ANALYSISApril 13, 2026
Should religious business owners have exemptions from anti-discrimination law?
The question of whether religious business owners may claim exemptions from anti-discrimination laws remains actively contested in U.S. courts and legislatures, with a series of major rulings and new legislation reshaping the legal landscape through 2025-2026. The Fifth Circuit's decision in Braidwood Management v. EEOC, left intact after the Supreme Court declined to hear it, allows religious private employers in Texas, Louisiana, and Mississippi to discriminate against LGBTQ+ workers under RFRA. In June 2025, a unanimous Supreme Court ruled in Catholic Charities Bureau v. Wisconsin that Wisconsin had improperly denied a religious exemption to Catholic Charities from the state unemployment insurance system.
When a business owner's sincere religious belief collides with a customer's legal right not to be turned away — whose freedom wins, and who gets to decide which faiths qualify for the exception?
Not familiar with this story? Get context →
Whether RFRA demands individualized adjudication
C
RFRA was passed unanimously by the Senate and signed by Bill Clinton in 1993 — this is not a right-wing invention. It explicitly requires government to demonstrate a compelling interest pursued through the least restrictive means before overriding religious conscience. The question is not whether religious claims always win; it is whether government must even try to justify itself, and the current enforcement trajectory says no.
L
You're right that RFRA demands justification, but you're glossing over what Braidwood actually did: it used that balancing test to carve a private employer exception to Bostock without Congress ever authorizing one. That's not government failing to justify itself — that's a court rewriting civil rights law through a procedural side door that RFRA's authors almost certainly never intended.
C
Calling it a 'side door' assumes RFRA's balancing test shouldn't apply to Title VII claims — but Congress wrote no such limitation into the statute. If the government has a compelling interest in LGBTQ+ employment protection, it can prove that in court; RFRA doesn't prevent it, it just makes the government show its work.
L
Making the government show its work sounds neutral until you notice that 'showing its work' in three states currently produces a result where gay workers can be fired — so the mechanism may be clean in theory while being corrosive in practice.
Substantial burden of employing gay workers
C
The Catholic Charities decision, written unanimously by Justice Sotomayor, found that Wisconsin's line-drawing between sufficiently religious and essentially secular work was textbook denominational discrimination. Government agencies have repeatedly treated religious exercise as an obstacle rather than a right, and courts including liberal justices have said so.
L
Catholic Charities was decided on narrow grounds about Wisconsin's specific denominational line-drawing — and it was probably correct on those facts. But it doesn't answer the prior question: how does paying a salary to a competent gay employee constitute a substantial burden on religious exercise in the first place? Not performing a ceremony, not endorsing a marriage — just employment.
C
That question — what counts as a substantial burden — is exactly what individualized adjudication is for. You're asking courts to rule categorically that employment can never burden religion, which is the same presumptive subordination RFRA was designed to prevent.
L
There's a difference between courts being open to the question and courts answering it the way Braidwood did — which is to treat the mere presence of a gay employee as a cognizable religious injury, a standard with no obvious limiting principle.
Geographic patchwork of worker protections
C
The Braidwood ruling applies only in Texas, Louisiana, and Mississippi — it is a regional circuit decision subject to ongoing litigation, not a settled national outcome. The liberal argument is strongest as a geographic and political warning, but predicting a systemic collapse from a limited ruling treats a possible slope as an already-realized one.
L
You're right that Braidwood is regional, but the patchwork problem exists independently of whether Braidwood spreads: twenty-eight states have no explicit SOGI protections, meaning workers there already depend on Title VII coverage — the very coverage now compromised in three of those states. A gay teacher in Houston and the same teacher in Austin have different legal rights, and that divergence is current reality, not a prediction.
C
The 22-state divergence you're describing predates Braidwood and reflects democratic choices by state legislatures — which is itself an argument for Congress to act rather than for courts to impose a uniform outcome by refusing to apply RFRA.
L
Telling LGBTQ+ workers in Houston to wait for Congress while their colleagues in Austin are protected today is not a neutral procedural position — it's a choice about whose rights bear the cost of legislative delay.
Tax exemptions as coercion versus accountability
C
When the state conditions federal contracts and tax exemptions on religious organizations abandoning their theological positions, it crosses from regulating conduct to punishing doctrine. The IRS's Bob Jones mechanism — use tax status to reshape religious belief — has migrated into contexts where the theological claim is far more defensible and the government's interest far less clear.
L
The Bob Jones problem cuts the other way: both racial exclusion and LGBTQ+ exclusion involve sincere theological claims producing documented harm to identifiable groups, and the mechanism used to defund one is identical to the mechanism you're now contesting for the other. You can't argue Bob Jones was justified and the current application is coercion without explaining why the theology is different enough to produce a different result.
C
Bob Jones turned on a specific judicial finding that racial discrimination had 'no legitimate justification' capable of overriding a fundamental national public policy — a threshold the Court has not applied to LGBTQ+ discrimination with equal clarity. That distinction is available, even if it's contingent rather than self-evident.
L
Contingent on evolving legal consensus means the line moves — and right now it's moving toward insulating religious institutions from public accountability while they remain eligible for public funding, which is a direction, not a balance.
Ministerial exception versus commercial employment
C
Title VII's existing framework already protects genuine religious practice — religious organizations have a ministerial exemption for core functions, and courts have upheld it. What Braidwood extends is not some novel right but the application of RFRA's balancing test to employment contexts, requiring courts to weigh competing interests rather than automatically subordinating religion.
L
The ministerial exception is precisely the line you're erasing. It protects a church's right to choose its own clergy — nobody disputes that. What Braidwood adds is categorically different: the right to exclude a class of workers from secular employment in the private market while remaining subject to every other provision of federal law. That's not religious autonomy; that's a license to discriminate against a protected class in commercial activity.
C
You're drawing the line at 'secular employment in the private market,' but that begs the question — a religiously affiliated school or hospital is not a secular employer, and RFRA's balancing test exists precisely to sort out which contexts warrant which level of protection.
L
If 'religiously affiliated' is enough to trigger exemption from employment discrimination law in any commercial context, then the category of secular employer shrinks every time a business adds a doctrinal statement — and the protected class absorbs all the costs of that expansion.
Conservative's hardest question
The strongest challenge to this argument is the Bob Jones precedent itself: if religious racial discrimination could be defunded through tax exemption revocation, it is genuinely difficult to draw a principled line preventing the same mechanism from applying to religious organizations that exclude LGBTQ+ people, since both involve sincere theological claims that produce real exclusionary harm to identifiable groups. The conservative position must explain why one application of government leverage was justified and the other is not — and the answer, while available, is not comfortable or clean.
Liberal's hardest question
The most serious challenge to this argument is that RFRA was passed by Congress with broad bipartisan support precisely to protect religious minorities from neutral laws with unintended coercive effects, and courts have historically applied it case-by-case rather than as a blanket exemption — which means the liberal alarm about Braidwood 'hollowing out' civil rights law may overstate how far courts will actually extend it beyond specific factual settings. If Braidwood remains a limited regional anomaly rather than a spreading precedent, the systemic threat I am describing is real but not yet realized, and invoking the slippery slope before the slope has fully materialized weakens the urgency of the argument.
Both sides agree: Both sides accept that RFRA's balancing test — requiring government to show a compelling interest pursued through the least restrictive means — is the correct legal framework for adjudicating these conflicts, and neither argues it should be abolished.
The real conflict: They disagree on a factual-legal question about what constitutes a 'substantial burden' on religious exercise: conservatives argue that being compelled to employ someone whose identity conflicts with religious teaching meets that threshold, while liberals argue that paying a salary for competent secular work cannot plausibly qualify.
What nobody has answered: If the government's compelling interest in preventing racial employment discrimination was sufficient to override sincere religious objections at Bob Jones University, what is the principled legal basis — not a political one — for treating the government's interest in preventing LGBTQ+ employment discrimination differently, given that both involve sincere theology producing documented exclusionary harm to identifiable groups?
Sources
- U.S. Court of Appeals for the Fifth Circuit, Braidwood Management, Inc. v. EEOC (2023)
- U.S. Supreme Court, Catholic Charities Bureau v. Wisconsin (June 2025)
- U.S. Supreme Court, Bostock v. Clayton County, 590 U.S. 644 (2020)
- U.S. Supreme Court, Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission, 584 U.S. 617 (2018)
- U.S. Supreme Court, Burwell v. Hobby Lobby Stores, Inc., 573 U.S. 682 (2014)
- Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq.
- Religious Freedom Restoration Act of 1993, 42 U.S.C. § 2000bb et seq.
- Fair Treatment of Religious Organizations Act, introduced by Rep. Blake Moore (R-UT), 2026
- Web search results provided as primary source material for this briefing