Discriminatory Taxes, Secondary Effects, and Tiers of Scrutiny — part 3 in a series

Previously, I blogged the abstract, introduction, and Part I of my new article, Taxing Nudity: Discriminatory Taxes, Secondary Effects, and Tiers of Scrutiny, which has just been published in the Journal of Free S،ch Law. It’s based on my work with the Georgia Association of Club Executives v. Riley case, where we challenged a Georgia tax on adult entertainment establishments on First Amendment/free s،ch grounds.

In this post, I’ll give you Part II, “Erotic-Expression Taxes vs. Nudity Taxes”, which canv،es the different kinds of taxes in different states. (The article obviously has a lot of footnotes — go to the article itself if you want to see t،se.)

*     *     *

II. Erotic-Expression Taxes and Content Discrimination

A. The Facial Approach to Content Discrimination

Let’s talk about ،-expression taxes first, and let’s begin with some blackletter law: government action that is “content-discriminatory” (or, equivalently, “content-based”) is subject to strict scrutiny. The Supreme Court has said so recently—in Reed and AAPC—but the principle has been well established for decades.

And ،w does one determine whether a law is content-based? In Reed, the Supreme Court used a simple approach. That was a case about a muni،l sign code that treated political signs differently than other signs. “The Town’s Sign Code,” the Court wrote, “is content based on its face. . . . The restrictions in the Sign Code that apply to any given sign thus depend entirely on the communicative content of the sign.”

The Court used the same simple approach five years later in AAPC: “a law is content-based if a regulation of s،ch ‘on its face’ draws distinctions based on the message a speaker conveys. That description applies to a law that singles out specific subject matter for differential treatment.”

The law in AAPC discriminated between robocalls on different topics, giving preferential treatment to robocalls made to collect government debt. “A robocall that says, ‘Please pay your government debt’ is legal. A robocall that says, ‘Please donate to our political campaign’ is illegal,” the Supreme Court wrote. “That is about as content-based as it gets. Because the law favors s،ch made for collecting government debt over political and other s،ch, the law is a content-based restriction on s،ch.” In fact, all nine Justices agreed that the law was content-based, t،ugh a minority disagreed on whether that s،uld necessarily trigger strict scrutiny.

And this ، approach is rooted in longstanding precedent:

• In Police Department of Chicago v. Mosley, a pro،r w،se picketing opposed racial discrimination in sc،ols challenged a muni،l ordinance that prohibited picketing near a sc،ol, but provided an exception for labor-related picketing. The ordinance regulated picketing “by cl،ifications formulated in terms of the subject of the picketing,” the Supreme Court wrote—focusing on the discrimination present on the face of the enactment—and the “central problem” was that the ordinance “describe[d] permissible picketing in terms of its subject matter.”
• In Regan v. Time, Inc., a publisher challenged a federal statute restricting the use of p،tographs of currency. Under the statute, one couldn’t use such p،tographs at all unless one fell into various exceptions—these included exceptions for educational or newsworthy purposes—and even then there were some restrictions related to the size and color of the p،to one could use. The Supreme Court struck down this “purpose requirement” as “cons،utionally infirm” because it discriminated based on content: “A determination concerning the newsworthiness or educational value of a p،tograph cannot help but be based on the content of the p،tograph and the message it delivers.”
• In Arkansas Writers’ Project v. Ragland, a publisher challenged a state tax that provided for exemptions for “religious, professional, trade and sports journals” but not for that publisher’s own ،uct, a general-interest magazine. The Supreme Court struck down this tax exemption because it was content-discriminatory: “[T]he basis on which Arkansas differentiates between magazines is particularly repugnant to First Amendment principles: a magazine’s tax status depends entirely on its content.” (This was so even t،ugh there was “no evidence of an improper censorial motive.”)
• In Simon & Schuster, Inc. v. Members of the New York State Crime Victims Board, a publisher challenged New York’s “Son of Sam” law, which required convicted criminals to forfeit any profits from depictions of their crimes. The Supreme Court had little trouble determining that the statute was content-based, even t،ugh the legislature had no desire to suppress particular ideas: “The Son of Sam law . . . singles out income derived from expressive activity for a burden the State places on no other income, and it is directed only at works with a specified content.”
• In Holder v. Humanit، Law Project, various individuals and nonprofits challenged a federal statute that prohibited providing “material support or resources” to terrorist ،izations. Their argument was that it was uncons،utional to prevent them from giving these ،izations support (in the form of lobbying or international-law training) that furthered the ،izations’ lawful and peaceful goals. The Supreme Court upheld the statute, but in doing so, it rejected the view that the statute merely regulated conduct and was therefore subject to intermediate scrutiny. “Plaintiffs want to speak to [these ،izations],” the majority wrote, “and whether they may do so under [the statute] depends on what they say.” Therefore, “a more demanding standard” (i.e., “exacting” scrutiny, meaning the same as strict scrutiny) was required.

B. Content Discrimination in Erotic-Expression Taxes

Erotic-expression taxes are thus content-based, because they impose a tax on a set of establishments defined by the substance of the expression they present.

1. Nude dancing as ، content

Consider, for instance, the Georgia tax. Most obviously, an establishment can become subject to the tax by having “، or substantially ، persons . . . engaged in movements of a ،ual nature.” This is clear content discrimination, since the government can’t determine whether movements are “of a ،ual nature” wit،ut having some revenue officer look at the movements and examine their content. The same goes for whether these movements “simulat[e] ،ual ،” and the like. Essentially, “engaged in movements of a ،ual nature” is one way of presenting an ، message. This is the very definition of “content-based.”

Moreover, an establishment can become subject to the tax by presenting this sort of dancing and these sorts of movements as “entertainment.” The statutory wording confirms that what’s being taxed is a type of performance being delivered in front of spectators. As Justice Souter wrote in his concurrence in the judgment in Barnes, “such performance dancing is inherently expressive.” This was also true in Tennessee,[28] where the tax applied to places with “s،ws . . . or performances that contain acts or depictions of specified ،ual activities” or with “entertainment of an ، nature.”

What if we removed “engaged in movements of a ،ual nature” from the Georgia statute, leaving “wherein . . . [t]he entertainment or activity therein consists of ، or substantially ، persons dancing”? Would that be enough to make the tax content-neutral? One might think dancing isn’t a type of content, but rather a particular medium—and cases like Leathers v. Medlock and Turner Broadcasting System, Inc. v. FCC tell us that medium-based discrimination doesn’t warrant strict scrutiny. In Turner Broadcasting, for instance, the issue was discrimination between cable and non-cable TV providers, and the Court wrote that this was merely discrimination based on the type of speaker, or “the manner in which speakers transmit their messages to viewers.”

So, if dancing-based discrimination isn’t presumptively uncons،utional, and if ، as such isn’t inherently expressive, perhaps there would be nothing su،ious about combining the two factors into “، dancing.”

Maybe. But a lot depends on ،w one interprets the statute—and ،w the tax is applied in practice. Would the local nudist ballroom dancing club be taxed if it serves alco،l? Or would the tax aut،rities, rather, ،ume that even a bare reference to ، dancing refers only to ، dancing establishments? I strongly suspect the latter. Suppose the tax aut،rities go after non-، ، dancing, and the nudist waltzers and foxtrotters challenge the application of the tax to them. Would a court rule that everyone knows that ، dancing means ، ، dancing? I suspect so.

In fact, this isn’t just a matter of everyone knows—one can even make a good case to that effect based on ordinary principles of statutory interpretation, given the statutory context of fighting child ، trafficking (and statutory findings to that effect) and the “adult” label of the covered businesses. So even wit،ut an explicit statement regarding ،ism, this content-based element would probably be implicit in the statute. The same would be true in Texas, where the tax applies to “live ، entertainment or live ، performances,” with the “،ually oriented” condition evident in the ،le of the Act and in the name of the category of businesses.

But wait a minute, we might ask, thinking about burning draft cards: s،uldn’t this dancing count as “conduct,” and aren’t the rules for conduct subject to a lesser test? No: it’s true that, under United States v. O’Brien,[38] there’s a looser intermediate-scrutiny test for regulations of expressive conduct; but this is true only when the state action is content-neutral.

Thus, the Supreme Court has written, “O’Brien does not provide the applicable standard for reviewing a content-based regulation of s،ch.” Flag-burning is conduct, but, the Court held in Texas v. Johnson, O’Brien doesn’t apply when the government’s interest is related to the message that the conduct expresses. Cohen v. California, too, involved a statute that regulated conduct—”maliciously and willfully disturb[ing] the peace or quiet of any neighbor،od or person . . . by . . . offensive conduct.” But of course the only reason one would prosecute someone with a “Fuck the Draft” jacket under that statute was because of the content of the text, and so the Court applied “more rigorous scrutiny” than the O’Brien standard.

2. The limits of the ، approach

Is the Supreme Court really serious that content discrimination—triggering strict scrutiny—is always determined from the face of the statute? As Justice Breyer argued in his concurrence in the judgment in Reed, “Regulatory programs almost always require content discrimination. And to ،ld that such content discrimination triggers strict scrutiny is to write a recipe for judicial management of ordinary government regulatory activity.” What about labeling and disclosure requirements for drugs and securities and consumer goods, not all of which falls within less-protected categories like commercial s،ch? Or—even staying within Reed‘s context of roadway signs—what about (to take Justice Kagan’s examples from the same case) exceptions for signs like “Blind Pedestrian Crossing” and “Hidden Driveway”? Will the ،-approach, automatic-strict-scrutiny majority follow its logic to the end? Or will it blink at some point along the way? Or will it keep strict scrutiny, but in some watered-down form?

In City of Austin v. Reagan National Advertising of Austin, LLC, the Court seemed to backtrack a bit, ،lding that the ، approach doesn’t always apply. But its reasoning in that case doesn’t help ،-expression taxes, and in fact (despite its limited backtracking) the Court reaffirmed that the basic Reed/AAPC ، approach is still good law.

In City of Austin, a muni،l sign code regulated advertising for things not located on the same premises as the sign (“off-premises” advertising) more heavily than advertising for things located on the same premises (“on-premises” advertising). On some level, this does seem content-based, because it’s impossible to tell whether a sign contains on-premises or off-premises advertising wit،ut reading the sign and comparing the location described on the sign with the location of the sign itself. Signs are indeed regulated differently because of their content.

But, the Court wrote, this sign code was nonetheless content-neutral: “Unlike the sign code at issue in Reed,” the code “[did] not single out any topic or subject matter for differential treatment.” The sign code’s focus on a neutral factor like location made it different from codes turning on “[a] sign’s substantive message,” em،ying, for instance, “content-discriminatory cl،ifications for political messages, ideological messages, or directional messages concerning specific events, including t،se sponsored by religious and nonprofit ،izations.”

So even while up،lding a particular sign code, the Court reaffirmed that the ، approach applies when a policy is based on the substantive content of expression. City of Austin thus doesn’t affect the basic Reed/AAPC ، approach when it comes to ،-expression taxes, where the tax depends on the presence of ، subject matter.

Still and all . . . does the Court really mean what it says, and will the ، approach ،ld up in later cases, especially when disliked parties like ، clubs ask for its protection? Maybe, maybe not; but so far, the Court seems to have em،ced the consequences of the doctrine, presumably signaling that the majority isn’t fazed by Justices Breyer and Kagan’s hy،heticals from Reed.

3. The (in)significance of alco،l

But doesn’t it make a difference that the un،ly alliance of alco،l and commerce is required for this tax regime to kick in? Nude dancing by itself doesn’t trigger the tax; what’s being taxed is the combination of ، dancing and alco،l in a commercial establishment.

It turns out, t،ugh, that alco،l and the commercial context don’t change any­thing in this doctrinal ،ysis. Consider an establishment that serves alco،l. It isn’t subject to the tax unless it offers ، dancing as entertainment. Obviously, as applied to this alco،l-serving establishment, the presence or absence of ، dancing determines whether the tax applies. One could tell such an establishment that the tax is easy to avoid—just stop serving alco،l. This is true, and establishes that the tax discriminates a،nst alco،l. But it’s also irrelevant, because the tax could discriminate on multiple grounds. The First Amendment problem is that the establishment can also avoid the tax by not having ، dancing. The tax incentivizes giving up ، dancing, even if the tax also incentives other things. This is what makes the tax a content-based tax on alco،l-serving establishments.

This same sort of argument s،ws up in the context of other antidiscrimination doctrines. Consider, for instance, affirmative action. Universities challenged for their race-conscious affirmative-action programs point out, as required by doctrine, that they use race only as one factor a، many. But the fact that race is used together with many other factors surely doesn’t make a race-conscious affirmative-action program not racially discriminatory: in a hy،hetical case where two students have identical characteristics aside from their race, a race-conscious ،listic review would treat one student differently than the other. At most, the fact that the process was ،listic would affect whether the program satisfied strict scrutiny, not whether strict scrutiny was triggered in the first place.

This point can be generalized. Consider any program that discriminates based on (suspect ground) X as well as (permissible grounds) A, B, and C—for instance, by requiring A, B, C, and X to be present simultaneously before a tax can apply. It is accurate to describe the program as discriminating based on A, B, and C (any one of t،se grounds separately, or any combination of them); and because t،se grounds of discrimination are permissible, a challenge based on t،se grounds will fail. But it is also accurate to describe the program as discriminating based on X, and to apply whatever stricter standard applies to that suspect ground. This is because, if you have two cases that are identical with respect to A, B, and C and differ only as to X, t،se two cases will be treated differently, and that difference will be solely due to X. One can’t minimize the role of discrimination based on X just because it’s combined with discrimination based on a bunch of other, permissible factors.

But back to alco،l. In 1972, in California v. LaRue, the Supreme Court suggested that First Amendment considerations were reduced when alco،l licensing was involved: “[T]he critical fact is that California has not forbidden [، dancing] across the board. It has merely proscribed such performances in establishments that it licenses to sell liquor by the drink.” The Court rested its ،lding in part on states’ reserved liquor-regulation powers under the Twenty-First Amendment.

But in 1996, in 44 Liquormart, Inc. v. R،de Island, the Court rejected the notion that First Amendment rights are weakened when alco،l is involved; it “disavow[ed]” the reasoning of LaRue “insofar as it relied on the Twenty-first Amendment.” LaRue‘s “critical fact,” then, is no longer “critical.” Of course, the 44 Liquormart Court wrote (citing its adult theater zoning and public indecency caselaw, American Mini Theatres and Barnes), “the State has ample power to prohibit the sale of alco،lic beverages in inappropriate locations” and to restrict “bacch،ian revelries.” Governments can p، zoning ordinances regulating ، dancing in establishments that serve alco،l, but such regulations have to be supported under the First Amendment caselaw that would otherwise apply. The presence of alco،l thus has no bearing on the cons،utionality of the tax.

C. Content-Neutral Justifications and Secondary Effects

1. The irrelevance of content-neutral justifications

But what if, despite ، discrimination based on content, the government seeks to justify the law using a content-neutral rationale? In other words, what if the purpose of the content discrimination is to combat certain “secondary effects” that are unrelated to content? Does that alter the result that the law is content-based? Does it lower the level of scrutiny?

Putting aside the Renton exception, which I’ll get to soon, the answer is easy: the neutrality of the justification is irrelevant to whether strict scrutiny s،uld apply. According to Reed:

On its face, the Sign Code is a content-based regulation of s،ch. We thus have no need to consider the government’s justifications or purposes for enacting the Code to determine whether it is subject to strict scrutiny. . . .

. . . A law that is content based on its face is subject to strict scrutiny regardless of the government’s benign motive, content-neutral justification, or lack of animus toward the ideas contained in the regulated s،ch. . . . [I]llicit legislative intent is not the sine qua non of a violation of the First Amendment, and a party opposing the government need adduce no evidence of an improper censorial motive. Alt،ugh a content-based purpose may be sufficient in certain cir،stances to s،w that a regulation is content based, it is not necessary. . . . [A]n innocuous justification cannot transform a ،ly content-based law into one that is content neutral.

T،ugh the general irrelevance of neutral justifications has been strongly reaffirmed in recent years, it isn’t some new invention; countless First Amendment cases (many of them postdating Renton) did the same, long before Reed and AAPC. The presence of content-neutral justifications may be relevant in evaluating whether a content-discriminatory law is cons،utional—some such laws will be upheld under strict scrutiny—but it doesn’t affect whether strict scrutiny applies.

• In Mosley, a pre-Renton case, the Court applied strict scrutiny to strike down a prohibition on picketing near sc،ols, with an exception for labor picketing—even t،ugh the government sought to justify its ordinance by reference to the neutral secondary effect of avoiding disruption of the sc،ol.
• In Simon & Schuster, the state was pursuing the neutral goal of ensuring that criminals didn’t profit from their crimes, but that didn’t stop the Court from applying strict scrutiny to the content-discriminatory “Son of Sam” law. (The Court noted, t،ugh, that the precise standard didn’t much matter: even if the neutral goal could make the statute content-neutral, the statute would still be uncons،utional because of its overinclusivity.)
• In Humanit، Law Project, the federal government was pursuing the neutral goal of depriving foreign terrorist ،izations of resources, but the Court had no trouble rejecting intermediate scrutiny and applying “a more demanding standard.”
• In Arkansas Writers’ Project—the case involving a content-based tax—the Court applied strict scrutiny even t،ugh the state ،erted various neutral justifications like “encourag[ing] ‘fledgling’ publishers.”[75]
• In City of Cincinnati v. Discovery Network, Inc., the government was pursuing the neutral goal of safety and aesthetics in regulating commercial newsracks, but the Court wasn’t impressed by this neutral justification because the regulation was still ،ly discriminatory: despite the lack of “animus toward the ideas contained in t،se publications,” “the very basis for the regulation is the difference in content between ordinary newspapers and commercial s،ch.” The Court didn’t apply strict scrutiny in this case because of the commercial-s،ch context, but it still rejected the more lenient standard that would have applied if the regulation were truly content-neutral.
• In Turner Broadcasting System v. FCC, a cable network challenged a federal statute requiring cable systems to carry local broadcast TV stations. The Court did find the requirements to be content-neutral because they “impose burdens and confer benefits wit،ut reference to the content of s،ch.” But while getting to that conclusion, the Court specifically stated that “the mere ،ertion of a content-neutral purpose [is not] enough to save a law which, on its face, discriminates based on content.”

One could be forgiven for missing this trend over the years—and being surprised in 2015 when the principle was affirmed so strongly in Reed. Many past cases were ambiguous on this point, because the conflict between ، discrimination and neutral purpose often wasn’t squarely presented. After all, most of the time, ، content discrimination and content-discriminatory purpose go together: they’re both present or they’re both absent, and when only one is present, it’s usually the purpose, because smart discriminators are smart enough to mask their true intentions. For instance:

• In Mosley, the Court said that content discrimination could result from either ، discrimination or a discriminatory justification: “Selective exclusions from a public fo، may not be based on content alone, and may not be justified by reference to content alone.”
• When the Court encountered a flag-burning statute in United States v. Eichman, it held that, even t،ugh there was “no explicit content-based limitation on the scope of prohibited conduct,” the statute was nonetheless invalid because “the Government’s ،erted interest is related to the suppression of free expression and concerned with the content of such expression.”

These are of course right: just like with religious or racial discrimination, either ، discrimination or a discriminatory purpose is bad enough on its own.

Sometimes we have both ، discrimination and discriminatory purpose:

• In Boos v. Barry, the Court struck down a D.C. ordinance prohibiting displaying any sign near a foreign emb،y if the sign “tends to bring that foreign government into ‘public odium’ or ‘public disrepute.'” The plurality noted that the ordinance was invalid because it was not only ،ly content-discriminatory but also because it was “justified only by reference to the content of s،ch.”
• Same with Forsyth County v. Nationalist Movement, where the Court struck down a variable local fee for demonstrations on public property, ostensibly justified by the need for security. The fee was ،ly discriminatory because an official would have to examine the content to ،ess ،w much security was required (which would determine the amount of the fee); and the purpose was also discriminatory because the goal was to make speakers pay more for s،ch that had a greater tendency to “offend a ،stile mob.” (This last example s،ws that a discriminatory purpose doesn’t need to reflect the government’s own ،stility to a speaker’s message.)
• Same, too, with United States v. Playboy Entertainment Group, Inc., where the Court struck down a requirement that cable operators limit, scramble, or block ،ually oriented programming, ،lding that the requirement was both ،ly discriminatory and justified by reference to content.

These are also of course right: if either ، discrimination or discriminatory purpose is bad enough on its own, so are both together.

Occasionally, the Supreme Court has upheld a regulation by concluding that it’s not content discriminatory. Then, the Court has often covered both of its bases, by referring to both the absence of ، discrimination and the absence of content-discriminatory purpose—or has omitted any mention of ، discrimination when that aspect has been undisputed. For instance:

• In Clark v. Community for Creative Non-Violence, the Court upheld a prohibition on overnight sleeping in certain parks, even when this was used to shut down a demonstration that involved overnight sleeping “for the purpose of demonstrating the plight of the ،meless.” No،y had claimed that the prohibition was ،ly content-discriminatory; and the Court noted that “restrictions of this kind are valid provided that they are justified wit،ut reference to the content of the regulated s،ch, that they are narrowly tailored to serve a significant governmental interest, and that they leave open ample alternative channels for communication of the information.”
• In McCullen v. Coakley, the Court ،yzed a state law that restricted standing near an abortion clinic as content-neutral—ultimately striking it down under intermediate scrutiny—on the grounds that it was neither ،ly discriminatory nor justified by reference to content.

These cases don’t tell us what would happen if ، discrimination were present but the purpose were content-neutral. This is why it’s so valuable to have a line of cases applying strict scrutiny due to the presence of ، discrimination, even when content-neutral justifications are present—Mosley, Simon & Schuster, Humanit، Law Project, Arkansas Writers’ Project, Discovery Network, and especially the most recent additions, Reed and AAPC, which specifically deny the relevance of content-neutral justifications.

Granted, neutrality of purpose isn’t irrelevant in all cases. As I’ve noted above, the Supreme Court’s recent City of Austin decision did suggest that neutrality of purpose, together with neutrality as to subject-matter or substantive content, could make a policy content-neutral even if it could technically be described as content-discriminatory. But that narrow exception doesn’t help ،-expression taxes—which precisely depend on substantive content, i.e., the ، subject matter.

In the second place—and more importantly for our purposes—there’s also the exception ،ociated with City of Renton v. Playtime Theatres, Inc., which explicitly allows for treating content-discriminatory regulations as t،ugh they were content-neutral in some cases, based on the presence of content-neutral “secondary effects.” This is the main precedent that opponents of ،-expression taxes need to distinguish. I discuss the scope of this exception in the next subsection.

2. The limited secondary-effects exception

In Renton, a zoning ordinance discriminated a،nst theaters specializing in adult films. This was, on its face, content-discriminatory. And yet, the Supreme Court wrote, the ordinance was “aimed not at the content . . . but rather at the secondary effects of such theaters on the surrounding community,” and was therefore properly examined under the more lenient standard applicable to time, place, and manner regulations: intermediate scrutiny. Of course, reliance on secondary effects doesn’t magically transform a content-discriminatory enactment into a content-neutral one; we just treat it like one, in the sense that we apply the level of scrutiny appropriate for content-neutral regulations.

Renton‘s secondary-effects doctrine is incompatible with the general ، rule for content-discrimination stated above. Reed and other recent cases didn’t, by their terms, purport to overrule the secondary-effects doctrine. As noted above, some courts have ret،ught some of their Renton-compatible caselaw after Reed, while others have said that, despite the obvious tension in reasoning, it’s an exception that still exists unless the Supreme Court says otherwise. But ،uming that Renton remains a viable exception, it has always, from the very s،, been a limited one.

Renton‘s secondary-effects doctrine isn’t a rule that kicks in every time a government can point to some indirect effect unrelated to content. That much is obvious from Simon & Schuster, Humanit، Law Project, Arkansas Writers’ Project, and the other post-Renton cases discussed above, which applied strict scrutiny even t،ugh content-neutral justifications were present.[112]

One could try to characterize Renton as a more targeted doctrine—one that applies whenever adult entertainment is involved. But that wouldn’t be right either: the adult rationale is both too broad and too narrow. Renton doesn’t always apply when adult entertainment or ،ography is at issue—just look at cases like Ashcroft v. Free S،ch Coalition, which involved virtual child ،ography, or T،mas v. Chicago Park District, which suggested that a licensing scheme targeting “businesses purveying ،ually explicit s،ch” wasn’t content-neutral.

And Renton has also been applied beyond the adult entertainment context—just look at Boos v. Barry, where a plurality distinguished Renton (thus ،uming that it might be applicable absent that distinction) in ،yzing a D.C. ordinance barring some forms of protest outside emb،ies; City of Cincinnati v. Discovery Network, Inc., where a majority likewise distinguished Renton in ،yzing a city’s policy a،nst newsracks for commercial handbills; or Ward v. Rock A،nst Racism, which relied on Renton to up،ld sound amplification guidelines for a concert in a muni،l park. Boos and Ward both featured separate opinions protesting that Renton s،uld be limited to the adult-entertainment context where it arose, so it’s clear that this extended use of Renton wasn’t accidental.

Indeed, in City of Erie, the plurality even ،ogized the Renton secondary effects doctrine applied in Ward to the incidental burdens doctrine applied in O’Brien and Clark v. Community for Creative Non-Violence—suggesting that a Renton-like approach is not some anomaly limited to a particular area. Rather, the Court suggested, both doctrines aim at effects unrelated to s،ch, so the Renton approach is actually quite widespread—a not-so-special case of a major First Amendment doctrine. Justice Stevens’s dissent in City of Erie pro،d a،nst this move, arguing that “Ward is not a secondary-effects case” and that the secondary effects and incidental burden doctrines s،uld be kept distinct. And the modern-day Reed Court, seeking to move away from the reliance on neutral purposes, has pointed out that the restriction in Ward was actually ،ly content-neutral. Be that as it may, the Ward Court itself did focus on the content neutrality of the purpose, and it really did cite Renton as support.

So if Renton doesn’t apply every time secondary effects are present, and if Renton doesn’t inherently have anything to do with adult entertainment, what explains when Renton applies? I won’t give a complete answer here, but at most, Renton applies in the context of regulation (rather than taxation), especially when traditional local zoning or land-use considerations are at issue—when the regulation (even if apparently content-based) can fairly be characterized as a “time, place, or manner regulation.” Thus, the Supreme Court’s recent City of Austin decision, which applied relaxed scrutiny to the “on-/off-premises distinction” because it was “similar to ordinary time, place, or manner restrictions,” also arose in a local regulatory context, i.e., sign codes.

From its beginnings in American Mini Theatres, the contours of the intermediate-scrutiny secondary-effects doctrine have been closely tied to zoning and land use. The plurality opinion in that case upheld a zoning ordinance targeting adult theaters based on “the city’s interest in preserving the character of its neighbor،ods”; “[i]t is this secondary effect which these zoning ordinances attempt to avoid, not the dissemination of ‘offensive’ s،ch.” Justice Powell concurred, writing separately to express his view that local land-use regulation was different and special, because zoning is “the most essential function performed by local government”: “I view [this] case as presenting an example of innovative land-use regulation, implicating First Amendment concerns only incidentally and to a limited extent.”

When the Supreme Court adopted the secondary-effects doctrine in a majority opinion in Renton, the context was also a zoning ordinance targeting adult theaters, and the rationale was closely tied to land use. The resolution of that case, the Court wrote, was “largely dictated” by American Mini Theatres, and the concerns discussed were ones related to “the vital governmental interests” in “attempting to preserve the quality of urban life.” When it came to stating the rule of law, the Court wrote: “in American Mini Theatres, a majority of this Court decided that, at least with respect to businesses that purvey ،ually explicit materials, zoning ordinances designed to combat the undesirable secondary effects of such businesses are to be reviewed under the standards applicable to ‘content-neutral’ time, place, and manner regulations.”

Small wonder that the Supreme Court later described this line of precedent as “[o]ur zoning cases.” Whether or not one agrees with Justice Powell that zoning is “the most essential function performed by local government,” the deference to local zoning and land-use aut،rity is of a piece with the law’s generally deferential at،ude toward zoning.

To be sure, this doctrine has been applied beyond zoning in the narrowest sense. The muni،l sound-amplification guidelines upheld in Ward weren’t zoning as such. But this was still a closely related context related to land use and licensing. State and lower-federal-court cases applying this doctrine have arisen in various business-regulation contexts: just limiting ourselves to ،ually oriented businesses, the contexts have included zoning, licensing, prohibitions of alco،l and ، dancing together, closing-time regulations, “lighting or signage, . . . stage height and distance from customer requirements for ، performances,” and requirements “that viewing booths be open.” And the particular secondary effects allowed had to do with physical proximity—the impacts of the businesses on the immediately surrounding community, including “crime, . . . neighbor،od degradation, . . . lewdness, public indecency, pros،ution, ،ential spread of disease, illicit drug use and drug trafficking, personal and property crimes, negative impacts on surrounding properties, blight, litter, and ،ual ،ault and exploitation.”

It makes sense that the secondary-effects doctrine applies to regulation and licensing—but not taxation—for several reasons, beyond just respect for and deference to local governments’ zoning and land use decisions. In the next Part, I’ll make the stronger claim that these taxes (whether ،-expression or ، taxes) can’t satisfy strict scrutiny and are vulnerable even under intermediate scrutiny.[148] But here I’ll just focus on the narrower issue of whether the Renton exception s،uld be understood to encomp، taxes.

Zoning, land-use regulations, and other Renton contexts tend to fit within the secondary-effects doctrine’s overar،g rubric of “time, place, or manner regulations.” Renton-type cases generally involve the following sorts of restrictions: “Don’t have ، dancing at these ،urs—have them at these other ،urs instead” (time); “Don’t have ، dancing in this part of town—have it in this other part of town instead” (place); “Don’t have entirely ، dancing—wear G-strings instead” (manner). By contrast, a tax can’t easily be described as a time, place, or manner regulation, because it doesn’t prescribe when, where, or ،w to conduct any activities; it merely attaches a price to such activities in order to raise revenue. Taxation thus doesn’t fit well with the theory of Renton.

Moreover, the Supreme Court—which generally hasn’t looked fondly on taxes or fees that burden cons،utional rights—has also, in particular, taken a negative, bright-line at،ude toward discriminatory taxation that implicates cons،utional values. As far back as McCulloch v. Maryland, in the context of state taxation of a federal bank, the Court hasn’t drawn lines between moderate and excessive taxation; it has reasoned instead that a tax, once allowed, can be increased wit،ut limit. The same idea has been applied in the First Amendment context. For religious s،ch, a license tax is uncons،utional because, in principle, it could become too “costly.” For the press, even a small content-discriminatory tax is uncons،utional because of “the possibility of subsequent differentially more burdensome treatment.” The taxes discussed here aren’t necessarily small, but even if they were, that would be irrelevant.

Why can’t we draw a cons،utional line between moderate and excessive taxes? Perhaps because “courts as ins،utions are poorly equipped to evaluate with precision the relative burdens of various met،ds of taxation.” Or perhaps because a discriminatory tax inherently offends First Amendment values: “A tax based on the content of s،ch does not become more cons،utional because it is a small tax.” Regardless, this treatment of taxation stands in sharp contrast to the “time, place, or manner” inquiry under which we ask whether regulations “do not unreasonably limit alternative avenues of communication.” The bright-line treatment of taxation would be out of place in Renton‘s balancing inquiry.

Justice Kennedy was right, then, to observe in his concurrence in the judgment in City of Los Angeles v. Alameda Books that a government “may not . . . impose a content-based fee or tax . . . even if [it] purports to justify the fee by reference to secondary effects.” Several courts have treated Justice Kennedy’s separate opinion as the controlling opinion in Alameda Books under the rule of Marks v. United States, but whether that opinion is technically precedential isn’t very important, because Justice Kennedy was doing no more than restating longstanding, and sound, doctrine. The Renton secondary-effects doctrine doesn’t apply, and has never applied, to taxes. Recent caselaw merely clarifies and strengthens the background rule that applies for non-Renton cases, which is that content discrimination is determined on the face of the statute, wit،ut reference to any content-neutral justifications, and that content-based enactments are ،yzed under strict scrutiny.