4B. RFRA Strict Scrutiny: The Argument from Secular Exceptions

(For an introduction to this series of posts, see here.)

On then to another general argument — the argument that, regardless of the specific interests the government asserts in support of the employer mandate, those interests don’t count because the law has secular exceptions. One version of the argument is that such exceptions make the law “underinclusive,” and thus prove that the government interest can’t be compelling. A related version is that, under RFRA, religious exemption claims are entitled to be treated as well as the most favored other claims for exemption. Here is how the Tenth Circuit en banc majority opinion in Hobby Lobby put it:

The government asserts two interests here: “the interests in [1] public health and [2] gender equality.” We recognize the importance of these interests. But they nonetheless in this context do not satisfy the Supreme Court’s compelling interest standards….

[T]he interest here cannot be compelling because the contraceptive-coverage requirement presently does not apply to tens of millions of people. As noted above, this exempted population includes those working for private employers with grandfathered plans, for employers with fewer than fifty employees [who don’t have to provide any health insurance -EV], and, under a proposed rule, for colleges and universities run by religious institutions. As the Supreme Court has said, “a law cannot be regarded as protecting an interest of the highest order when it leaves appreciable damage to that supposedly vital interest unprohibited.” Lukumi; see also O Centro (citing Lukumi as instructive in determining whether exemptions undermine a compelling government interest for purposes of RFRA). The exemptions at issue here would yield precisely this result: they would leave unprotected all women who work for exempted business entities.

Now it’s not clear to me that all the exceptions “leave[] appreciable damage to [both] supposedly vital interest[s] unprohibited.” As I understand it, the exception for employers with fewer than fifty employees wouldn’t single out contraceptives (and it is this singling out that is said to implicate the “gender equality” interest); those employers just don’t have to provide health insurance for their employees at all, but if they do, they too have to cover contraceptives. And as I understand the college and university exception, it assures employees coverage for the contraceptives through the insurance companies, though with no direct payment for it by the employers; that exception thus wouldn’t implicate either interest. Still, I agree that one large exception — for grandfathered plans — at least temporarily “leaves appreciable damage to [both] supposedly vital interest[s] unprohibited,” and the other one, for smaller employers, does the same as to the public health interest.

But let’s set all this aside, because the problem with the “cannot be regarded as protecting an interest of the highest order” argument is deeper than that.

1. Nearly all important laws, have a large set of exceptions. Those laws, even when they serve compelling interests, leave appreciable damage to the interests unprohibited. Yet this can’t mean that religious exemptions must be granted from such laws.

Let’s begin with tax laws. Some people have religious objections to paying particular kinds of taxes, or taxes that would go to particular activities. They don’t get exemptions, though, because of the “broad [and compelling] public interest in maintaining a sound tax system” (Hernandez v. Commissioner (1989), relying on United States v. Lee (1982)).

Yet of course the tax system is chock full of a vast range of exceptions — many more than the employer mandate has. That doesn’t entitle objectors to a religious exemption from tax obligations, or make the interest in tax collection less than compelling. “The fact that Congress has already crafted some deductions and exemptions in the Code … is of no consequence, for the guiding principle is that a tax ‘must be uniformly applicable to all, except as Congress provides explicitly otherwise.’” Hernandez (emphasis in original).

Or consider draft law. Gillette v. United States (1971) rejected the claim that people who had a religious objection to fighting in what they saw as unjust wars were constitutionally entitled to a draft exemption. (The statutory conscientious objector exemption applies only to people who object to fighting in all wars.) Gillette has since been cited (for instance, by Bob Jones Univ. v. U.S. (1983)) as an example of the application of Sherbert/Yoder strict scrutiny, based on the compelling interest of raising manpower for the military. That the draft likewise had many secular exceptions that covered many people didn’t lead the Court to view the interest as any less compelling.

Likewise for a vast range of other laws that the Court hadn’t considered during the Sherbert/Yoder era, but that I think should clearly withstand RFRA exemption claims. Federal employment discrimination laws have various exemptions, such as for small employers (which implicates millions of employees), for certain bona fide occupational qualifications, and more. The duty to testify when subpoenaed is subject to many exceptions in the form of testimonial privileges, privileges that people routinely use.

Statutory rape laws often except acts committed by someone who is close enough in age to the minor (which likely covers millions of acts each year), or acts committed by the minor’s spouse. Breach of contract law has exceptions galore. The Copyright Act contains one operative section followed by more than a dozen sections of exceptions, many of which are taken advantage of constantly. The list could go on.

All these laws do, I think, serve government interests that the legal system would view as compelling. But they contain exceptions because most laws serve multiple interests. Accurate factfinding at trial, for instance, is very important, and people shouldn’t get an exemption just because they feel a religious obligation to, for instance, not testify in a way that may harm a relative, a coreligionist, or someone who the witness thinks is innocent. And this is so despite the present of various secular exceptions to that duty. Those exceptions don’t undermine the significance of the government interest — they merely reflect the fact that the legal system also tries to serve other interests, such as (for instance) also encouraging candid disclosures to spouses, psychiatrists, and lawyers.

Interpreting RFRA as embodying the view that, whenever a law “leaves appreciable damage to [an] interest unprohibited,” the interest can’t be compelling, would require granting religious exemptions to all these laws. Yet I don’t think that’s right. I don’t think the Court is likely to do this. And I don’t think it would be consistent with Lee and Gillette, two of the precedents from the Sherbert/Yoder era that Congress was trying to restore.

2. Now I agree that the Court has indeed held that the underinclusiveness of a law may make it fail strict scrutiny. This, though, has happened with the nearly-fatal-in-fact strict scrutiny used in cases involving discrimination based on the content of speech or the religiosity of conduct. The Court expects such discrimination to be extremely rare, and any content-based or religion-based discrimination is thus extremely hard to justify, and raises grave suspicions about the government’s motivation.

But secular exceptions from generally applicable laws are extremely common, and are generally seen by the Court as legitimate and well-intentioned. That’s why, even when generally applicable laws were subject to Free Exercise Clause scrutiny, the presence of such secular exceptions wasn’t seen as casting doubt on the strength of the government’s interests.

The Hobby Lobby Tenth Circuit decision called on Gonzales v. O Centro (2006) for support:

On this question, O Centro is particularly instructive. In that case, a religious group sought an exemption for the sacramental use of hoasca, a hallucinogen classified as a Schedule I(c) controlled substance under the Controlled Substances Act. The question in O Centro was limited to whether the government could show a compelling governmental interest under RFRA to justify what was indisputably a substantial burden on the plaintiffs’ exercise of religion. The government in part relied on its interest in promoting public health and safety and upon Congress’s determination that hoasca “‘has a high potential for abuse,’ ‘has no currently accepted medical use,’ and has ‘a lack of accepted safety for use … under medical supervision.’”

The Supreme Court refused to credit this argument, however, in part because the CSA and related regulations contained an exemption for the religious use of another substance categorized as a Schedule I hallucinogen, peyote. As the Court reasoned, “Everything the Government says about the [dangerous chemicals] in hoasca … applies in equal measure to the [dangerous chemicals] in peyote.” Because both the Executive Branch and Congress had decreed a religious exemption for Native American use of peyote, the Court concluded that “it [was] difficult to see how” those same concerns could “preclude any consideration of a similar exception for” the religious use of hoasca. If the peyote exemption in O Centro, which applied to “hundreds of thousands of Native Americans” was enough to undermine the government’s compelling interest argument in that case, we conclude the exemption for the millions of individuals here must dictate a similar result.

But this, I think, misreads O Centro. In O Centro, the government argued that denying a religious exemption to the ban on the hallucinogen contained in hoasca was necessary to serve compelling interests in “protecting the health and safety of UDV members” and “preventing the diversion of hoasca from the church to recreational users.” The lower court held that the government didn’t prove the factual predicate for these claims; the court “concluded that the evidence on health risks was ‘in equipoise,’ and similarly that the evidence on diversion was ‘virtually balanced.’” The Court relied on these factual findings as to those interests.

The government, though, sought to avoid these factual findings by arguing that “there is no need to assess the particulars of the UDV’s use [of hoasca] or weigh the impact of an exemption for that specific use, because the Controlled Substances Act serves a compelling purpose and simply admits of no exceptions” (emphasis added). It is this claim that uniformity was absolutely necessary that led to the Court’s discussion of peyote:

Congress’ determination that DMT should be listed under Schedule I simply does not provide a categorical answer that relieves the Government of the obligation to shoulder its burden under RFRA…. The fact that the Act itself contemplates that exempting certain people from its requirements would be “consistent with the public health and safety” indicates that congressional findings with respect to Schedule I substances should not carry the determinative weight, for RFRA purposes, that the Government would ascribe to them.

And in fact an exception has been made to the Schedule I ban for religious use [as to peyote]…. Everything the Government says about the DMT in hoasca — that, as a Schedule I substance, Congress has determined that it “has a high potential for abuse,” “has no currently accepted medical use,” and has “a lack of accepted safety for use … under medical supervision” — applies in equal measure to the mescaline in peyote, yet both the Executive and Congress itself have decreed an exception from the Controlled Substances Act for Native American religious use of peyote. If such use is permitted in the face of the congressional findings in § 812(b)(1) for hundreds of thousands of Native Americans practicing their faith, it is difficult to see how those same findings alone can preclude any consideration of a similar exception for the 130 or so American members of the UDV who want to practice theirs. See Church of Lukumi Babalu Aye, Inc. v. Hialeah, 508 U.S. 520, 547 (1993) (“It is established in our strict scrutiny jurisprudence that ‘a law cannot be regarded as protecting an interest ‘of the highest order’ … when it leaves appreciable damage to that supposedly vital interest unprohibited’”)….

The Government argues that the existence of a congressional exemption for peyote does not indicate that the Controlled Substances Act is amenable to judicially crafted exceptions. RFRA, however, plainly contemplates that courts would recognize exceptions — that is how the law works. Congress’ role in the peyote exemption — and the Executive’s, see 21 CFR § 1307.31 (2005) — confirms that the findings in the Controlled Substances Act do not preclude exceptions altogether; RFRA makes clear that it is the obligation of the courts to consider whether exceptions are required under the test set forth by Congress.

So the Court was rejecting the government’s argument that the whole strict scrutiny analysis, with its factual inquiries about harm and less restrictive alternatives, could be avoided simply on the grounds that drug laws must be uniform. Drug laws already aren’t uniform, the Court said; Congress concluded that uniformity isn’t necessary. The government needs to actually prove why denying this exemption request is the least restrictive means of serving a compelling government interest. It can’t just rely on the compelling government interest behind the Controlled Substances Act as a whole, coupled with a claim that this interest applies to all possible applications of the Act.

Indeed, it’s noteworthy how the Court deals with the rejection of claimed exemptions from tax laws, and the argument that uniformity is necessary there:

The Government points to some pre-Smith cases relying on a need for uniformity in rejecting claims for religious exemptions under the Free Exercise Clause, but those cases strike us as quite different from the present one. Those cases did not embrace the notion that a general interest in uniformity justified a substantial burden on religious exercise; they instead scrutinized the asserted need and explained why the denied exemptions could not be accommodated. In United States v. Lee, 455 U.S. 252 (1982), for example, the Court rejected a claimed exception to the obligation to pay Social Security taxes, noting that “mandatory participation is indispensable to the fiscal vitality of the social security system” and that the “tax system could not function if denominations were allowed to challenge the tax system because tax payments were spent in a manner that violates their religious belief.” See also Hernandez v. Commissioner, 490 U.S. 680, 700 (1989) (same)…. These cases show that the Government can demonstrate a compelling interest in uniform application of a particular program by offering evidence that granting the requested religious accommodations would seriously compromise its ability to administer the program.

Here the Government’s argument for uniformity is different; it rests not so much on the particular statutory program at issue as on slippery-slope concerns that could be invoked in response to any RFRA claim for an exception to a generally applicable law.

So read in context, the O Centro decision doesn’t generally hold that a law can’t serve a compelling interest if it has broad secular exceptions (despite the quote from Lukumi). Rather, such exceptions simply help show that the government can’t categorically rely on the need for uniformity. Courts are still “obligat[ed] … to consider whether exceptions are required under the test set forth by Congress.” Indeed, the government can even rely on the uniformity interest, in the face of secular exceptions, if it “offer[s] evidence that granting the requested religious accommodations would seriously compromise its ability to administer the program.”

O Centro is thus consistent with Lee and Hernandez, which it cites and treats as authoritative for RFRA purposes. A law may serve a compelling interest, so that exemption requests from it may be denied, even if it has many secular exceptions.

Yes, the employer mandate excludes employers with grandfathered plans. But this doesn’t itself give other employers a RFRA right to reject the mandate, just as the fact that tax law has many exceptions doesn’t give nonexcepted taxpayers a RFRA right to not pay certain taxes.

Yes, the employer mandate excludes small employers. But this doesn’t itself give large religious employers a RFRA right to reject the mandate, just as the fact that Title VII excludes small employers doesn’t itself give large religious employers a RFRA to reject the ban on race discrimination. (Church and religious schools employers do have a right under Hosanna-Tabor Evangelical Lutheran Church & School (2012) to discriminate in choice of clergy and other religion-propagating employees, but that’s a First Amendment for a narrow set of employers and a narrow set of employees, not a general RFRA right applicable to all employers who religiously object to antidiscrimination law.)

All this doesn’t mean, of course, that Hobby Lobby should necessarily lose. It just means that the Tenth Circuit’s rationale, based on the secular exceptions from the law, doesn’t itself make Hobby Lobby win. The government may still try to demonstrate that denying the religious exemption request is the least restrictive means of serving a compelling government interest — albeit a compelling government interest that the employer mandate serves alongside other interests (e.g., in preventing excessive disruption, or excessive burden on small employers). More on that argument tomorrow.