Symposium: Title VII did not and does not extend to sexual orientation or gender identity — in 1964 or today
Richard Epstein is the inaugural Laurence A. Tisch Professor of Law at NYU School of Law, the Peter and Kirsten Bedford Senior Fellow at the Hoover Institution, and the James Parker Hall Distinguished Service Professor of Law Emeritus and Senior Lecturer at the University of Chicago.
Political and legal discourse has changed much since 1964, when sex was added, almost as an afterthought, to the list of forbidden grounds of discrimination under Title VII of the Civil Rights Act of 1964. What seemed risqué in 1964 seems positively old guard in 2019. So one looming battle of the current Supreme Court term is whether Title VII covers the sexual-orientation claims raised in Bostock v. Clayton County, Georgia and the gender-identity claims raised in R.G. & G.R. Harris Funeral Homes Inc. v. EEOC. The short answer is: It doesn’t.
These two cases offer a stunning contrast in styles of judicial thought. Bostock affirms in a few short sentences a 1979 precedent that “discharge for homosexuality is not prohibited by Title VII.” The use of the term “homosexuality” dates the opinion. Harris, on the other hand, takes 35 dense pages to establish, first, that gender identity is a protected category under Title VII, and second, that the defendant’s religious practices and beliefs provide no constitutional refuge from Title VII under the free exercise clause.
Note the differences. Bostock takes stare decisis seriously. It is not willing to read any change in social mores into Title VII. Its implicit subtext is that only Congress, not the courts, should make fundamental changes in a venerable 55-year old statute. Harris takes the opposite judicial stance. Now, the longstanding Title VII rule is a liability, not an asset. The opinion’s constant use of such voguish terms as “stereotypical behavior” shows its real impatience with obsolete legal norms. Harris eagerly updates older statutes to reflect modern sensibilities.
One year ago, the odds were substantial that the Supreme Court would have followed the approach taken in Harris. The operative Supreme Court view on interpretation was encapsulated in Auer v. Robbins, a 1997 case that showed extreme deference to a federal regulator in the construction of a federal statute. In an opinion that he later came to regret, Justice Antonin Scalia held that deference was the norm, so that the latest interpretation of a disputed position trumped all prior ones. Auer migrated into areas of civil rights in Gloucester County School Board v. G.G., in which the U.S. Court of Appeals for the 4th Circuit performed conceptual somersaults under Title IX to uphold the Office of Civil Rights’ conclusion that educational facilities should “generally treat transgender students consistent with their gender identity.”
The temporal differences between Bostock and Harris reflect a deep difference in how to think about administrative — and indeed all — law. There is abundant evidence, as Professor Aditya Bamzai has shown, that the deference afforded to administrative agencies during the 19th and early 20th centuries was a deference to a consistent practice on such personnel issues as promotion and retirement benefits for government employees. This regnant theory tied in closely to the law of contract, which uses just these tools to construe open terms in various government contracts. On this view, it is the recent case — Harris — that deviates from the earlier practice, which makes it an apt target for a Supreme Court reversal. After all, judges interpret, not make, law.
The modern constitutional law under Chevron v. Natural Resources Defense Council, Inc. and National Cable & Telecommunications Association v. Brand X Internet Services reverses the flow of deference by privileging the most recent deviation from any consistent past practice, just as in Harris. The ability for trendy interpretations to remake the substantive law is evident in this move, and the question is whether the presentist version of administrative law can survive the all-critical recent Supreme Court decision in Kisor v.Wilkie, in which Justice Elena Kagan’s heroic efforts to salvage Auer from extinction resulted in a set of strategic concessions that should shipwreck today’s effort to expand Title VII. Her key passage reads:
First and foremost, a court should not afford Auer deference unless, after exhausting all the “traditional tools” of construction [citing Chevron], the regulation is genuinely ambiguous. A court must carefully consider the text, structure, history, and purpose of a regulation before resorting to deference. If genuine ambiguity remains, the agency’s reading must still fall “within the bounds of reasonable interpretation [citing Arlington v. FCC].”
But Auer stood for no such proposition. It sustained the narrow reading of the statutory exemption from the overtime provision of the Fair Labor Standards Act for “bona fide executive, administrative, or professional” employees, without once bothering to look at the text, structure, history or purpose of the FLSA. So just what is it that happens if we look at any of these issues in Bostock and Harris?
The basic statutory provision prevents discrimination “on the basis of race, color, religion, sex, or national origin.” The case of sex has always proved trickier than, say, race. There are so many legitimate reasons to discriminate on the basis of sex that Title VII acknowledges “bona fide occupational qualification[s]” (BFOQs) to limit the scope of that prohibition. And “good faith” did not impose strict scrutiny on the scope of the exception, at least in 1964. So actresses had no right to play male roles and men could be prevented from selling lingerie. BFOQs were part of a large tradition that allowed for sex differences in dormitories, athletics and the military. There is not a single word in the 1964 Civil Rights Act, or in its legislative history or in contemporary social commentary, that cuts against the then-conventional wisdom that sex refers to the dichotomous division between men and women. The 1964 Act banned the use of “help wanted female” for secretaries or nurses.
The same conclusion applies, only more so, when looking at the attempt to read sex as sexual orientation. At the time of the statute, it was widely accepted that the state had the power to criminalize homosexual conduct, which it often did. In 1986, Bowers v. Hardwick read the history correctly when it accepted the old order, all to massive protests. But it becomes odd in the extreme to posit that sexual orientation both could have subjected someone to criminal punishment and simultaneously could have formed a protected category under Title VII.
The situation with gender identity is even clearer. In the context of Title VII there was no discussion of that topic at all in 1964. The issue jumped from nowhere onto center stage within the last decade or so—40 years too late to support the statutory claim in Harris. It cannot be that self-identification matters for sex, but is irrelevant for race or ethnicity. The statute just does not parse if identity claims carry over to race, color, religion or national origin.
I have little doubt that some form of statutory protection for gender identity will work itself into the law. But I am hardly sanguine that it will do so in a sensible way. The ease with which Harris condemns these sex stereotypes is deeply troublesome. The word “stereotype” is plagued with deep ambiguity because the conventional definition speaks of “a widely held but fixed and oversimplified image or idea of a particular type of person or thing.” To say “oversimplified” suggests that these hasty generalizations are wrong. But many so-called stereotypes are true generalizations, whether they deal with height or some psychological attribute of men and women. It is not that all men are taller than all women. It is true that the two distributions have different patterns, so that it is possible to set up a one-to-one correspondence by which each man is paired with a shorter woman.
The great danger today is that the state will find discrimination where there is none, because of its refusal to take into account these systematic differences. That same difficulty can apply to social attitudes. A funeral home is a sensitive locale, and it surely has some interest in the way in which its “public facing” employees appear to its anxious patrons coping with the loss of a loved one. It is fine for an appellate court to write coolly and acontextually that “the district court correctly determined that Stephens [the transgender worker] was fired because of her failure to conform to sex stereotypes.” But that court could not care less about the impact that Stephens’ dress could have on customers, and on the business success of Harris Funeral Homes. In this case, the Supreme Court should not put a match to the kindling, because no defensible mode of statutory construction, post Kisor, supports the claim of either plaintiff.