Religious Schools, Collective Bargaining, & the Constitutional Legacy of NLRB v. Catholic Bishop

It would be difficult to find a corner of American labor law more anomalous than the one covering religious schools. Nearly half a century ago, in National Labor Relations Board v. Catholic Bishop,[1] the Supreme Court excluded those schools from the Board’s jurisdiction. It did that by reading the National Labor Relations Act narrowly: it reasoned that because the Act never mentioned religious schools, Congress must have meant to exclude them. In other words, the Court anticipated Justice Neil Gorsuch’s Canon of Donut Holes.[2]

That logic was, to put it generously, unorthodox. But the Court had its reasons. It paid little attention to the statutory language, focusing instead on the effect any other interpretation would have had on the schools’ constitutional rights. Had the Board been given jurisdiction over the schools, it would have been responsible for policing collective bargaining and investigating alleged unfair labor practices in religious schools. Both activities would have forced it to question the schools’ motivations in various contexts, which would have led it into disputes often grounded in religion. And in that way, the Board risked colliding with core First Amendment activity. Unwilling to stomach that risk, the Court avoided it by reading an exception into the law.

The Court’s decision, however, was hardly the last word. In the decades that followed, the Board launched effort after effort to reassert jurisdiction over the schools. It formulated multiple tests and theories, each of which aimed to bring the schools back under its purview. Perhaps predictably, those theories were rebuffed by lower courts. The courts saw the theories for what they were: post hoc attempts to limit Catholic Bishop’s scope. And the courts proved more than willing to defend Catholic Bishop’s core holding, despite its counterintuitive rationale.

They proved less willing, however, to apply the same rigor to similar efforts by the states. Even as the Board was trying and failing to reassert jurisdiction, states were rushing to fill the gap. New York, New Jersey, and Minnesota applied their own labor-relations laws to religious schools. They reasoned that Catholic Bishop addressed only the scope of federal statutory law; it had nothing to say about state law. And courts gave that logic their stamp of approval. They held Catholic Bishop’s black-letter holding dealt only with the NLRA. It had no import for questions of state law.

That approach presents us with a puzzle. It is well accepted that one should not read a decision only for its core holding.[3] The rationale producing that holding is at least as important.[4] And Catholic Bishop’s rationale should have led courts to reject the application of state labor-relations laws to religious schools. At Catholic Bishop’s core was the doctrine of constitutional avoidance: the Court strained to read the Act as it did because a different reading would have produced an unacceptable risk to First Amendment rights. And courts have long recognized that the First Amendment applies with the same force to the states as it does to the federal government. The same analysis, then, should have applied whether jurisdiction was being asserted by the Board or by a state agency. In either case, Catholic Bishop should have led lower courts to avoid a conflict by denying states regulatory jurisdiction.

Yet for whatever reason, they failed to approach the question that way. And so a dichotomy has persisted in the law. Even today, after the Board has given up any hope of reinserting itself into religious schools, state agencies continue to regulate them. That is, states continue to do exactly what Catholic Bishop said the Board could not. And with each passing year, the dichotomy grows harder to defend. The Supreme Court has repeatedly emphasized that the First Amendment protects religious schools’ control over their internal affairs—including their relationships with their employees. Meanwhile, scholars, lower courts, and the Supreme Court itself have questioned one of the key precedents used to justify state involvement in religious schools—Employment Division v. Smith.[5]

This tension cannot hold. At some point, courts will recognize the illogic of allowing states to do what the Board cannot in this context. The Board cannot require religious schools to bargain with a union because to do so would put constitutional rights at risk. That risk is no less present when the regulator is a state agency. And though courts currently distinguish between the two situations, that distinction is untenable. It has no principled undergirding. It appears to be no more than an unnoticed inconsistency—a wrinkle in the law yet to be ironed out.

 In an ideal world, the states would wield the iron themselves. They would recognize the potential damage to First Amendment rights and would withdraw from religious schools. But in the real world, they have shown no inclination to do so. More likely, they will back out only when they face the coercive force of a court order. Courts perpetuated this inconsistency by failing to give Catholic Bishop its full effect. It will be up to courts to set things right.

I. Catholic Bishop and Its Aftermath: The Board Moves Out, and States Move In

The story of the Board’s jurisdiction is one of slow mission creep. Technically, the Board has jurisdiction over nearly all private employers.[6] Its reach extends to the full scope of Congress’s power under the Commerce Clause, which means that the Board can regulate any employer whose activity has a substantial effect on interstate commerce.[7] But the Board has always exercised less than that full power. It has adopted voluntary jurisdictional limits, stated in terms of annual revenue.[8] Depending on the type of employer, these revenue floors can go as high as $500,000 a year.[9] If an employer generates less than that, it falls outside the Board’s jurisdiction.[10]

Besides these revenue limits, the Board has sometimes declined jurisdiction over certain industries or activities on an ad hoc basis. For example, it has excluded horse-racing tracks from its scope by regulation.[11] And at various points in its history, it has declined jurisdiction over student athletes, explaining that its involvement in their relationship with schools wouldn’t advance the NLRA’s goals.[12]

For years, that was the approach the Board took toward nonprofit schools.[13] The schools, it reasoned, had only a limited effect on interstate commerce, and it made little sense to dedicate resources to policing their labor relations. So the Board declined jurisdiction over them, effectively carving out an ad hoc exception.[14]

But in the 1970s, it abandoned that approach.[15] Instead, it decided that some schools had a large enough effect on commerce to justify regulation.[16] So it started asserting jurisdiction.[17] But even then, it continued to impose some limits. In particular, with religious schools, it drew a line between “completely religious” institutions and those that were merely “religiously associated.”[18] It declined jurisdiction over the former, but claimed full authority over the latter.[19]

Catholic Bishop brought that approach to the Supreme Court. The case involved two sets of high schools: one operated by the Catholic Church in Chicago, the other by the Diocese of Ft. Wayne–South Bend.[20] Both sets of schools offered secular and religious instruction, using both lay and religiously trained teachers.[21] In the mid-70s, two unions petitioned to represent the teachers. The Board accepted the petitions and certified election units comprising all full- and part-time teachers. It excluded, however, all “religious faculty,” a term it did not define.[22] Despite this carveout, the schools resisted the petitions. They argued that government-mandated bargaining—even limited to lay teachers—would violate their First Amendment rights. As religious institutions, they enjoyed a protected sphere of autonomy over their internal affairs. And bargaining, they said, would drain their discretion over those affairs and invade their autonomy.[23]

The Board disagreed. Applying its “completely religious” test, it found that the schools were too secular to qualify for an exemption. For example, they had sought and received accreditation from a secular regional authority.[24] They had also admitted non-Catholic students, employed non-Catholic teachers, and offered a mix of religious and secular instruction.[25] Indeed, their secular instruction looked much like the instruction found in any secular college-prep course.[26] So the schools could not claim to be “completely religious”; they were merely associated with a religious institution, which would not justify an exemption.[27]

On review, the Seventh Circuit rejected the Board’s reasoning.[28] The court saw no proper way to distinguish between “completely religious” and “religiously associated” schools. To draw that distinction, the Board would have to measure an institution’s “degree of religiosity,” and such an inquiry “would perforce involve [the Board] in answering the sensitive question as to how far religion pervades that institution.”[29] That was a question the Board, as a government agency, had no constitutional competence to answer.[30]

But the Board’s test wasn’t the only problem. The court reasoned that even had the Board developed a more workable test, government-mandated bargaining would still have interfered with the schools’ internal affairs. By definition, collective bargaining takes some control from management and gives it to a union.[31] Management and the union effectively share control over key decisions affecting wages and working conditions. And in a religious institution, control over management can have doctrinal significance. For example, canon law gave the bishop complete control over parochial schools.[32] His discretion over their activities was a matter of doctrine.[33] But mandatory bargaining would have forced him to share his discretion with a third party.[34] Important institutional decisions would no longer be his to make: they would instead require consultation and negotiation.[35] That kind of shared control could not be squared with the church’s internal law.[36]

Nor did the problem stop with bargaining itself. To ensure bargaining proceeded apace, the Board would have to investigate alleged unfair labor practices.[37] And those investigations would inevitably draw the Board into religious disputes. For example, in Catholic Bishop, the schools had terminated three teachers. The schools offered religious reasons for the terminations: one teacher had exposed biology students to unapproved sexual theories; another had married a divorced Catholic; and a third had refused to restructure a course according to instructions from the religion department. The Board admitted that had these been the schools’ true motivations, it would have owed them some kind of “reasonable accommodation.”[38] Yet it still ordered the schools to put the teachers back to work. In the Board’s view, the schools had acted not for their purported religious motivations, but for unlawful discriminatory ones—they had retaliated against the teachers for union activity.[39]

For the court, that kind of second-guessing was inappropriate when dealing with religious schools. The Board could not properly assess the schools’ motivations when those motivations implicated religious doctrine.[40] Government officials have no competence in religious matters; they cannot inquire into the veracity or sincerity of an asserted religious belief. Yet under the Board’s approach, that kind of inquiry would occur whenever the Board investigated an action the school took for ostensibly religious reasons.[41] To decide whether those reasons were sufficient, the Board would have to make a judgment call about the veracity or importance of the school’s beliefs.[42] And that was a course barred to the government by the First Amendment.[43]

The court saw no way through this constitutional thicket.[44] There was no way to command the school to bargain while still respecting its autonomy. There was no way to evaluate unfair labor practices without digging into the school’s beliefs. And so there was no way for the Board to properly supervise bargaining in religious schools.[45]

The Supreme Court affirmed that decision, but on different grounds.[46] Like the Seventh Circuit, it saw serious constitutional problems with extending the Board’s jurisdiction to religious schools. The schools were religious institutions: they existed only because the church wanted to offer a religious alternative to secular education.[47] Their main purpose was to help the church pass its faith on to the next generation. Religious authority, then, “necessarily pervade[d]” their operations.[48]

The Board would deeply entangle itself in those operations by enforcing mandatory bargaining. Bargaining would touch on all manner of school policies. To resolve disputes over those policies, the Board would often have to ask questions about religious doctrine.[49] And though it might try to answer those questions in a way respectful to the schools’ beliefs, merely asking the questions would draw it onto shaky constitutional ground.[50]

The Court saw no clean way around this problem. There was “no escape from conflicts flowing from the Board’s exercise of jurisdiction over teachers in church-operated schools and the consequent serious First Amendment questions that would follow.”[51] The Board’s presence in religious schools, no matter how limited or tailored, would repeatedly cause constitutional conflicts.[52] 

But rather than address those constitutional questions directly, as the Seventh Circuit had, the Court avoided them through statutory interpretation. It observed that although the NLRA applied to “employers” generally, nothing in the text directly addressed religious schools.[53] Nor had Congress addressed those schools at any point in the legislative process.[54] In other words, there was no evidence that Congress wanted the Board to wade into such a constitutionally fraught workplace.[55] And absent strong evidence on that point, the Court refused to assume that the Board’s authority reached so far. So it read an exception into the Act and denied the Board jurisdiction over religious schools.[56]

A. The Board Tinkers with the Regulation of Religion

On its face, Catholic Bishop’s conclusion was absolute: the Board had no jurisdiction over religious schools. And while its holding was statutory, its analysis was constitutional. Whenever the Board exercised jurisdiction over religious schools, it risked violating the First Amendment. It therefore had no business regulating those schools. There was no gray area. 

But not everyone read the decision that way. The Board, for one, thought Catholic Bishop left open a gap—a gap the Board would spend the next fifty years trying to pry open.

At first, the Board tried to limit Catholic Bishop to primary and secondary schools. It distinguished those schools from colleges and universities, where the students were less impressionable and the faculty more independent.[57] Indeed, the teachers there enjoyed “academic freedom,” further insulating them from the school’s institutional (i.e., religious) views.[58] So, the reasoning went, the teachers were less entwined in the institution’s religious mission, and their relationship with the institution was more grounded in mundane workplace realities. That meant the Board could assert jurisdiction over them without bumping up against the First Amendment.[59]

But that approach ran aground in the courts. In Universidad Central de Bayamon v. NLRB,[60] the First Circuit rejected the Board’s distinction between high schools and universities.[61] In an opinion by then-Judge Stephen Breyer, the court reasoned that even in a university, religion could still permeate an educational environment. Religion could still inform the university’s instruction, course offerings, and academic decisions. And in such an environment, the Board would still have to draw knotty lines between religious and secular matters. These lines would present themselves whenever the Board certified a bargaining unit, enforced bargaining obligations, or investigated unfair labor practices. Nothing about the nature of higher education suggested that the lines would disappear. They had nothing to do with how advanced the students were, or whether the teachers enjoyed academic freedom. But they had everything to do with the school’s religious mission—a mission that could pervade a university just as much as a high school.[62]

Undeterred, the Board changed tack.[63] While the courts continued to block it from asserting jurisdiction over religious schools, it still had to decide whether a school was religious in the first place. This, it thought, gave it another opening. So it started asking whether the school in question had a “substantial religious character.”[64] If so, the Board would decline jurisdiction. But if not, it would regulate at will.[65]

That approach fared no better in court. In Great Falls University v. NLRB,[66] the D.C. Circuit held that even this new approach veered too far into forbidden territory. The court took it as a given that the government has no competence in religious affairs: no public official can evaluate a person’s or institution’s beliefs, let alone decide whether those beliefs are “substantial.”[67] Yet the Board’s new test called for just that kind of distinction. To apply its new standard, it would have to comb through a school’s practices and draw a conclusion about the school’s fundamental character. That kind of evaluation was exactly what Catholic Bishop meant to avoid.[68]

To cut off any more maneuvering, the court announced a bright-line test. A school would be beyond the Board’s jurisdiction if it (1) held itself out as providing a religious educational environment; (2) was organized as a nonprofit; and (3) was affiliated with a religious institution.[69] Those three criteria comprised the entire inquiry. If all three were present, the Board could ask no more questions.[70]

Yet ask the Board did. Several years later, the Board shifted its focus again, this time from the schools to the employees. In Pacific Lutheran University,[71] it held that to avoid regulation, a school would have to show that the individual employees played a role in the school’s religious mission. If they didn’t, the Board would assert jurisdiction regardless of the school’s overall character. That is, rather than focus on the institution’s mission, the Board would evaluate the employees’ duties.[72]

The D.C. Circuit swiftly rejected that approach as well. In Duquesne University v. NLRB,[73] the court reiterated that Catholic Bishop and Great Falls left no loopholes. Catholic Bishop meant what it said: the Board had no business in religious schools.[74] And the Great Falls test was absolute: if a school satisfied its three criteria, the Board lacked jurisdiction.[75] The Board could not evade that result by shifting its focus. It was no less invasive to ask about the religiosity of individual jobs than it was to ask about the religiosity of whole institutions.[76] In either case, the Board would have to wade into questions about religious belief and doctrine—questions it had no competence to answer.[77]

Finally, after more than five decades of resistance, the Board accepted defeat. In a 2020 decision, Bethany College,[78] it recognized that Catholic Bishop stripped it of all jurisdiction over religious schools. Going forward, it would follow the bright-line test from Great Falls.[79] It would no longer try to police the relationship between religious schools and their teachers.[80] Instead, it would leave the schools to manage their own internal affairs.[81]

B. States Step into the Breach

As Aristotle famously (and perhaps apocryphally) said, nature abhors a vacuum. That is no truer in nature than it is in law. For even as the Board was struggling to find a foothold in religious schools, states recognized an opening, and they rushed in to fill it.

In most workplaces, states have no authority to regulate collective bargaining. The NLRA is a comprehensive regulatory system, and so it preempts state efforts to regulate the same subjects.[82] A state cannot, for example, provide additional remedies for federal unfair labor practices.[83] Nor can it require bargaining over subjects federal law leaves to the interplay of free-market forces (for example, strikes).[84] In fact, some courts have held that states cannot even encourage collective bargaining, as any effort to rebalance the incentives set by Congress would interfere with the federal scheme.[85] This principle is quite broad, and it leaves states with little if anything to say about labor relations in most workplaces.[86] 

The principle does, however, admit a few exceptions. For one, states are free to regulate workplaces over which the Board has declined jurisdiction—or over which it lacked jurisdiction in the first place.[87] So for example, states can create collective-bargaining systems for their own employees.[88] They can also create systems for agricultural workplaces or businesses too small to qualify for federal jurisdiction.[89] In these cases, federal law either does not reach the workplace or the Board has decided, as a matter of policy, to leave the workplace unregulated. That regulatory gap leaves a space for states to act.[90]

Some states saw just such a gap in the wake of Catholic Bishop. They reasoned that the Court denied the Board jurisdiction not because of any constitutional problem, but because Congress had provided no statutory authority.[91] In other words, they argued, the only problem was that Congress had not been clearer about its intent to regulate religious schools. And in that sense, religious schools were really no different than agricultural or public workplaces. The schools may have been outside the Board’s remit, but they were fair game for states.

1. New York

The first state to act was New York. When the state originally adopted its labor-relations law in the 1930s, it exempted charitable, educational, and religious employers.[92] But in the late 1960s, it amended the law to cover those institutions.[93] That amendment gave the state’s agencies a statutory hook for regulating the schools—exactly the hook the Board lacked in Catholic Bishop. That is, whereas federal law withheld authority implicitly, state law supplied it explicitly.[94]

Unions wasted no time in taking advantage of New York’s more explicit coverage—an effort that eventually brought the issue to the U.S. Court of Appeals for the Second Circuit. In Catholic High School Ass’n of Archdiocese v. Culvert,[95] the court considered whether the state could apply its law to a group of Catholic high schools. The schools had a history of voluntarily bargaining with a union representing their lay teachers. Over the years, they had signed several collective-bargaining agreements with the union. But in 1980, they adopted a new substitution policy without bargaining about it first. Some of the teachers went on strike in protest. The schools suspended those teachers, prompting the union to file unfair-labor-practice charges for the first time.[96] In response, the schools argued that despite the state statute, New York had no jurisdiction over their internal affairs. Exercising jurisdiction, they said, would run afoul of Catholic Bishop and the First Amendment.[97]

The Second Circuit disagreed. It saw Catholic Bishop as addressing only a statutory question.[98] It thought the Court had declined to answer the constitutional question—whether mandatory collective bargaining in religious schools violated the First Amendment.[99] And on that question, the Second Circuit saw no conflict between the First Amendment and state labor law. Neither mandatory bargaining nor unfair-labor-practice investigations infringed on religious exercise.[100] Bargaining, for one, caused no excessive entanglement or interference with religious affairs. The state dictated no particular outcome in bargaining; it had no say in the terms the parties reached. Instead, it merely brought them to the table and left them to their negotiations.[101] And as for unfair labor practices, the court viewed them as inherently secular.[102] A state could forbid anti-union practices without interfering with religious exercise. True, there would be cases presenting conflicting motivations: the union would say the school acted out of anti-union animus, and the school would say it acted out of religious conviction. But the state could resolve that kind of conflict by applying a mixed-motives analysis. That is, the state could decide whether the alleged unlawful motivation would have led the school to act even without the religious one. And if the school would have done the same thing, the state could grant relief.[103]

The same reasoning prevailed in New York’s state courts. A decade later, in Nyserb v. Christ King School,[104] the New York Court of Appeals took up the constitutional question and reached the same answer. The court relied not only on the Second Circuit’s opinion in Culvert, but also the Supreme Court’s intervening opinion in Employment Division v. Smith.[105] Decided in 1990, Smith had given constitutional approval to neutral, generally applicable laws, even those laws burdening religious practices.[106] Drawing on that principle, the Court of Appeals found that New York’s labor law passed constitutional muster. The law applied neutrally and generally across all employers. It did not target religious practice.[107] It aimed instead at promoting collective bargaining across the state’s economy.[108] And so whatever incidental burdens it placed on religious exercise were of no constitutional significance.[109]

That view prevailed over the following decades. Though the state shifted regulatory responsibility among various agencies,[110] it held firmly to its view that collective bargaining could be mandated in religious schools.[111]

2. New Jersey

The same year Nyserb came down, the New Jersey Supreme Court reached a similar result. In South Jersey Catholic School Teachers v. St. Teresa, it held that religious-school teachers had a right to bargain collectively under the state constitution.[112] It also held that the federal constitution presented no bar or competing mandate.[113]

St. Teresa involved a group of elementary schools run by the Diocese of Camden. When teachers at these schools formed a union, the Diocese refused to bargain, and the teachers sued for recognition. The teachers pointed to a provision of the state constitution guaranteeing the right to collective bargaining.[114] They argued that the provision applied to all “private employment,” including employment in religious schools.[115] In response, the Diocese argued that forcing it to bargain under the state constitution would violate its rights under the federal one. That is, according to the Diocese, Catholic Bishop barred the state from asserting jurisdiction.[116]

Although a trial court sided with the schools, the New Jersey Supreme Court reversed.[117] Like its sister court in New York, the New Jersey court saw Catholic Bishop as no barrier. Catholic Bishop, the court reasoned, dealt only with statutory interpretation.[118] The U.S. Supreme Court had been able to avoid the constitutional question because Congress had failed to clearly signal its intent.[119] That type of constitutional avoidance, however, was unavailable in New Jersey, where the state’s constitution explicitly guaranteed the right to bargain in all “private employment.”[120] So the court had no choice but to answer the First Amendment question itself.[121]

Relying largely on Smith, the court found no free exercise problem.[122] The state constitutional guarantee applied to religious and non-religious employers alike. It in no way targeted religion.[123] And its goals were obviously secular: it aimed to promote collective bargaining in all employment, and thus to strengthen all workers’ positions vis-à-vis their employers. It was, in other words, neutral and generally applicable.[124] It therefore passed the Smith test and raised no concerns under the Free Exercise Clause.[125]

Even so, the court recognized that the scheme, if pursued too far, could raise constitutional concerns. For example, the state probably could not force a school to negotiate over overtly religious topics, such as a teacher’s moral qualifications.[126] That kind of mandate would drag the state directly into religious disputes. So the court drew a line between religious and secular subjects. The latter could be the subject of mandatory bargaining, while the former could not.[127]

Separating secular from religious subjects was, of course, no easy task. But the court still concluded that it could be done. As evidence, it pointed to a collective-bargaining agreement the Diocese had voluntarily negotiated for some of its high schools.[128] That agreement dealt only with financial terms, such as salaries and benefits.[129] It explicitly reserved the Diocese’s authority over potentially religious subjects, such as educational policies, discipline, assignments, accountability, class ratios, and other canonical or religious matters.[130] The court reasoned that if the Diocese could negotiate such an agreement with its high-school teachers, it could surely negotiate a similar one with its elementary teachers.[131] It therefore ordered the Diocese to bargain over the same subjects with the union.[132]

3. Minnesota

This mode of reasoning prevailed outside the Northeast as well. In Hill-Murray Federation v. Hill-Murray High School,[133] the Minnesota Supreme Court likewise held that the state could constitutionally mandate bargaining between a religious school and its teachers. And like its northeastern counterparts, it relied heavily on Smith.

Hill-Murray involved a high school run by a nonprofit corporation associated with the St. Paul Priory. About eighty-five percent of the school’s students were Catholic.[134] Along with secular instruction, the school offered religion courses and monthly mass services.[135] Many of its teachers, however, were of different faiths.[136] Unless they worked in the religion department, they could practice any faith they chose. They could also use an internal grievance procedure, which the school had voluntarily adopted. (Teachers in the religion department, by contrast, could be fired at the Archbishop’s discretion.)[137]

Seeking to represent the teachers, a union petitioned the Minnesota Bureau of Mediation Services. The school resisted the petition, making the now-familiar argument that Catholic Bishop barred the state from asserting jurisdiction.[138] The Bureau disagreed and certified an election unit of teachers outside the religion department.[139] A court of appeals refused to enforce the Bureau’s judgment, but the Minnesota Supreme Court reversed.[140] 

Like the New York and New Jersey courts before it, the Minnesota court saw the question largely as a free exercise issue. Relying on Smith, it characterized Minnesota’s labor-relations law as a neutral law of general applicability.[141] The law on its face applied to all employers, religious and non-religious. It targeted no religious practice. And were the court to let the school opt out, it would, in Smith’s words, make the school a “law unto itself.”[142] That result was unacceptable to the court, and so it enforced mandatory bargaining.[143]

II. An Unstable Dichotomy: State Jurisdiction over Religious Schools

And so, a half century of litigation has brought us to the unstable status quo. Time and time again, religious schools have beaten back the Board’s efforts to insert itself into their internal affairs. The schools have consistently argued—and consistently convinced federal courts—that the Board’s presence in their hallways cannot be squared with the First Amendment. Yet even as they protected their autonomy on one flank, they saw it overrun on another. No sooner had federal officials retreated than state officials appeared in their place.[144]

That result was surely not what the Supreme Court envisioned when it handed down Catholic Bishop. It is true that the Court framed its holding in terms of federal statutory interpretation. The Court’s holding said nothing explicit about state law or, for that matter, the Constitution.[145] But its rationale implied more than that. Its mode of reasoning—and indeed, its counterintuitive interpretation of the statute—showed that it was doing more than parsing language. Language alone could not have produced an exemption for religious schools. Nowhere did the NLRA exempt those schools; it didn’t even mention them. Instead, it applied to “employers” and “employees.”[146] And when a statute uses broad terms like those, ordinary principles of statutory interpretation warn against reading in exceptions where none appear.[147]

Yet find an exception the Court did. It reasoned that because Congress hadn’t expressly mentioned religious schools, it must have meant to exclude them.[148] In other words, it reversed the normal presumption against implied exceptions. Such an approach turns statutory interpretation on its head, and in most other contexts would have been laughable—another Holy Trinity[149] destined for the historical dustbin. But instead, Catholic Bishop has survived, and it has survived because there were other considerations at play. As the Court spelled out plainly in its opinion, it took pains to read the statute as it did because the more natural reading—one giving the Board jurisdiction over religious schools—would have risked violating the First Amendment. In other words, the Court engaged in “constitutional avoidance.”[150] To treat its decision as merely a statutory one is thus to ignore the major thrust of its reasoning—to deprive it of its central force, its rational glue.

Yet that is just what courts have done in cases involving state jurisdiction over labor relations in religious schools. They have minimized Catholic Bishop by giving it only its literal force, treating it as if it had nothing to say about the Constitution. That is the wrong approach, one that smacks of willful ignorance, or even malicious compliance.

But put that point aside. Even if these courts were right—even if Catholic Bishop had said nothing about the Constitution—the constitutional question would still remain. And it is by now beyond serious debate that the First Amendment applies equally to the federal government and the states.[151] The states are no freer to invade religious autonomy than the Board is.[152] So if states want to regulate the schools, courts must, at a minimum, confront the constitutional question themselves.[153] And their answer should be the same one that produced Catholic Bishop: there is no way, consistent with the First Amendment, to mandate collective bargaining in religious schools.

As the Seventh Circuit recognized, the state’s involvement in mandatory bargaining threatens the First Amendment in two ways: through mandating bargaining itself, and through investigating alleged unfair labor practices.[154] In the former case, the state interferes by requiring the school to share authority over the terms and conditions of teachers’ employment with a third party, even when those terms and conditions potentially raise religious questions. And in the latter, the state interferes by probing the school’s motives for a particular action, even when the school justifies its action on religious grounds. By reading Catholic Bishop narrowly, courts considering state regulations have ignored both of these problems, but they haven’t resolved the constitutional questions.

A. Policing Collective Bargaining

To understand why bargaining interferes with religious autonomy, we first have to understand what bargaining entails. Like the NLRA, most state labor-relations laws require employers to bargain over three topics: wages, hours, and working conditions.[155] The first two topics include a relatively limited universe of subjects. They’re about when employees show up to work and how much employees get paid. The third topic, however, is more expansive. It includes, of course, things central to the employment relationship, such as workloads, promotions, and discipline.[156] But it also includes more attenuated items, such as parking spots and prices in the office cafeteria.[157] Nearly any decision affecting an employee’s work life is fair game.[158]

That can lead to especially expansive bargaining in a school, where nearly all managerial decisions affect a teacher’s work.[159] For example, consider the choice of which courses to offer. If a school decides to offer a wide range of courses, there will be more classes, and so more teaching work. The school can address the additional workload in a few ways: it can hire more permanent staff, hire more adjuncts, or assign more work to its current teachers. Any of these choices will affect the teachers’ experience at work, and so will require bargaining.[160]

But in a religious school, the same choice may also take on a religious character. Assume the school decides to offer a new divinity course. Most people would say that the decision to offer such a course is, on some level, religious. Yet as we’ve just seen, the decision also affects the teachers’ working conditions. So in any other workplace, the decision would be a mandatory bargaining subject.[161] But in the religious workplace, is this a decision about working conditions, or is it about the school’s religious mission?

The answer, of course, is that it’s both—and therein lies the problem. To avoid a conflict with the First Amendment, the state has to avoid inserting itself into religious decisions. And to do that, it has to draw clear lines between subjects affecting religion and subjects affecting working conditions.[162] But in practice, it can’t draw that distinction, because the distinction is illusory. Just as nearly everything a school does affects teachers’ working conditions, nearly everything a religious school does involves religion.[163] The subjects blend together in ways that make them impossible to disentangle.

Consider a few more examples. Suppose a school decides to offer a course in humanist moral theory. Whereas the choice about the divinity course looked religious on its face, this one looks “secular.” The humanist course will involve teachings from outside the church’s doctrines—maybe even antithetical to those doctrines.[164] So a government official might initially react by considering it an appropriate subject of bargaining. But that initial reaction would minimize the potential religious significance of teaching even apparently non-religious ideas. Maybe, in fact, the school wants to teach humanist theory because it sees the theory as compatible with its own beliefs. Or maybe it wants to illustrate a contrast between its own views and those of the secular world. Or maybe one of its central tenets is tolerance of other worldviews. Any of these goals could be characterized as religious. And because the goals are potentially religious, so is the decision to offer the course. You cannot put the course neatly into a “secular” box even when its subject is facially secular.[165] 

You can find the same issue with many common workplace decisions. Dress codes, weekly schedules, codes of conduct—all of these can take on a religious character in some settings. A dress code may carry religious significance when it requires the wearing (or not wearing) of a hijab. A schedule may change its character when it forbids work on the Sabbath. A code of conduct may mix with doctrine when it requires good moral behavior. There is no way to sort these subjects neatly into secular and religious buckets. They are not working conditions or religious matters; they are both.[166]

You might think these examples are outliers, cherry-picked to prove a point. For the moment, let’s assume that’s right. Let’s say that there are actually three categories of potential bargaining subjects: The first includes subjects that are clearly secular, the second those that are clearly religious, and the third those that are a mix of the two. The state cannot order bargaining over the second category because doing so would insert it directly into religious decisionmaking.[167] And the third category will at minimum present the same knotty line-drawing problems we just explored.[168] But couldn’t the state simply put those two aside? Without wading into the difficult line-drawing questions, couldn’t it limit its own authority and order bargaining only over items in the clearly secular category?[169]

While that approach may be tempting, the answer is still no. The problems pile up as soon as you start the analysis. To even create the three categories, a government official has to comb through the school’s practices and label them accordingly.[170] And to do that, the official has to make some initial judgment about their substance.[171] Even a “clearly” religious subject requires her to recognize it as religious, and even a “clearly” secular one requires the opposite judgment. The problem isn’t the ease or accuracy with which the official can make the distinction; it’s that she is making the distinction in the first place.[172] She is telling the school which of its practices are religious and which are not. Even if she is fairhanded, careful, and even correct, she is still evaluating the substance of the school’s beliefs.[173]

The official might try to avoid that problem by deferring to the school. For example, she might order bargaining only on subjects the school itself labels secular. Of course, that might not work, as most institutions don’t make lists of all the secular things they do. So maybe more realistically, the official might require the school to object to bargaining when it sees a subject as religious. And whenever the school objects, the official might take the school at its word and set the subject aside. That approach would require her to make no judgment for herself; the school, not the official, would sort subjects into secular and religious buckets.

But even that solution would be hollow. If the official always defers, she effectively leaves the school in control. The school can decide which subjects are fit for bargaining and which are not. And at that point, we could reasonably ask why the official is involved at all. Schools can already bargain over the things they want to; the whole point of government intervention is to make them bargain over the things they don’t want to discuss.[174]

You might think that the official could solve the problem by deferring only when the school makes a reasonable objection. But of course, to decide what’s reasonable, the official still has to make some decision about the merits. And that kind of decision brings us back to the original problem. Again, government officials have no competence in religious affairs; they cannot evaluate the merits of a religious belief, reasonable or unreasonable.[175] Deference avoids that kind of evaluation only when it is universal. It works only when the official defers every time—in which case it is worthless.

Some courts have looked for a third way around the problem. In St. Teresa, the New Jersey Supreme Court used the Diocese of Camden’s prior agreements as a kind of crib sheet. The court ordered the Diocese to bargain with its elementary-school teachers, but only about subjects contained in a prior agreement with its high-school teachers.[176] The court reasoned that if the Diocese could bargain about those subjects for its high schools, then surely it could do the same for its elementary schools.[177]

Admittedly, the St. Teresa approach has a superficial appeal. After all, why can’t a state require a school to bargain about terms it already agreed to bargain over? The school can hardly complain that those terms are categorically off-limits. It can’t say that its religious beliefs prevent it from discussing the terms with its employees or a union. It is being asked to do only what it has already done. And that, it seems, is about as modest a burden as the school could hope for.

But in fact, as a general policy, the St. Teresa approach is inadequate in almost every way. For one, not every religious school will have a prior agreement to crib from. And even if every school did, St. Teresa would still approach the problem from the wrong direction. The court assumed that mandatory bargaining is a problem only when it interferes with some specific religious practice or belief.[178] But that’s wrong. In many cases, bargaining and belief are completely consistent. The Catholic Church, for one, has vocally supported collective bargaining.[179] No, mandatory bargaining is a problem only because it’s mandatory.[180] The problem comes not from some specific term in a collective-bargaining agreement, but from the government-backed interference the term implies. The command to bargain interferes with a religious school’s autonomy to control its own internal affairs.[181] And it is that autonomy, not some specific religious practice, that the First Amendment protects in this context.  

And let’s be clear: constitutional protection for this kind of autonomy is not a new concept. More than half a century ago, the Supreme Court recognized that the government has no business telling religious institutions how to manage their internal affairs. In Kedroff v. St. Nicolas Cathedral of the Russian Orthodox Church, the Court held that New York could not insert itself into a dispute between the Orthodox Church in Moscow and a North American religious corporation.[182] The dispute concerned control of the St. Nicholas Cathedral. Because the matter related to internal church hierarchy, any attempt by the state to weigh in interfered with the church’s right of self-determination. The First Amendment, the Court said, “radiates . . . a spirit of freedom for religious organizations, an independence from secular control or manipulation—in short, power to decide for themselves, free from state interference, matters of church government as well as those of faith and doctrine.”[183]

This sphere of independence comes from the very nature of a church. Churches embody the religious beliefs of their members.[184] So every time the government interferes in a church’s internal organization, it to some extent interferes with religious practice.[185] That is true regardless of the nature of the interference. Interference occurs when the government tells the church whom to hire, whom to promote, or how to allocate its resources.[186] The problem isn’t that the government is telling the church to act inconsistently with some specific belief; it’s that the government is telling the church how to organize itself at all.

To think about it in another way, imagine if the government ordered a religious school not to bargain with its teachers. The government would still be interfering with internal school affairs. And that kind of interference would be no less unconstitutional than the opposite command.[187] It is the existence, not the substance, of the command that offends the First Amendment.  

It follows, then, that St. Teresa was wrong to conflate mandatory bargaining with the voluntary kind. Without government involvement, a school can bargain about any subject it wants, even overtly religious subjects. It can bargain about the curriculums in divinity courses, qualifications for ministers, or even the admission of nonbelievers. When discussed voluntarily, none of these subjects causes a First Amendment problem. The First Amendment puts no limits on a church’s decisions about its own affairs.[188] The subjects become constitutionally problematic only when the government gets involved.[189] And for that reason, voluntary bargaining is an unreliable guide for mandatory bargaining. The government cannot simply order a religious school to bargain about any subject it has bargained about on its own initiative.[190] The same bargaining subject might be perfectly fine when voluntary, but constitutionally suspect when mandatory.[191]   

The St. Teresa approach, then, offers us no way around the constitutional problem. Any government order to bargain interferes with a school’s autonomy. And that is true whether the order comes from the federal government or a state. 

B. Investigating and Remedying Unfair Labor Practices

No less problematic is the state’s involvement in unfair-labor-practice investigations. Like the federal government, most states run their investigations through an administrative agency.[192] If the agency finds evidence of an unfair labor practice, it brings the case before a hearing officer or a judge.[193] This judge is responsible for weighing evidence and evaluating credibility. She reviews documents, hears testimony, and resolves disputes between differing narratives. These narratives often conflict when they describe why the employer took some action. The employer will offer a business motive, the agency an unlawful one. It’s up to the judge to decide which is true.[194]

To do that, the judge sometimes has to decide whether the employer’s motives are pretextual—whether it made its decision for a reason different from the one it offers.[195] In a normal case, that kind of judgment call causes no constitutional problems. The judge can simply conclude that the employer is lying.[196] But the same judgment presents real problems in cases involving religious schools. Suppose the agency alleges that the school fired an employee for union activity. In response, the school says it fired the employee for violating certain religious tenets. To side with the agency, the judge has to conclude that the school’s justification is pretextual. And to do that, the judge has to decide either that the school’s reason wasn’t sufficient to justify the firing or that the school doesn’t believe its own reasons. Either way, she has to make some determination about the substance of the school’s asserted religious beliefs.[197]

This is rocky constitutional territory. Again, government officials have no competence in religious affairs. The government cannot tell someone what she believes, much less whether her beliefs justify some specific action. And calling her asserted belief “pretext” comes quite close to that.[198]

Recognizing the problem, some courts have looked to “mixed-motives” analysis. That analysis asks whether, even without the asserted religious element, the school would have taken the challenged action anyway. If the school would have acted differently without the religious element, the official leaves things where they lie. But if the school would have done the same thing regardless of the religious element, then the unlawful motive was the true cause, and the official can order the school to reverse itself. So in our example, the judge can take the school at its word; she can accept that the school was motivated, at least in part, by religion. But she can then hypothetically remove that motive and reevaluate the situation. If, without the religious motive, the school would have fired the teacher, the official can put the teacher back to work. She doesn’t have to call the school’s religious beliefs into question.[199]

It’s easy to see why courts are attracted to this kind of solution. Ostensibly, it lets the government have its cake and eat it too. The government can avoid questioning religious beliefs while also remedying unlawful discrimination. The school gets to keep its religious autonomy, and the employee gets her job back.

But of course, nothing is quite that easy in real workplaces. Let’s assume now that the school suspends a biology teacher for a semester. The school says it suspended her because she taught a theory of evolution inconsistent with the school’s religious beliefs. The agency, by contrast, alleges that the school suspended the teacher because she attended a union meeting. Under a mixed-motives approach, a judge can accept both motives as true.[200] The school may have been upset by the teacher’s course materials, but also by her union activities. The judge would then apply mixed-motives analysis to decide whether the school would have suspended the teacher even if she hadn’t attended the meeting.[201] That is, the judge still has to decide whether the discussion of evolution was important enough to justify the suspension on its own. That means the judge still has to weigh the school’s religious motivations; she still has to make some judgment about the strength of the school’s beliefs.[202] Mixed-motives analysis can’t get us around this problem.[203]

In fact, the example we just considered offers an unusually clean scenario. In most real scenarios, the motivations won’t be so easy to segregate. For example, suppose a Catholic school decides not to renew the contracts of several teachers. It does that, it says, because the teachers engaged in “un-Christian” behavior: they went out on strike. How is the judge to apply a mixed-motives analysis here? The alleged religious and unlawful motives are not discrete; they are the same. The strike was protected, but also, according to the school, “un-Christian.” So to reverse the school’s action and put the teachers back to work, the judge has to conclude either that the school is being disingenuous or that labor law overrides the religious concern. In other words, the judge has to balance the school’s religious beliefs against government policy.[204]

It’s tempting to dismiss this scenario as unlikely, even fanciful. But we know it happens in real schools. In fact, it was exactly the scenario presented in Nyserb.[205] The Nyserb court dealt with it by endorsing a mixed-motives analysis.[206] But as we now see, that solution was too facile. It failed to recognize, much less resolve, the conflict between religious beliefs and union activity. Mixed-motives analysis couldn’t resolve that conflict because the school’s motives were never really mixed. There was only one motive, both religious and prohibited.[207]

Now, we could imagine a rule resolving the conflict by prioritizing legal compliance over religious exercise. Indeed, such a rule prevails in most of American life. Under Smith, neutral and generally applicable laws often override religious practices.[208] And the same rule could play out in the halls of religious institutions, including religious schools. General laws could govern the internal affairs of those institutions just as they govern the affairs of so many others. Such a rule might even fit better with neutrality-centered views of the First Amendment, which tend to prioritize equal treatment over accommodation.[209] 

But for better or worse, that has never been the rule when it comes to a religious institution’s internal affairs. The Supreme Court has long recognized that at least in their internal governance, religious institutions enjoy a sphere of autonomy unlike anything enjoyed by the public at large.[210] State courts have consistently missed this distinction.[211] Yes, Smith allows some types of interference with religion. But not all interference is the same. And when it comes to interference with internal institutional autonomy, Smith has almost nothing to say. 

C. Three Types of Interference: The Irrelevancy of Smith

Decided in 1990, Smith revolutionized free exercise jurisprudence. For decades, the Court had analyzed laws burdening the free exercise of religion under a compelling-interest standard. That is, whenever a state burdened religious exercise, it had to provide a sufficiently compelling reason for doing so.[212] But that standard had drawn withering criticism. Many, including some of the Justices, thought it offered too little guidance to lower courts and state officials.[213] There was no way for these officials to decide, objectively, whether a particular interest was sufficiently compelling. And without guidance, they were left to their own devices; they were free to decide how important the government’s interests were based on their own intuitions.[214] The Court took those criticisms to heart and, in Smith, discarded the compelling-interest approach.[215] It instead announced that it would uphold laws burdening religious exercise as long as they were neutral and generally applicable.[216] Discriminatory laws would fail that test, but most others would pass.[217]

This revolution came just as states were considering whether to extend their labor laws to religious schools. When these laws were challenged, courts in Minnesota, New Jersey, and New York all looked to Smith to help them resolve the constitutional question.[218] And in each case, they upheld the laws. The laws, they reasoned, were neutral and generally applicable. They applied to religious and non-religious employers alike. They singled out no religious practice or belief. And their purpose was self-evidently secular: they promoted collective bargaining to improve the wages and working conditions of all employees. As a result, they passed muster under Smith, and whatever incidental interference they caused was of no constitutional significance.[219] 

This analysis, however, elided a distinction between different kinds of interference. In a classic article on employment law in religious schools, Towards a General Theory of the Religion Clauses: The Case of Church Labor Relations and the Right to Church Autonomy, Professor Douglas Laycock divided interference with religion into three categories.[220] The first was government interference with religious belief: the government tells the believer what he or she can or cannot think.[221] The second category was state interference with specific religious practices: the government tells the believer she cannot sacrifice animals, cannot use certain drugs, cannot dodge the draft, etc.[222] The third was state interference in the operation of religious institutions: the government tells believers how to administer the entities through which they practice their faith.[223]

The first type of interference is rare in this country. We seldom see examples of the government proscribing beliefs or dictating matters of faith.[224] The second, however, occurs almost daily. The government tells people when and where they can gather, what substances they can consume, whom they can marry. It was this type of interference that the Court dealt with in Smith. There, the Court held that the state could deny unemployment benefits to believers who lost their jobs for smoking peyote, even though peyote was part of their religious faith.[225] State law thus clashed with a specific religious practice.[226]

Catholic Bishop, however, involved the third kind of interference.[227] The schools never argued that collective bargaining itself violated any particular religious practice or tenet. Indeed, the Catholic Church enthusiastically supported collective bargaining.[228] Instead, the schools objected to the state’s interference in their internal affairs. By commanding them to bargain over conditions of employment, the state sapped their authority over their internal governance. In other words, it wasn’t bargaining that violated the schools’ rights; it was the government’s command to bargain.[229]

Smith, then, has little to say about whether a state can dictate the terms of a religious school’s relationship with its employees. Smith dealt with a different kind of interference—state interference with individuals’ religious practice. It never suggested that a state could insert itself into a religious institution’s internal administration, even if the state did so in a neutral and generally applicable way. Smith tells us, in short, almost nothing about the debate over mandatory collective bargaining in religious schools.

This conclusion will, no doubt, raise some eyebrows. After all, if Smith doesn’t allow states to regulate religious schools, does anything? Surely the school must comport itself according to normal commercial and regulatory laws. It must, for example, pay its vendors on time, comply with local zoning laws, and observe general building codes.[230] We cannot let a school flout those laws simply because it associates with a religious institution. So some will ask: Can a state do anything to rein in a religious school, or is the school “a law unto itself”?[231]

But so stated, the question presents a false choice. Not even churches claim that they can ignore all laws simply by virtue of their religious affiliation. Society can—and does—recognize a church’s general duty to comply with the law while still respecting its sphere of internal autonomy.[232] The real question, then, is where autonomy ends and general obligation begins.

To draw the line, we have to distinguish between a church’s behavior toward those outside its community and its behavior toward those within it. When the church complies with contracts, zoning laws, and building codes, it is acting externally: it is operating in the market just like any other person, business, or other entity.[233] But when it acts internally, its governance is its own, and the members of its community voluntarily submit to its authority.[234] That is no less true of employees than it is of congregants. Like congregants, employees in a religious school voluntarily join the community and accept the church’s leadership.[235] They are no longer pure outsiders dealing with the school at arm’s length, as a member of the public might.[236] They have taken up a role—a vital one—in the school’s religious mission.[237]

This last point is, of course, not uncontroversial. There are no doubt some who think teachers are more like vendors than congregants, more like arm’s-length contracting parties than members of the religious community. But that view overlooks the teachers’ role in carrying out a school’s religious mission.[238] Religious schools exist only when a community decides to offer an alternative to secular education.[239] Religious schools, then, owe their existence to a community’s desire to project its religious message, and in particular, to hand that message down to the next generation.[240] The community’s primary agents in that mission are its teachers. Teachers stand at the front lines, speaking for the whole group. They may not always embrace the community’s teachings, but they do serve as its voice.[241]

That special role has been recognized for decades. For example, in Lemon v. Kurtzman, the Supreme Court held that a state could not subsidize teacher salaries at religious schools even when the teachers taught only secular subjects.[242] The Court reasoned that even lay teachers would inevitably be affected by the school’s religious mission and character.[243] The teachers were products of their environment, and the state could not legitimately expect them to expel religion from their classrooms.[244] That is, they were religious agents even when teaching subjects other than religion.

Similarly, in a pair of recent cases, Hosanna-Tabor Evangelical Lutheran Church & School v. EEOC[245] and Our Lady of Guadalupe School v. Morrissey-Berru,[246] the Court held that the First Amendment protects religious institutions from interference with their relationship with “ministerial” employees—i.e., employees who play important religious roles. Both cases involved attempts to apply antidiscrimination laws to teachers in religious schools. In rejecting those attempts, the Court reemphasized that religious schools enjoy a sphere of autonomy over their internal affairs, including their relationships with their teachers.[247] Laws regulating those relationships sapped the schools of their internal authority and entangled the state in school administration: “When a school with a religious mission entrusts a teacher with the responsibility of educating and forming students in the faith, judicial intervention into disputes between the school and the teacher threatens the school’s independence in a way that the First Amendment does not allow.”[248]

Teachers, then, are more than just ordinary employees. They are part of the school’s internal religious community. In fact, they are often the school’s most important agents in its religious mission. Their relationship with the school is a matter of internal governance, over which the school enjoys constitutionally protected autonomy.[249]

That being the case, when a state regulates the teachers’ relationship with a school, it necessarily interferes with the school’s internal authority.[250] And that is true even when the regulation is neutral and generally applicable, and even when these laws interfere with no specific religious practice. Again, Hosanna-Tabor and Our Lady of Guadalupe offer prime examples. There, the schools never argued that their religious practices required them to discriminate on the basis of some protected characteristic. No one claimed that antidiscrimination laws failed the Smith test. Instead, the only question was whether the state could apply neutral, generally applicable employment laws to the schools’ internal affairs. The answer was no.[251] That was the answer not because the schools had a First Amendment right to discriminate, but because they had a First Amendment freedom to manage their own internal relationships.[252]

The same, then, must be true for mandatory-bargaining laws. Those laws interfere with a school’s autonomy at least as much as antidiscrimination laws, probably more. Antidiscrimination laws have only a moderate effect on management’s decisionmaking: they limit the bases on which management can make certain employment decisions, but still leave those decisions in management’s hands. Bargaining laws, by contrast, limit management’s authority across an array of subjects. They require bargaining over every term and condition of employment.[253] And again, when it comes to teachers, those terms and conditions encompass nearly everything the school does.[254] Class sizes, course offerings, curriculums—they all affect teachers’ work environments, and so are proper subjects for bargaining.[255] Mandatory bargaining thus represents a far greater loss of autonomy for religious schools.[256]

If Smith meant to limit that longstanding sphere of autonomy, you might have expected the Court to at least mention it. But it never did. Nor has the Court suggested at any point since that Smith gave the state an entryway into church administration. To the contrary, the Court has affirmed and reaffirmed the importance of church autonomy over internal affairs, including employment relationships. On that subject, Smith has nothing to teach us.

III. Conclusion: All Roads Lead to Consistency

As we’ve now seen at length, a dichotomy persists in the law governing religious schools. While the courts have recognized that the First Amendment denies the Board jurisdiction over church-run schools, they have failed to apply that same rule to state agencies. Worse, they have done so without making any serious effort to explain the difference. Instead, they have swept Catholic Bishop into a jurisdictional corner, dismissing it as a decision only about statutory interpretation. And they have justified the states’ own actions with logic that fails to address Catholic Bishop’s core concern: protecting the autonomy of religious schools over their internal affairs, as required by the First Amendment.

There is no way to square Catholic Bishop with that result. Nor is there any way to justify the distinction based on the Court’s later precedents. The First Amendment protects religious schools’ autonomy over their relationships with their employees, and that protection extends just as much to state agencies as it does to the Board. There is no constitutionally coherent way to deny the Board jurisdiction over the schools while allowing it to the states.

To paraphrase Abraham Lincoln, the law divided cannot stand. It must become all one thing, or all the other. Ideally, states would recognize the illogic of the existing divide and withdraw on their own accord. But more likely, the courts will have to make them. Courts will have to recognize the dichotomy and order states to stand down. With such an obvious imbalance, we might expect that decision to come sooner rather than later. We are now entering our fifth decade since Catholic Bishop, and the Supreme Court appears more solicitous of religious autonomy than ever. If ever there were a time to give Catholic Bishop its full force, it is now.

Note from the Editor: The Federalist Society takes no positions on particular legal and public policy matters. Any expressions of opinion are those of the author. Whenever we publish an article that advocates for a particular position, we offer links to other perspectives on the issue. We also invite responses from our readers. To join the debate, please email us at [email protected].