Updated at 11:15 a.m. ET on February 8, 2021.
Over the past few years, the Supreme Court has been sketching the outline of a broad compromise on LGBTQ rights. Civil-rights protections will shield people from discrimination on the basis of sexual orientation and gender identity. At the same time, religious objectors will have their own set of robust rights. For example, the Court recently clarified that Title VII, the federal antidiscrimination employment law, covers LGBTQ employees.* The Court also, in another case, found in favor of a baker who would not provide a wedding cake to a same-sex couple. This term, the Court is considering granting another religious exemption from civil-rights law, this time for an adoption-licensing agency in Philadelphia that refuses to consider same-sex couples as prospective parents.
Is this compromise stable? Supporters say yes: Religious objectors, they argue, are a negligible minority in a society growing ever more affirming of LGBTQ equality. Exempting them, the thinking goes, will not expand discrimination against same-sex couples. But critics worry that the Court’s religious exemptions normalize discrimination and thereby encourage it.
The trouble is that all of this is conjecture. There have been almost no data to clarify whether religious exemptions really do increase discrimination. Particularly missing are data on the effects of religious exemptions granted in Supreme Court decisions.
In a forthcoming journal article, I set out to fill that hole. I looked at this question through Masterpiece Cakeshop, the case involving a devout baker who declined to create a custom wedding cake for a same-sex couple. In that decision, the Court handed down a 7–2 decision favoring the baker. This was a narrow ruling that did not settle the big constitutional questions at hand. Yet it received extensive media coverage, and conservative groups hailed Masterpiece as a significant victory that confirmed the rights of people of faith to pursue their beliefs whether at home or at work.
In May 2018, anticipating that the Court would rule in favor of the baker, I began a field experiment that measured the willingness of wedding vendors to provide services to same-sex and opposite-sex couples. The wedding vendors in the study—1,155 photographers, bakers, and florists—were sampled in areas of the United States that differ in their approach to religion-equality conflicts. (Across the nation, some states or cities prohibit businesses from discriminating on the basis of sexual orientation, whereas others do not, and some states facilitate religious exemptions via Religious Freedom Restoration Acts, while others do not.) I then sent vendors queries purporting to be from potential clients, such that each vendor received queries from both same-sex and opposite-sex couples, first in the run-up to the Court’s decision and then in the two weeks following Masterpiece. (In total, I contacted each vendor four times. Twice before—once as a same-sex couple, once as an opposite-sex couple—and twice after, with the same breakdown.) I measured whether vendors’ willingness to do business with the couples changed after the decision was rendered. To attempt to isolate the causal effect of Masterpiece from other events that could have influenced the vendors, all communications were conducted in this short time span.
What were the real-world consequences of Masterpiece? In short, the decision seems to have exposed same-sex couples to heightened risk of discrimination in the organization of their weddings. First, my field experiment revealed that Masterpiece appears to have generally reduced vendors’ willingness to provide wedding services to same-sex couples. Whereas before Masterpiece, same-sex couples had received positive responses from 64 percent of vendors, after Masterpiece only 49 percent of all vendors responded positively. It’s important to note that opposite-sex couples do not receive uniformly positive responses to their queries, because not all vendors provide tip-top service and some may be genuinely busy. But same-sex couples seemed to fare worse after Masterpiece. Focusing on those businesses that prior to the decision were willing to serve same-sex couples, I found that 75.5 percent responded positively to opposite-sex couples after Masterpiece, whereas only 66 percent of these previously gay-friendly businesses responded positively to same-sex couples after the decision. The effect did not grow or shrink with the conservativeness of the county, and it was just as strong in big cities (despite the common belief that discrimination against LGBTQ people doesn’t occur very much in metropolitan areas). An identical effect was found in a “control group” of businesses that were contacted for the first time after Masterpiece. However, the discriminatory effect of the decision was significantly more pronounced in counties with relatively more religious congregations per capita.
The negative Masterpiece effect is likely to add up to a meaningful increase in discrimination experienced by LGBTQ people. Couples of all identities typically contract with about 10 types of vendors in the process of organizing their wedding (e.g., reception venues, wedding planners, bakers, florists, photographers, videographers, bridal/groom salons, jewelers, DJs, and calligraphers). Many inquire with several vendors from each category, often amounting to 15 to 20 encounters. Taking these factors into account, I estimate that about three out of every four wedding-planning LGBTQ couples will experience discrimination they would not otherwise have encountered, had it not been for the Masterpiece decision. Of course, this risk can vary depending on the number of vendors a couple encounters.
These conclusions have several implications for the debate on religious exemptions. First, they discredit the argument that the effect of religious exemptions is negligible and that exemptions will not increase discrimination. Second, the results complicate the conventional portrait of religious objection as fixed and unyielding to change. Instead, the experiment finds that people’s behavior is influenced by signals from the government; if the government creates exemptions, people who had not discriminated before might start doing so.
Third, the finding that exemptions increase discrimination has legal implications. Under the most demanding legal standard, the government must justify laws that substantially burden religion, by showing that the laws are the least restrictive means to achieve a compelling state interest. If religious exemptions increase discrimination, as I have found, then enforcing antidiscrimination laws without exception may be the best way to promote equality, and perhaps the only way. (I discuss some potential ways forward in a forthcoming article in the Harvard Civil Rights–Civil Liberties Law Review.) Granting exemptions frustrates the government’s goal of ending discrimination. Indeed, Supreme Court justices on the right and on the left have always considered the consequences of exemptions when deciding whether to grant or deny them.
These findings should prompt the Supreme Court to proceed with great care as it sets to deciding the Philadelphia case and any future religion-equality conflicts. Undoubtedly, the Court faces an acute dilemma: Both equality before the law and religious liberty are fundamental constitutional rights, and setting their respective boundaries is no simple task. Yet however the Court decides to resolve the constitutional issues at hand, it must take into account that even a narrow and case-specific decision might have a significant detrimental effect on the broader population that stands to lose from the exemption—and it is the Court’s duty to avoid causing this harm.
*This article previously misstated which law is the federal discrimination employment law. It is Title VII, not Title IX.