latter‐day Saints v. Amos, 483 U.S. 327, 334‐335 (1987)). The Supreme Court “has long recognized that the government may … ac‐ commodate religious practices … without violating the Es‐ tablishment Clause.” Id. at 713 (quoting Hobbie v. Unemploy‐ ment Appeals Comm’n of Fla., 480 U.S. 136, 144‐145 (1987)). 72 Nos. 14‐1430 & 14‐1431

Administering a government‐run program for contracep‐ tive coverage in order to relieve the nonprofits of the burden on their religion imposed by the accommodation would not “devolve into ‘an unlawful fostering of religion.’” Id. at 714. A government‐run program would provide the contracep‐ tion coverage on a cost‐free basis. Any burden resulting from an employee’s participation in the program would be de min‐ imis because it truly would be nothing more than additional paperwork (unlike the self‐certification and alternative no‐ tice). Furthermore, such a small burden would be no differ‐ ent than the burden experienced by the many who obtain dental and vision care benefits from different plans and fill their prescriptions at pharmacies unassociated with their health care providers. It would be absurd to say that such a de minimis burden even came close to the establishment of religion. Finally, any burden would be within the employee’s power to avoid by changing employment to an employer that provides the coverage. According to the government, when the contraceptive services mandate was enacted, 85% percent of large employers and 62% percent of small em‐ ployers already covered contraceptives services under the health plans. Even more plans will cover contraceptives and that coverage will be copayment‐free now that the mandate is in force. V. Conclusion This dissent explores the road less traveled by. As de‐ tailed above, the detour exposes two serious misrepresenta‐ tions. First, the so‐called accommodation is nothing but a mirage. The government strung together the complicated Nos. 14‐1430 & 14‐1431 73

details to create a lengthy and twisted extension cord. The end result is the de facto imposition of a provision offering “free” birth control into the nonprofits’ necessary health plans. The unwanted provision is very offensive and contra‐ ry to the nonprofits’ sincerely held religious beliefs. The im‐ position does not occur if the nonprofits refuse to plug in the extension cord by refusing to self‐certify or otherwise indi‐ cate consent through the alternative notice. But this refusal causes enormous, existential monetary penalties. So, there are substantial burdens at both ends of the accommodation. Second, deep into the detour is the falsehood behind the government’s claim that increasing the availability of contra‐ ceptive services furthers a “compelling governmental inter‐ est.” That label is needed to overcome the nonprofits’ sin‐ cerely held religious beliefs that no one disputes. But, con‐ traceptive services are already widely available from the great majority of employers. And, for the primarily targeted poor and/or unemployed women, whom the mandate does not affect, there are already programs like Medicaid and Ti‐ tle X that offer free contraceptive services. At its center, the IOM Study recognizes that babies are medically very expen‐ sive, so the government endeavors to reduce “unexpected” pregnancies to save money. In effect, the government con‐ siders pregnancy a preventable disease. Aside from the fact that the government desires to sub‐ stantially burden the nonprofits’ religious exercise in fur‐ therance of an exaggerated, misnamed, and misdirected in‐ terest, there are, no doubt, less restrictive means of further‐ ing its interest. But why even go there? The government cer‐ 74 Nos. 14‐1430 & 14‐1431

tainly has no compelling interest in forcing contraceptive coverage into the nonprofits’ otherwise wanted and needed health plans when they unanimously assert they don’t want the coverage and don’t need it. The obvious solution for these plaintiffs (and likely for the plaintiffs involved in the similar—and similarly expensive—litigation in at least six other federal circuits, see supra p.42) is for the government to extend the religious employer exemption to all religious nonprofits that object to the coverage. 45 C.F.R. § 147.131(a). The nonprofits have shown a likelihood of success on their claims that the HHS accommodation violates RFRA. 42 U.S.C. § 2000bb‐1. The preliminary injunction granted by the district court should be affirmed. Korte, 735 F.3d at 666 (“Alt‐ hough the claim is statutory, RFRA protects First Amend‐ ment free‐exercise rights, and in First Amendment cases, the likelihood of success on the merits will often be the determi‐ native factor.” (internal quotation marks omitted)). For all these reasons, I dissent.