Preferring Works Without Faith, Nine Federal Agencies Would Make Religious Charities More Secular

Report Religious Liberty

Preferring Works Without Faith, Nine Federal Agencies Would Make Religious Charities More Secular

April 27, 2023 36 min read Download Report
Jack Fitzhenry
Legal Fellow, Meese Center
Jack is a Legal Fellow in The Heritage Foundation’s Edwin Meese III Center for Legal and Judicial Studies.


The Notice of Proposed Rulemaking from nine different federal agencies is a collective bid to rewrite the terms on which they partner with local religious organizations to provide an array of social services ranging from workforce training to (among others) homeless outreach and educational enrichment. The proposals have little justification and still less to recommend them. The agencies raise the alarm of Establishment Clause concerns while in the same breath proposing to insert themselves into questions of faith affecting religious organizations. For all of the Notice’s paeans to expanding beneficiaries’ access, one gets the impression that the agencies would sooner see such services interrupted or withdrawn rather than have certain beneficiaries encounter overt religious faith in the process.

Key Takeaways

Nine federal agencies are attempting to rewrite the terms on which they partner with local religious organizations to provide an array of social services.

The proposal creates needless doubt about religious organizations’ ability to retain their faith-based character and could violate their free exercise rights.

The agencies’ supposed Establishment Clause concerns are inadequate justification for the changes because these concerns are unfounded.

Regulators are required to justify the rationality of their actions to the public. Thus, whenever an agency proposes to act—for instance by adopting a new binding rule—it must identify some problem or lurking concern that its proposed action is meant to address. So often, however, when an agency’s explanation is examined, there is no real substance to the asserted problem—just a ghost haunting the otherwise barren imagination of the bureaucratic class.

Such is the case with the imagined “Establishment Clause concerns” invoked by nine different federal agenciesREF in their collective bid to rewrite the terms on which they partner with local religious organizations to provide an array of social services ranging from workforce training to (among others) homeless outreach and educational enrichment.

In their Notice of Proposed Rulemaking (Notice), the nine agencies propose to revise the definition of “indirect federal funding” by adding the words “wholly,” “genuinely,” and “private” to modify the word “choice” and by adding a sentence “stating that the availability of adequate secular alternatives is a significant factor in determining whether a program affords true private choice.”REF According to the agencies, these revisions are necessary to eliminate confusion caused by the prior 2020 rule, “to avoid Establishment Clause concerns,” and to “promote maximum participation by beneficiaries and providers in the Agencies’ covered programs and activities.”REF

However, the agencies offer no evidence that the current norms for partnering with religious organizations are an obstacle to beneficiaries’ maximal participation in federally funded services. Moreover, the putative “confusion” and “Establishment Clause concerns” are based on readings of U.S. Supreme Court jurisprudence that are nearly two decades out of date. The agencies also give short shrift to the Court’s more recent Establishment Clause decisions, which recognize that the once vogue “wall of separation” between church and state cannot be reconciled with the nation’s extensive history of partnership between the government and religious organizations to address society’s most persistent concerns and assist its most vulnerable members.

When regulators tilt at ghosts, they often create problems uglier than the ones they imagine they are fighting. The agencies’ proposal raises two very real specters: pervasive violations of the Free Exercise Clause and the en masse flight of religious organizations from participating in the agencies’ social service programs. The proposed revisions would create significant confusion for religious service providers who wish to retain the teachings of their faith as cornerstones of the services they provide to beneficiaries. Under the proposed rule, their ability to do so would hinge on an agency-by-agency assessment of whether “providers that offer secular programs are as a practical matter unavailable” to “particular beneficiaries.”REF That highly individualized, context-dependent assessment, which lacks any specific guardrails, would likely produce confusion among religious organizations and beneficiaries alike.

These effects would undermine the nine agencies’ stated goal of increasing both providers’ and beneficiaries’ participation in their social service programs. Rather than expanding access, the revisions would dissuade some religious service providers from participating in the agencies’ programs and induce others to withdraw in order to maintain their fundamentally religious character. The likely result would be fewer service providers available to beneficiaries. That is not merely an incidental cost: The regions where service providers are scarcest, and thus most needed, are precisely the regions where service providers would most likely be dissuaded from participating.

Little acknowledgment of these costs appears in the Notice. That does not mean, however, that the agencies are unaware of them. The rule’s unavoidable effect would be to make more religious organizations operate as though they were merely secular service providers. Because this is the rule’s obvious result, it is reasonable to assume that it also was the agencies’ intended result. That bias against religion and in favor of secularization is exemplified elsewhere in the rule, particularly in the agencies’ interpretation of Title VII’s religious organization exemptions, which ensure that a religious organization retains the right to employ persons based on their adherence to the tenets of the organization’s faith without incurring liability for employment discrimination under that title. Contrary to the text of Title VII and the weight of existing case law, the agencies opine that these exemptions do not apply whenever another protected classification such as race or sex (including sexual orientation or gender identity) is implicated. This would allow employee litigants to bypass their employer’s legitimate religious rationales and recast their employer’s adverse personnel decisions as unlawful bigotry.

Seen in this light, the proposed rule is one part of a larger evidentiary record demonstrating that this Administration will not tolerate rivals in the sphere of social policy. Chief among those rivals are organizations that dissent from secular progressive orthodoxy on personhood, sexuality, and the moral boundaries of health care. Where, as here, those organizations are indispensable in providing social services to the public, their distinctive religious character must be muted or eliminated so that they appear to be another de facto arm of the secular state.

The agencies may be betting that religious organizations are sufficiently dependent on the federal funds that flow through the agencies that they will choose to secularize instead of choosing not to participate. They may or may not be correct in that assessment, but if the result is that fewer beneficiaries rely on overtly religious organizations for care, then the implied goal of secularization is served, and the agencies are willing to force others, mostly beneficiaries, to bear the costs.

Reliance on Unsupported Assumptions

The agencies’ reasoning proceeds from several unstated but dubious assumptions. The logical connection between the proposed revisions and the goal of maximizing beneficiary participation is that the agencies’ current rules for partnering with religious service providers erect some obstacle between the agencies’ programs and eligible beneficiaries. What the obstacle is and who the affected beneficiaries are is nowhere expressly stated in the Notice. The Notice pays de rigueur obeisance to “historically marginalized communities,” but that vague designation informs the reader not one whit about how these differently situated communities are disserved by the current state of relations between the federal government and religious service providers.

Two further (but also unstated) assumptions offer possible answers. The first is that some beneficiaries are dissuaded from accessing the agencies’ programs because they object to the religious activities of their local service provider, either from a dislike of religion in general or from disagreement with the service provider’s particular faith expression. The second is that service providers’ religious convictions cause them to discriminate against certain beneficiaries, thereby making the agencies’ programs inaccessible to some.

The first problem is that the agencies have not offered so much as a single anecdote supporting the existence of either phenomenon. The second is that if one assumes that one or both of these situations exists, one also has to assume that the affected beneficiaries have no reasonable access to a secular service provider for the current state of affairs to be a cause of low participation. The agencies offer no evidence of the prevalence of this problem–if it even exists. Finally, to link the whole chain together, one must assume that the 2020 rule that the agencies seek to revise contributes to, or at least enables, the phenomenon causing lowered participation and the lack of secular options. Again, the Notice is devoid of evidence for these propositions.

Thus, the agencies begin their reform project on very unsure footing. If the goal is to increase beneficiary participation, the underlying problem is that too few eligible persons are partaking. If that is so, then the agencies must identify the cause and adopt a measure that is reasonably calculated to address the cause. But here doubt exists at each step, and a regulation responding to a specific problem is “highly capricious if that problem does not exist.”REF The agencies are not entitled to act on the basis of “unsupported speculation,” but instead must provide some “factual basis for this belief” that the 2020 rule and the relationship it establishes with religious service providers cause a concrete problem of underparticipation in the agencies’ programs.REF Where an agency lacks a factual basis for the belief motivating its regulatory choice, it necessarily lacks the reasoned explanation required for an agency’s action to be valid.

Misinterpreting the Establishment Clause

The agencies cite “Establishment Clause concerns” as their reason for revising the definition of “indirect federal funding.” The concern, premised on language in the Supreme Court’s 2002 decision in Zelman v. Simmons-Harris, is that where it is impractical for a beneficiary to access a secular service provider, the beneficiary is arguably compelled to direct his funds to a religious service provider, and that decision is therefore not truly voluntary.REF Supposedly, under Zelman, the lack of secular alternatives results in an impermissible use of federal funds because the government in discrete instances would be affecting an unconstitutional establishment of religion. That is not a plausible interpretation of the 20-year-old case. In addition, the Notice unpersuasively attempts to deny that more recent Supreme Court decisions compel a different assessment of the Establishment Clause issue.

Failure of Zelman and Its Progeny to Support Agencies’ Establishment Clause Concerns. The agencies misread Zelman by failing to understand the level at which a federal program must be assessed under the Establishment Clause. The question is not whether certain beneficiaries, for reasons not attributable to government action, feel constrained to use federal funds at a religious service provider. The question is simply whether the government funding program uses selection criteria that are neutral toward religion such that “recipients generally [a]re empowered to direct the aid to” service providers of their choosing.REF

The Zelman Court surveyed several prior Establishment Clause cases and repeatedly found that a program’s compliance with the Clause must be assessed at the level of its general operation, not at the level of individual beneficiaries. For instance:

  • The Court noted that it was able to uphold a “program authorizing tax deductions for various educational expenses” against an Establishment Clause challenge by “viewing the program as a whole” and determining that it enabled beneficiaries to make private choices among religious and secular options.REF
  • The same was true of a challenge brought against a “vocational scholarship program.” “Looking at the program as a whole,” the Court had no difficulty concluding that it allowed for private choice.REF
  • Finally, the Court upheld federal funding for sign-language interpreters in religious schools because, “[l]ooking once again to the challenged program as a whole, we observed that the program distributes benefits neutrally to any child qualifying as disabled.”REF

To satisfy the Establishment Clause, an individual’s choice need only be “independent” of the government’s choice, not from every practical constraint.REF Only when the “single” “unmediated” will of government is the force impelling beneficiaries away from secular options and toward religious ones is the Establishment Clause implicated.REF That is not the case where personal circumstances such as geography, lack of transportation, or work schedules place practical constraints on a beneficiary’s ability to access secular service providers. The neutrality of the programs’ criteria for approving service providers remains unaffected by these circumstances.

Zelman also referenced the “reasonable observer” standard for assessing whether a government action endorses religious practice. That standard’s roots in the now discredited Lemon v. KurtzmanREF decision ought to have warned the agencies that Zelman is perhaps not the best guide for applying the Establishment Clause.REF Setting that concern aside, however, Zelman still cautioned that “the reasonable observer in the endorsement inquiry must be deemed aware of the history and context” underlying a challenged program.REF A reasonable observer, aware of the neutrality and breadth of the programs administered by the agencies, would not interpret the circumstantial difficulties individual beneficiaries may face in reaching a secular service provider as a federal effort to establish religion. Rather, where, as here, “the religious, irreligious, and areligious are all alike eligible for governmental aid, no one would conclude that any indoctrination that any particular recipient conducts has been done at the behest of the government.”REF

The lower court cases cited in the Notice do nothing to detract from that reasoning.REF While those cases restated Zelman’s private choice standard, most of them sustained government programs against Establishment Clause challenges, and none relied solely on the isolated experience of an individual beneficiary to hold a funding program unconstitutional. The only (non-overruled) case to countenance an Establishment Clause challenge arose in a state penitentiary, where a Christian rehabilitation program was the sole option available to inmates.REF In that state-controlled environment, the complete absence of secular alternatives was readily attributable to the unmediated will of the state government, all the more so because the state failed to use neutral criteria when selecting the prison’s sole provider.REF Consequently, the general operation of the state program was not neutral toward religion, making it fundamentally distinct from the programs offered by the agencies here.

Returning to Zelman itself, the decision makes it clear that a standard based on the endlessly varied, ever-changing circumstances of individual beneficiaries cannot be the measure of a federal funding program’s compliance with the Establishment Clause: “Such an approach would scarcely provide the certainty that this field stands in need of, nor can we perceive principled standards by which such [] evidence might be evaluated.”REF Rather than avoiding Establishment Clause concerns, the agencies’ proposed approach would tend to manufacture concerns where none should exist. Zelman does not allow, let alone require, that approach.

Agencies’ Establishment Clause Concerns also Belied by More Recent Cases. Of course, Zelman was hardly the last or latest word on the constitutionality of federal funding to religious organizations. The agencies’ views on the Supreme Court’s more recent Establishment Clause and Free Exercise jurisprudence raise additional concerns about the justification for the proposed revisions and how they would be applied.

The Supreme Court’s decisions in Trinity Lutheran Church v. Comer,REF Espinoza v. Montana Department of Revenue,REF Carson v. Makin,REF and Kennedy v. Bremerton School DistrictREF make it still more evident than Zelman that the agencies’ purported Establishment Clause concerns are overwrought if not wholly imaginary. The disconnect is most apparent in the agencies’ concern “that the Government [] not [be] responsible for the use of the aid to support explicitly religious activities.”REF The agencies attempt to bolster this concern by stating that neither Carson v. Makin nor anything else in the Supreme Court’s recent decisions “affects the longstanding doctrine that the Establishment Clause generally prohibits the use of aid received directly from the government for ‘specifically’ or ‘inherently’ religious activities” as described in Bowen v. Kendrick.REF That interpretation of Carson’s effect and Bowen’s continued viability is sorely mistaken.

In Carson, the Court rejected attempts by the State of Maine, the lower courts, and the dissenting justices to justify discrimination against religious schools based on the schools’ use of government funds for, among other things, religious instruction. The Court stated unequivocally that its decisions have “never suggested that use-based discrimination is any less offensive to the Free Exercise Clause” and that “the prohibition on status-based discrimination under the Free Exercise Clause is not a permission to engage in use-based discrimination.”REF In light of this, the idea that government can still discriminate based on religious use if that religious use is “inherently religious” is fanciful if not disingenuous.

How, as a legal matter, the agencies (or any government arm) could consistently distinguish an inherently religious use from ordinary or acceptable religious use is a mystery. It is equally unclear how the agencies could separate the two in practice, but “[a]ny attempt to give effect to such a distinction by scrutinizing whether and how a religious [provider] pursues its [] mission would also raise serious concerns about state entanglement with religion and denominational favoritism.”REF Thus, the agencies’ solution would create more First Amendment concerns than it resolves.

Contrary to the agencies’ assertion,REF it is Carson that applies to their proposal and Bowen that is irrelevant. Bowen is a relic that “assess[ed] the constitutionality of an enactment by reference to the three factors first articulated in Lemon v. Kurtzman,” which decision the Supreme Court has since repudiated as a misguided attempt at a “grand unified theory” for assessing Establishment Clause claims.REF Bowen’s ban on funding “inherently” religious activities derived from the Lemon framework and thus is no more alive than Lemon itself. Bowen’s paradigmatic example of an inherently religious activity was one in which grantees “use[d] materials that have an explicitly religious content or are designed to inculcate the views of a particular religious faith.”REF This is precisely the use of government funding that the Court blessed in Carson without so much as a word of concern for whether that activity was “inherently,” and thus unacceptably, religious. That is because the distinction is a fiction and a dead one at that.

Still more recently, the Supreme Court affirmed in Kennedy v. Bremerton School District that the “Establishment Clause does not…compel the government to purge from the public sphere anything an objective observer could reasonably infer endorses or partakes of the religious.”REF Thus, it is untenable to maintain, as the agencies do, that Bowen’s carve-out for inherently religious activities somehow survived the Supreme Court’s most recent Establishment Clause decisions.

The sole exception that the Carson Court left intact was not Bowen, but Locke v. Davey,REF and that decision in no way supports the agencies’ contorted views of the Establishment Clause. Locke held that a state government did not violate the Free Exercise Clause when it denied a theology student a publicly funded scholarship to prepare for ministry. However, concerning the Establishment Clause, the Locke Court explained that “there is no doubt that the State could, consistent with the Federal Constitution, permit Promise Scholars to pursue a degree in devotional theology,” thereby belying the agencies’ concerns over religious use.REF If that were not clear enough, Carson reminds us that “Locke cannot be read to generally authorize the State to exclude religious persons from the enjoyment of public benefits on the basis of their anticipated religious use of the benefits.”REF Carson’s discussion of Locke forecloses the agencies’ attempt to rely on fears of indirect funding of “explicitly” or “inherently” religious activities as a valid basis for the proposed changes.

To interpret the Establishment Clause, the agencies must embrace, not resist, the application of Carson v. Makin to their efforts. That decision covered much the same problem that the agencies address here: the permissible intersection between government funding and religious activities where beneficiaries in remote regions would struggle to access necessary services because the government would not or could not provide them directly. Nothing in Carson suggested that the constitutionality of Maine’s program hinged on a beneficiary-by-beneficiary assessment of how practical it was for students to reach secular schools from their rural Maine towns. The Agencies’ attempts to escape the shadow of Carson here are therefore unavailing.

Agencies’ Failure to Consider the Nation’s History and Traditions. Agencies fail to engage in reasoned decision-making when they fail to consider a relevant aspect of the problem at issue.REF Where, as here, the Establishment Clause is motivating agency action, recent Supreme Court precedent dictates that history is a factor that must be considered. The Supreme Court has made clear that Establishment Clause objections to a particular government program or practice cannot be divorced from a historical inquiry into the role religion has played in the sphere under scrutiny.REF Here, the relevant history is the robust tradition from the Founding era onward of partnership between government and religious organizations in pursuit of vital social ends.

There is no doubt that ever since the nation’s Founding, religious organizations have played an indispensable role in providing social welfare services to the public.REF There is even evidence that religious organizations received public funds, including federal aid, to provide education and other social services to the public both before and after the Civil War.REF Portions of that historical record, particularly those relating to the pervasive public funding of religious educational institutions, have been discussed at length in the Supreme Court’s recent jurisprudence.REF

Such deeply rooted historical practices tend to belie concerns that federal funding of religious social service providers raises any Establishment Clause concerns. Yet no consideration of history prior to 2002 is apparent from the nine agencies’ Notice. That failure only highlights the deficiency of their Establishment Clause analysis.

Serious Free Exercise Concerns Raised by Agencies’ Proposals. The agencies’ fundamental error is that they read the Establishment Clause too broadly and the Free Exercise Clause too narrowly. In their zeal to maximize the rights of beneficiaries, they give short shrift to the Free Exercise rights of religious organizations.

Of particular concern is the agencies’ proposal to require religious providers to abide by the restrictions placed on direct funding of recipients when the agencies determine that a beneficiary’s decision to use the religious organization’s services is insufficiently voluntary.REF Compliance with those restrictions would require religious organizations to purge their services of all religious elements. The agencies envision this secularizing requirement as a solution to the obvious constitutional problem they would face if they terminated their relationship with a religious organization because of the organization’s religious character. But the Agencies are mistaken.

Requiring religious organizations to become secular in practice is no solution at all—it is simply religious discrimination in a subtler guise. The federal government “need not subsidize private [service providers]. But once a State decides to do so, it cannot disqualify some private [providers] solely because they are religious.”REF The Free Exercise Clause prohibits not only the termination of a service provider’s contract on religious grounds, but also the “indirect coercion or penalties on the free exercise of religion,”REF and “‘[i]t is too late in the day to doubt that the liberties of religion and expression may be infringed by the denial of or placing of conditions upon a benefit or privilege.’”REF Thus, the agencies cannot avoid the implications of cases like Carson and Trinity Lutheran simply by resorting to less drastic forms of religious discrimination than contract termination.REF

Moreover, while the Notice leaves the precise consequences of noncompliance unclear, the implication is that if an affected religious provider refused to segregate religious activities from its services, then the agencies would terminate their relationship and funding. The agencies indicate that such an outcome might be constitutionally acceptable because they would simply be treating religious organizations the same “as all other providers” are treated.REF That framing misunderstands the issue.

By requiring religious providers to become effectively secular as a condition of their continued receipt of funding, the agencies would be committing the same error as the State of Maine in Carson. Maine defended its policy of excluding religious schools from its funding program on the theory that the program was meant to provide “a rough equivalent of the public school education.”REF In this regard, religious schools were unlike the public schools and their secular private counterparts.

The Supreme Court did not allow the state to avoid the constitutional issue through a semantic reframing of the condition. At bottom, Maine sought to require that schools be secular to participate in an otherwise available public benefits program.REF In essence, that is what the agencies would require here: To remain eligible once an agency determined that some beneficiary had no secular alternatives, the religious organization would be forced to secularize or else be excluded. Here, as in Carson, “the definition of a particular program can always be manipulated to subsume the challenged condition,” but allowing the agencies to recast their secularization requirement as a form of impartial treatment “would be to see the First Amendment…reduced to a simple semantic exercise.”REF That outcome is not permissible.

Making a secular alternative available to the affected beneficiary would be a valid alternative whether it is accomplished by establishing a new service provider or by making resources such as transportation available to the beneficiary to ameliorate difficulties involved in accessing secular alternatives.REF When a religious service provider declines to separate out the religious aspects of its programming, it is these alternatives, not forced secularization or termination of the partnership, that should be pursued to avoid violating the Free Exercise Clause.

However, this solution raises other concerns. The agencies’ ability to establish secular options in all cases is doubtful. Of course, it is hard to assess just how practical that approach is because the agencies offer no sense of the frequency with which secular options are unavailable to beneficiaries. In any event, considerations of scarce resources—the same considerations that led the federal government to partner with private service providers in the first place—incentivize the agencies to compel religious service providers to purge the religious aspects of their services rather than bear the time and expense of creating a secular alternative. This is particularly so if one assumes that there will be a delay in creating the alternative during which the affected beneficiary is left to rely on the religious organization or go without services. One also must assume that the government would be willing to invest the resources necessary to create or accredit a new provider even though the circumstances giving rise to the need made it likely that the new secular provider would have few beneficiaries, presumably just the dissenters from the local religious program.

Thus, when, whether, and how the agencies plan to create new secular alternatives for isolated beneficiaries is unclear. This casts doubt on whether “expanding the universe of reasonably available…secular options”REF is a viable alternative to the impermissible compulsion of religious organizations.

Creating Needless Confusion for Religious Service Providers

The agencies assert that the 2020 rule sowed confusion among providers and beneficiaries. They contend that their proposed redefinition, by contrast, will offer providers “clarity” and “consistency.” But the proposal’s standardless, agency-by-agency, beneficiary-by-beneficiary approach to assessing the voluntary nature of every individual choice threatens to create considerable uncertainty, not to mention inconsistent—if not outright arbitrary—results.

The agencies propose to reintroduce a third step to the test for indirect federal funding that asks whether a “beneficiary has at least one adequate secular option for the use of the voucher, certificate, or other similar means of government-funded payment.”REF The agencies indicate that when “providers that offer secular programs are as a practical matter unavailable” to any beneficiary, a beneficiary’s choice of a religious organization would not be a “genuine and independent private choice.”REF

Putting aside the constitutional infirmities of this approach, it has few guiding principles and even fewer limiting principles. The agencies speak in terms of “the potential availability to beneficiaries of a practical [secular] option,” the absence of which will be “significant” in determining whether a beneficiary is capable of a “wholly…genuine and independent private choice” among providers.REF This grab bag of modifiers does nothing to elucidate the agencies’ intended approach; it merely creates interpretive space in which the agencies can muse in their discretion about whether an apparently independent choice is in fact not “wholly” voluntary.REF

  • What, for instance, does it mean for a secular option to be “practically” available?
  • What personal or circumstantial difficulties will an agency consider in determining availability?
  • What weight would geography, limited transit options, inflexible work schedules, or unreliable childcare options receive in assessing a beneficiary’s ability to reach a more distant secular option when a religious one is nearby?
  • At what point on the sliding scale of voluntariness does an individual’s choice become less than “wholly” independent?

Never mind that many of the considerations driving the inquiry will be well outside the control of religious organizations whose status and daily operations will now depend to a “significant” extent on these accidents of chance.

This is not an instance of agencies’ adopting a well-developed standard. Neither Zelman nor any of the lower court cases cited by the agencies provide a test for assessing when an individual’s choice is sufficiently free and independent. Nor could they because, as explained above, those decisions assessed the challenged government funding programs only at the level of their general operation, not at the level of an individual beneficiary’s access.

Apart from general confusion, the likely effect of this approach would be to reduce provider participation and, by extension, the availability of services offered under the agencies’ programs. The redefinition’s focus on individual beneficiaries’ difficulties would herd religious organizations out of the category of indirect federal funding recipients and into the category of direct recipients in which the agencies insist providers can be forced to secularize their activities. When one affected beneficiary is enough to bring about this change in status, many existing religious service providers will find their positions too precarious and, in order to maintain their religious activities, will decline to participate in the agencies’ programs.

The same logic would dissuade other religious organizations from becoming providers in the agencies’ programs, and attrition is likely to occur in areas where service providers are scarcest. It is in remote or rural regions that beneficiaries are less likely to have secular alternatives, making their choice of religious organizations not wholly voluntary according to the agencies. As a result, these will be the areas where religious organizations will most often face the choice of secularizing or losing federal funding. A decision to preserve their religious nature would cost religious providers revenues, but it would also cost rural beneficiaries, who might lose their only option for accessing services. This is contrary both to the agencies’ general goal of expanding participation and to their more specific goal of advancing support for “underserved communities” including rural populations.REF

The agencies should abandon their misguided proposal for the same reasons the Supreme Court abandoned Lemon: The test “invited chaos” in application, “led to differing results in materially identical cases, and created a minefield” for those bound by its assessments.REF The same outcomes are likely to result from so broad and formless an approach as the one the agencies’ propose in their Notice.

The practical implausibility of developing a principled, consistent means of applying this new standard points us back to the constitutional concerns. “The protections of the Free Exercise Clause,” as the Supreme Court has noted, “do not depend on a ‘judgment-by-judgment analysis’ regarding whether discrimination against religious adherents would somehow serve ill-defined interests.”REF Further, the agencies’ approach, which would allow the circumstances of one beneficiary to affect the religious practice of an entire organization, raises a considerable risk of reinstituting the “modified heckler’s veto, in which…religious activity can be proscribed based on perceptions or discomfort” of certain beneficiaries.REF That, like so many other effects of this proposal, would be unconstitutional. Because the agencies’ rulings will determine whether religious organizations can maintain the faith-based aspects of their services, too much is at stake for a discretionary ad hoc approach to prevail—all the more so because while the rule’s negative implications for providers and beneficiaries are real, the supposed Establishment Clause problems are not: “[I]n no world may a government entity’s concerns about phantom constitutional violations justify actual violations of an individual’s First Amendment rights.”REF

Making Religious Exemptions Impermissibly Narrow

Elsewhere in the Notice, the agencies propose to clarify the scope of Title VII’s religious exemptions by removing language from the 2020 rule that the agencies fear “could mistakenly suggest that Title VII permits religious organizations…to insist upon tenets-based employment conditions that would otherwise violate Title VII.”REF Here again, the agencies’ proposed solution is to be feared more than the purported problem it addresses. In practice, the reinterpreted religious exemptions would be viable defenses only against claims of religious discrimination. The agencies propose to eliminate the exemptions whenever the religious employer’s decision happens to affect an individual who qualifies for another protected classification under Title VII, such as race or sex (including sexual orientation or gender identity).REF

As many critical comments noted, the agencies’ efforts to reduce the exemptions’ scope are neither faithful to the law’s text nor consistent with the case law interpreting the exemptions.REF While religious employers are not wholly exempt from Title VII, they are entitled to invoke the religious exemptions when they seek to hire employees “of a particular religion.”REF Title VII defines religion to encompass more than adherence to a particular profession of faith; the statutory term “religion” appropriately “includes all aspects of religious observance and practice.”REF Thus, the “exemptions create, as to Title VII, a freedom on the part of religious employers to have religion-based employee conduct standards” that are applicable to all of the organization’s employees whether they are co-religionists or not.REF

By Title VII’s clear terms, if the religious exemptions apply, then “This title [Title VII] shall not apply,” and the religious employer may lawfully engage in faith-based hiring and retention practices “[n]otwithstanding any other provision of this title.”REF The statutory text does not state that particular protections or classifications no longer apply when the exemptions are invoked; it states that Title VII in its entirety, including various protected classifications and anti-retaliation provisions, shall not apply. The agencies’ contrary view that the exemptions evaporate whenever non-religious classifications are affected is therefore wholly inconsistent with the text of Title VII.

Case law is no more favorable to the agencies’ reading. Comments submitted in opposition to the rule offered a wealth of cases illustrating that the religious exemptions retain their full force and effect even where the affected employee is the beneficiary of another classification protected by Title VII.REF Moreover, the cases establish beyond doubt that religious employers may apply the standards of their faith to assess and punish employee conduct.REF

In addition, the problem raised by the agencies’ reinterpretation of the religious exemptions is much the same as the problem raised by their proposal to separate out the specious subcategory of “inherently religious” uses from other less religious uses, namely the problem of entanglement. To the agencies, it would be of no consequence that a religious employer can establish a valid religious reason for terminating an employee if that employee has the good fortune to be a member of some other protected class under Title VII. By asserting that certain identities overcome an otherwise valid religious reason, the agencies are proposing in effect to appoint themselves adjudicators of whether or not an employee’s conduct comports adequately with the tenets of the employer’s faith. The prospects for entanglement between bureaucrats and faith on questions of doctrine are legion, illustrating just how ill-conceived the agencies’ desired revisions and reinterpretations are. Then again, the agencies may be well aware of the prospects for entanglement; they may simply enjoy the prospect of playing ecclesial authority from their high secular perch.

Insufficiency of New Definitions of Financial Assistance

Finally, the agencies’ “Request for Comments on Regulatory Definitions of ‘Federal Financial Assistance’ or ‘Financial Assistance’” affords an inadequate basis for final rulemaking. First, the impetus for a change in the definition is unclear. The Notice criticizes the 2020 rule for adopting definitions of “Federal Financial Assistance” that deviated from the one contained in Executive Order 13279.REF After criticizing the 2020 rule for its deviation, the Notice then solicits comment on whether the agencies should deviate in another fashion by “us[ing] any definition other than that in Executive Order 13279.”REF

The putative problem with the definition in the 2020 rule is “confusion and possible misunderstandings” in some undefined sense. But the agencies offer no clue regarding their assessment of the definition contained in Executive Order 13279: whether it is overinclusive or underinclusive, whether it has been workable in application, whether the balance struck by the current definition aligns with agency priorities, etc.

Meaningful comment is not possible when parties are left to “divine [the agency's] unspoken thoughts.”REF Because the Notice contains no specific proposal, no discussion of the problems with Executive Order 13279’s definition, and not even a discussion of the ends to be achieved through redefinition of federal financial assistance, the public can only guess at what course the agencies may adopt and their reasons for doing so.REF Thus, any final rule would not be the “logical outgrowth” of this Notice.REF Any reformulation of the definition that the agencies intend to pursue must be the subject of a separate notice with additional opportunity for public comment.


The agencies’ proposals have little justification and still less to recommend them. The nine agencies raise the alarm of Establishment Clause concerns while in the same breath proposing to insert themselves into the questions of faith affecting religious organizations. If the agencies appreciate that irony, they are keeping the humor to themselves. For all the Notice’s paeans to expanding beneficiaries’ access, one gets the impression that the agencies would sooner see such services interrupted or withdrawn rather than have certain beneficiaries encounter overt religious faith in the process.

The agencies’ service programs assist vulnerable populations—the homeless, “dislocated workers,” students in high-poverty, low-performing schools, etc. Their material need is enough to attract the government’s attention, and the government therefore assumes that material need—and nothing more—should suffice to motivate those persons or groups that would serve these populations. But since well before our Founding, it has been the spiritual promptings of faith that have inspired many men and women to make the alleviation of human suffering and the cultivation of human potential their life’s work. Their collective efforts were, and still remain, so extensive that the federal government long ago recognized that it could best address persistent societal issues by entering into partnerships with the religious organizations that these men and women operate.

To the current Administration, however, it seems nearly intolerable that public funds should assist people or organizations that are motivated by religious considerations. Haunting the Administration is a fashionable caricature that depicts religion as an invitation to divisive bigotry rather than the inspiration of surpassing human kindness. Thus, to guard against religion’s imagined incivility, the Administration’s message is clear: Where the federal government is concerned, we shall have only works without faith.

Jack Fitzhenry is a Legal Fellow in the Edwin Meese III Center for Legal and Judicial Studies at The Heritage Foundation.


Jack Fitzhenry
Jack Fitzhenry

Legal Fellow, Meese Center