Hobby Lobby, RFRA, and a “Private” Establishment Clause

January 21st, 2014

Professor Gedicks graciously offered three probing responses to my post on his WaPo Op-Ed arguing that (in short) the government’s decision to sanction Hobby Lobby’s decision to not pay for contraceptives unconstitutionally imposes its religious beliefs on its employees. In this post, I will respond to some of Gedick’s points, and a raise a few more questions his thoughtful comments raised in my mind.

Before I do, I neglected previously to link to an article Gedicks and Rebecca Van Tassell published in the Harvard Civil Rights-Civil Liberties Law Review, titled “RFRA Exemptions from the Contraception Mandate: An Unconstitutional Accommodation of Religion.” Here is the abstract of that article:

Litigation surrounding use of the Religious Freedom Restoration Act to exempt employers from the Affordable Care Act’s “contraception mandate” is moving steadily towards resolution in the U.S. Supreme Court. Both opponents and supporters of the mandate, however, have overlooked the Establishment Clause limits on such exemptions.

The heated religious-liberty rhetoric aimed at the mandate has obscured that RFRA is a “permissive” rather than “mandatory” accommodation of religion — a government concession to religious belief and practice that is not required by the Free Exercise Clause. Permissive accommodations must satisfy Establishment Clause constraints, notably the requirement that the accommodation not impose material burdens on third parties who do not believe or participate in the accommodated practice.

While it is likely that RFRA facially complies with the Establishment Clause, it violates the Clause’s limits on permissive accommodation as applied to the mandate. RFRA exemptions from the mandate would deny the employees of an exempted employer their ACA entitlement to contraceptives without cost-sharing, forcing employees to purchase with their own money contraceptives and related services that would otherwise be available to them at no cost beyond their healthcare insurance premium.

Neither courts nor commentators seem aware that a line of permissive accommodation prohibits shifting of material costs of accommodating anti-contraception beliefs from the employers who hold them to employees who do not. One federal appellate court has already mistakenly dismissed this cost-shifting as irrelevant to the permissibility of RFRA exemptions from the mandate.

The impermissibility of cost-shifting under the Establishment Clause is a threshold doctrine whose application is logically prior to all of the RFRA issues on which the courts are now focused: If RFRA exemptions from the mandate violate the Establishment Clause, then that is the end of such RFRA exemptions, regardless of whether for-profit corporations are persons exercising religion, the mandate is a substantial burden on employers’ anti- contraception beliefs, or the mandate is not the least restrictive means of protecting a compelling government interest.

Part I summarizes the legal mechanics of the mandate and briefly describes the three classes of anti-mandate plaintiffs — churches, religious nonprofit organizations, and for-profit businesses owned by anti-contraception believers. Part II details Establishment Clause doctrine that prohibits permissive accommodations that impose material burdens on third parties. Part III applies this rule to RFRA exemptions from the mandate, showing that the cost- shifting entailed by such exemptions violates the Establishment Clause. We conclude that the existing regulatory regime that exempts churches, accommodates religious nonprofits, and leaves for-profit businesses subject to the mandate is the proper balance of private and government interests in the radically plural society that the United States has become.

I will repeat here the emphasized sentence, which seems to be the most important turn in the article:

Neither courts nor commentators seem aware that a line of permissive accommodation prohibits shifting of material costs of accommodating anti-contraception beliefs from the employers who hold them to employees who do not.

I’ll admit, before seeing this article, I was among those who were totally unaware that an argument existed that because of the Establishment Clause, the government cannot create an accommodation to allow an employers to shift costs onto their employees (through denying certain benefits) due to the employers religious beliefs.

The authors write:

When a legislature directly grants a specific “retail” permissive accommodation to a named class of religious adherents, it must comply with the Establishment Clause limit on negative religious externalities.99 It follows that when Congress indirectly grants permissive accommodations “wholesale” through a general statute like RFRA it must also work within Establishment Clause limitations.100

They argues that creating exemptions to RFRA would violate the establishment clause. In other words, the government providing an accommodation from RFRA to Hobby Lobby, because of Hobby Lobby’s religious beliefs (not required by Free Exercise), would in effect violate the Establishment Clause, because the government is sanctioning (in essence) an unconstitutional externality passed onto employees due to the employers religion. (I think I am stating this correctly).

They conclude in the article:

In short, the Court and academic commentators are united in disapproving permissive accommodations that generate negative religious externalities—that is, in condemning accommodations that shift significant financial and other costs of a religious practice from those who engage in it to those who do not.

All of the Establishment Clause precedents the article cite concern violations of the Constitution by government itself through passing and enforcing laws. The supposed violation of the Constitution at issue here is that a private employer is shifting costs onto employees due to the employers religious beliefs. The article connects the private conduct at issue in Hobby Lobby to the Establishment Clause (which only applies to the government) by positing that the exemption (sanctioning if you will) of this cost shifting is unconstitutional.

The reason why these exemptions would violate RFRA is, not because of anything the government is doing directly–such as passing a law telling employers they cannot force employees to work on the Sabbath. Instead, the unconstitutional conduct is based on what a private employer is doing–specifically shifting costs onto employees. For the government to grant this exemption–or more likely, for the Supreme Court to order the government to grant this exemption, would be unconstitutional, Gedicks argues.

First, as a prerequisite to considering whether the Establishment Clause places limits on the ability of private employers to shift costs to their employees as a result of the employers’ religious beliefs, I want to push a bit at whether there is state action. The article makes a fascinating case for what I would perhaps call a “private” establishment clause. This is the notion that, because private employers are subject to RFRA (a federal law), and seek to vindicate their arguments in federal court, they have effectively become state actors if the government grants them an exemption. As Professor Gedicks wrote in a comment, “a RFRA exemption for Hobby Lobby would indeed involve federal government action, and not just private action.” In other words, the conduct of private employers, exempted by the government, raises an establishment clause violation

This is somewhat reminiscent of the argument made in Shelly v. Kraemer: racial covenants could be written,  but if home owners sought enforcement of their contractual rights in court, they were employing the long arm of the law to implement them, and thus there was state action, bringing the action under the auspices of the Fourteenth Amendment. That argument–perhaps an exigency of the nascent Civil Rights movement–was never held to any scrutiny, largely because the Fair Housing Act now handles most of that type of litigation. But I always ask my students, how is it possible that seeking to vindicate a right in court transform you into a state actor. If that is the standard, who *isn’t* a state actor.

To borrow an unfortunate line from earlier ACS litigation, what is the limiting principle? If RFRA seeks to protect certain rights of free exercise, is seeking an exemption from a law that infringes those beliefs enough to transform the employers actions into state action, now bound by the Establishment Clause? If so, that is a very dramatic concept. This would, in effect, bootstrap an establishment clause challenge to a suit for a RFRA exemption. You can imagine that often seeking to exercise your religion may have an impact on someone else’s wallet. I need to wrestle further with the idea that granting exemptions violates the establishment clause because of the actions private employers take, but I’ll assume this is the case for now.

I’ll respond briefly to Professor Gedick’s other points concerning Title VII. Has there ever been case in where any court found that a religious employer denying an employee contraceptives violated Title VII? I understand that employers  have to reasonably accommodate the religious beliefs of employees. But do employees have to reasonably accommodate the religious beliefs of their employers?

Of course, if a religious employer creates a hostile work environment due to excessive preaching, or discriminates based on religion by only requiring employees of a certain faith, that is prohibited. But here we have it in reverse. The traditional case considers whether an employer can force an employee to work on Sunday. The authors write about the Court’s invalidation of a state law “granted employees an absolute right not to work on their chosen Sabbath, irrespective of the costs their choices might impose on their employer and co-workers.”

The statute invalidated in Thornton generated what economists call a “negative externality”—“a cost that one person, firm, or group imposes on others without their consent.”54 By giving employees an unqualified right not to work on their chosen Sabbath, the statute “externalized” the cost of accommodating Sabbath observance from the Sabbath-observing employees to employers and other employees who didn’t observe a Sabbath. This, the Court held, violated the Establishment Clause.

The SG makes a similar point in its brief (p. 39):

The court of appeals believed that the interests of corporate-respondents’ employees are entitled to no weight under RFRA, opining that “[a]ccommodations for religion frequently operate by lifting a burden from the accommodated party and placing it else- where.” Pet. App. 60a-61a. But this Court has never permitted a secular employer to obtain a religious accommodation that comes at the expense of employ- ees.9

9 Indeed, the Court has held that, under certain circumstances, an accommodation that imposes burdens on employees can violate the Establishment Clause. Compare Estate of Thornton v. Caldor, Inc., 472 U.S. 703, 708-711 (1985) (holding that statute requiring employers to accommodate an employee’s Sabbath observance without regard to the burden such an accommodation would im- pose on the employer or other employees violated the Establish- ment Clause), with Amos, 483 U.S. at 334-340 (concluding that Title VII’s exemption for religious employers from its prohibition on religious discrimination does not violate the Establishment Clause as applied to non-profit activities of a church).

But let’s consider the mirror image of that case. Instead of an employer requiring his employees to work on Sunday, the employer shuts down on Sunday due to a newfound religious conviction. As a result, an employee who previously worked on Saturday and Sundays, and could only work on weekends, has now had his hours cut in half. (Of course the employee could choose to work elsewhere where more hours are available, or work during the week, but under this frame, that is not material). This is, in Gedicks words, a material cost shifted to the employee as a result of the employers religious beliefs. In fact, having eight hours of salaries cut from a salary is *much* more significant than being denied the cost of contraceptives for the overwhelming majority of employees. Would this result in a violation of Title VII? This is the difference between an employer having to accommodate a religious employes religious beliefs, and an employee having to accommodate an employers religious beliefs. The argument advanced in the article would thus impose something a “private” establishment clause.

Granted, keeping your store open on Sunday is not required by any law (though some laws require their closures). But before the ACA, providing contraceptives was also not required. Let’s say a state passed a law requiring that stores that service a certain number of customers a day must be open seven days a week, in order to promote the general welfare under the police power. A narrow religious exemption is made for churches, but not for religious-owned businesses. If the government handed out an exemption to, let’s say a Kosher butcher shop that closes on Saturday, would this violate the Establishment Clause? Certainly employees would gain additional salaries as a result of this new law, by being eligible to work on a Saturday (though there may not be many customers). Wouldn’t this law create a new “baseline”?

Which leads me to the final point Professor Gedicks made, which I think is the meta issue lurking in aisles of Hobby Lobby.

I wrote:

Hobby Lobby is not some mere private actor, but is in essence a cog in the federal machinery that guarantees healthcare to anyone. Employees should no longer treat health insurance as some sort of benefit, but now must demand it–and the employers’ religious liberty should not stand in that way. … The significance of Hobby Lobby, I have come to realize, goes far beyond RFRA or birth control pills. Instead, the ruling would either legitimize or challenge the consolidation of the relationship between the individual, the employer, and the state, under the auspices of the ACA, in a way we have not yet seen.

Professor Gedicks replies that the ACA does no more than “the Social Security Act (funded 50% by mandatory employer payroll taxes), the Fair Labor Standards Act (mandating payment of minimum wage), the Family and Medical Leave Act (mandating job protection for specified employee leaves), or any number other federal laws that dictate minimum employee compensation and benefits.” After I wrote the post yesterday, and before I saw the comments, I started thinking more about Social Security and other laws that regulate the employer-employee-government trinity.

On that point, Joey Fishkin has an important argument that the article alludes to.

Some Mandate opponents have argued that RFRA exemptions for anti-Mandate employers would not shift costs to employees, because employees of RFRA-exempted employers are no worse off than they would have been in the absence of the Mandate—employees in both situations must use their own funds to purchase contraceptives. This makes no sense. It is like defending a denial of Social Security benefits by observing that it merely puts the disappointed claimant in the same position he or she would have been in had the Social Security program never been enacted.107 The problem, of course, is that we do not live in a world in which Social Security does not exist; Social Security has been enacted, and its enactment created a social welfare entitlement whose denial to any particular claimant deprives him or her of its benefits.

We are now living in a new world. The ACA created a massive new entitlement with a new “baseline.” Health insurance is now a condition of employment. It was not before. Previous laws governing the terms of employment–such as FMLA, FLSA, etc.–go to the actual terms of the job: what are you doing, and what and how are you getting paid, etc. The ACA mandates a new benefit (or if you like Marty Lederman’s argument, imposes a tax on not giving benefits), totally apart from the terms of employment. Sure, if an employer offered a pension in the past, it was governed by ERISA. But there is no requirement to offer pensions.

The stakes here are significant. If this argument is taken to its logical conclusion, it could create a private establishment clause, whereby an employer’s ability to run his business will be constrained not only by regulations requiring him to do certain things that may violate his religious beliefs (short of a free exercise claim), but also he must operate his business so as to prevent his own religious beliefs from shifting externalities employers. This represents a significant responsibility on behalf of an employer to provide for his employees, even if it conflicts with his religious beliefs, in a way that has not been done before. Putting aside the link Gedicks makes between the employer and state action, here the new constitutional burden would fall directly on the religious employer.

Lots of interesting food for thought here. I appreciate Professor Gedicks engaging the topic.