It is easy to articulate the core principle behind the First Amendment right to free speech: The government can’t restrict what you say or make you say what the government wants without a good reason. Ah, but how good a reason and what kind of reason? That is where it gets more complicated. How do we distinguish, for example, between a law that requires every man, woman, and child to declare “All hail the mighty New England Patriots” in public every day and a law that requires certain government-approved warnings with prescription drugs.
Both represent the government compelling speech, but one passes Constitutional muster in its current form while the other clearly would not. (Hint: The unconstitutional one is the hypothetical law compelling Patriot worship, which would be a bad idea with or without the First Amendment and might even run afoul of the prohibition on cruel and unusual punishment.) Whether government regulation of speech violates the constitution depends on the nature of the regulation and the standard under which courts review the law, and all of this is put on display in the Ninth Circuit’s recent en banc opinion striking down San Francisco’s sugar-sweetened beverage warning law. The case is American Beverage Ass’n v. City and County of San Francisco, — F.3d —, No. 16-16072, 2019 WL 387114 (9th Cir. Jan. 31, 2019), and while the entire eleven-judge panel agreed that San Francisco’s law violated the First Amendment, it filed four different opinions explaining why.
As far as public health policy goes, San Francisco’s beverage warning law was a clumsy effort. It required that advertisements for sugar-sweetened beverages include the following warning in a size no smaller than 20 percent of the ad and set off in a rectangular box—a sort-of “black box” warning:
WARNING: Drinking beverages with added sugar(s) contributes to obesity, diabetes, and tooth decay. This is a message from the City and County of San Francisco.
Id. at *1. The law applied to some sugary drinks—but not all. It applied to some advertisements—but not all. And, importantly, the warning was not factually true—sugary drinks do not contribute to Type 1 diabetes. We are no fans of sugary drinks, but we can’t help but think that the law has a certain arbitrary feel to it.
So maybe the law could have been written better and maybe it represented debatable public policy. But was it unconstitutional? The majority decided that it was, and it came to that conclusion after applying a form of “rational basis” review and rejecting the application of heightened scrutiny. That is to say, the government did not have to demonstrate that its regulation of commercial speech was “narrowly tailored” to a compelling governmental interest or that it met any other version of heightened scrutiny—strict, intermediate, or otherwise. Id. at *4-*5. Referring to Supreme Court precedent, the majority set the standard as follows:
[B]efore NIFLA [National Institute of Family & Life Advocates v. Becerra, 138 S. Ct. 2361 (2018)], we examined a similar health and safety warning and held squarely that Zauderer [v. Office of Disciplinary Counsel, 471 U.S. 626 (1985)] provides the proper analytical framework for considering required warnings on commercial products: “[T]he government may compel truthful disclosure in commercial speech as long as the compelled disclosure is ‘reasonably related’ to a substantial governmental interest.” CTIA, 854 F.3d at 1115-17. We rejected the argument that intermediate scrutiny—as required by Central Hudson, 447 U.S. 557, for situations in which speech is restricted or prohibited—should govern. We also rejected the argument that Zauderer applies only to situations in which the government requires disclosures to prevent consumer deception . . . .
Id. at *4 (Emphasis added). Let’s unpack that a little bit. At least when it comes to health and safety warnings, the Ninth Circuit held that the government can compel truthful disclosure so long as the disclosure is reasonably related to a substantial governmental objective. That situation is to be distinguished from laws restricting or prohibiting commercial speech, to which intermediate scrutiny under Central Hudson would apply. Then there is strict scrutiny, which should apply to certain other regulation of speech.
When it came to San Francisco’s beverage law, the Ninth Circuit majority applied the Zauderer three-part test to determine whether the law survived rational basis review: Whether the notice is (1) purely factual, (2) noncontroversial, and (3) not unjustified or unduly burdensome. Id. at *4. The law fell on the third requirement:
On this record, the 20% [size] requirement is not justified when balanced against its likely burden on protected speech. [¶] In addition, . . . [San Francisco has] not shown that the contrasting rectangular border containing a warning that covers 20% of the advertisement does not “drown[ ] out” Plaintiffs’ messages and “effectively rule[ ] out the possibility of having [an advertisement] in the first place. . . .
The required warnings therefore offend Plaintiffs’ First Amendment rights by chilling protected speech.
Id. at *5. We have no strong opinion on this particular law. We are, however, surprised that the Ninth Circuit did not apply a heightened level of scrutiny. The result would have been the same, but the court could justifiably have set the bar higher for laws that compel commercial speech.
That was the opinion of one judge who concurred in the result, but dissented “from most of the reasoning.” Id. at *6-*10. In this judge’s view, the San Francisco beverage law was a “content-based regulation of speech, which is subject to heightened scrutiny under the First Amendment unless the Zauderer exception applies.” Id. at *6. In other words, strict scrutiny applies unless the compelled notice is purely factual, noncontroversial, and not unjustified or unduly burdensome. Id. at *7, and see supra. Here, the factual accuracy of the required warning was disputed, and with regard to Type 1 diabetes the warning was literally false. Id. at 8.
Moreover, the law was anything but “noncontroversial.” To the contrary, “the warning here requires the advertisers to convey San Francisco’s one-sided policy views about sugar-sweetened beverages.” Id. Recall that the requirement that the compelled speech be “noncontroversial” comes from very recent Supreme Court authority, a 2018 opinion addressing mandated government messaging at prenatal clinics. See NIFLA, 138 S. Ct. 2361. If any “controversial” or “inaccurate” government-compelled message triggers heightened scrutiny, then why would that not apply to government-mandated drug warnings that similarly lack scientific support? Food for thought. Finally, requiring commercial speakers to “fight a government-scripted message that drowns out their own advertisements is unduly burdensome.” Id. Under these circumstances, heightened scrutiny should apply, and “[b]ecause the warning requirement is not narrowly drawn” to a substantial state interest, “it does not survive even intermediate scrutiny.” Id. at *9.
Two other judges wrote separately to note that they would find the ordinance unconstitutional on the sole basis that it was not purely factual, which they believed to be a threshold question. Yet another judge wrote that Central Hudson controlled and that “the Supreme Court held that regulation of commercial speech is evaluated under an intermediate scrutiny standard.” Id. at *13. Although this final opinion did not garner a majority, it cut to the core of the dispute with this conclusion: “I share . . . concerns that our current case law [applying rational basis review] will lead to a ‘proliferation of warnings and disclosures compelled by local municipal authorities’ that have ‘only a tenuous link to a ‘more than trivial government interest.’” Id. at *13 (internal citations omitted). This captures the classic First Amendment question: How good a reason and what kind of reason does the government need to restrict what you say or compel you to say what the government wants? The majority answered that the government can compel truthful disclosure that is reasonably related to a substantial governmental interest. A minority would have held the government to a greater level of scrutiny.
We have a feeling we have not heard the last of this, particularly as to what is, or is not “purely factual” and “noncontroversial,” and particularly as to the level of scrutiny required to satisfy the First Amendment test. In many drug and medical device cases, plaintiffs are positing liability on theories that defendants should have warned (thus seeking to compel speech in the form of warnings) about risks or other information that the FDA has concluded are not scientifically based. If scientific basis, or its absence, equates to either the “purely factual” or “noncontroversial” elements of an emerging First Amendment test, then the basis exists for a new constitutional defense in a significant number of prescription medical product liability actions. And as to California, requiring governmentally compelled warnings to have sufficient scientific basis as to be “noncontroversial” has obvious implications for that state’s Proposition 65.