Discrimination, the Speech That Enables It, and the First Amendment

Imagine that you’re interviewing for your dream job, only to be asked by the hiring committee whether you’re pregnant. Or HIV-positive. Or Muslim. Does the First Amendment protect your interviewers’ inquiries from government regulation? This Article explores that question.1

Antidiscrimination laws forbid employers, housing providers, insurers, lenders, and other gatekeepers from relying on certain characteristics in their decision-making.2 Many of these laws also regulate those actors’ speech by prohibiting them from inquiring about applicants’ protected class characteristics;3 these provisions seek to stop illegal discrimination before it occurs by preventing gatekeepers from eliciting information that would enable them to discriminate. Although these laws generated little if any First Amendment controversy for decades, they now face new constitutional attacks inspired by the antiregulatory turn in the Supreme Court’s Free Speech Clause doctrine.4

Part I of this Article starts by describing how gatekeepers’ inquiries about applicants’ protected characteristics enable illegal discrimination. It then outlines the wide variety of efforts by federal, state, and local legislatures to tackle thorny problems of inequality by restricting gatekeepers’ inquiries about applicants’ protected characteristics. Next, it identifies the potential collision course between these measures and the recent antiregulatory turn in First Amendment law and litigation.

Part II examines the theory and doctrine that support these laws’ constitutionality, explaining why the government’s restriction of the speech that enables conduct that the government has legitimately regulated triggers no First Amendment scrutiny. More specifically, the First Amendment permits the government to restrict speech that initiates or accomplishes conduct that the government has regulated—speech that does something and not just says something, to use legal scholar Kent Greenawalt’s vocabulary.5 As an illustration of speech that is unprotected because it initiates or accomplishes illegal conduct, the Court has repeatedly pointed to gatekeepers’ speech that enables illegal discrimination: “Congress, for example, can prohibit employers from discriminating in hiring on the basis of race. The fact that this will require an employer to take down a sign reading ‘White Applicants Only’ hardly means that the law should be analyzed as one regulating the employer’s speech rather than conduct.”6

In other words, a gatekeeper’s statement “White Applicants Only” is unprotected because it declares certain transactions and opportunities as off limits to protected class members; precisely because of gatekeepers’ power, their speech in these transactional settings thus does something and not just says something. Once we understand why the First Amendment does not protect those statements, we can see that the First Amendment similarly permits the government to regulate gatekeepers’ transaction-related inquiries about candidates’ protected class status—inquiries that enable illegal discrimination by deterring candidates based on their protected class status as well as by eliciting the information that facilitates gatekeepers’ discriminatory decisions.

Part II next explains how the Court’s longstanding commercial speech doctrine captures these insights by holding that the First Amendment does not protect commercial speech related to illegal activity. It then applies this doctrine to the antidiscrimination laws identified in Part I, concluding that the government’s restriction of gatekeepers’ inquiries about applicants’ protected class status triggers no First Amendment scrutiny because those inquiries constitute commercial speech related to the illegal activity of discriminatory employment, housing, and other transactions.

Part III briefly considers the First Amendment implications of other antidiscrimination provisions that regulate transactional parties’ speech in various ways, sometimes by restricting speech and sometimes by requiring it. It shows how here too the Court’s commercial speech doctrine provides the relevant analysis, with its focus on protecting speech that furthers listeners’ First Amendment interests while permitting the regulation of speech that frustrates those interests.

I.     Antidiscrimination Laws That Prohibit Gatekeepers’ Reliance On, and Inquiries About, Applicants’ Protected Characteristics

As this Part explains, gatekeepers’ inquiries that elicit candidates’ protected class status facilitate illegally discriminatory decisions about important opportunities and deter candidates from pursuing those opportunities.7 Legislatures thus often enact laws prohibiting gatekeepers not only from relying on, but also from inquiring about, applicants’ protected class status to stop illegal discrimination before it happens. Legislatures’ interest in stopping discrimination before the fact is especially strong because after-the-fact enforcement is frequently slow, costly, and ineffective.

A.    How Gatekeepers’ Inquiries About Applicants’ Protected Characteristics Enable Illegal Discrimination

Information about applicants’ protected characteristics enables gatekeepers to discriminate, intentionally or otherwise, against those applicants. When gatekeepers know (or think they know) candidates’ race, gender, or other protected characteristic, they too often rely on that information to discriminate in their decisions about jobs, housing, credit, and other opportunities and services.8 Consider, for instance, a Harvard Business School study, which found that Airbnb hosts used information collected and shared by Airbnb to discriminate against prospective guests with “distinctively African-American names.”9 In the same vein, Facebook recently settled complaints filed by nonprofit civil rights organizations alleging that Facebook used information about its users’ protected class status to enable housing providers to steer users to—or away from—certain housing opportunities based on that status.10

Gatekeepers often acquire the information that enables discrimination by asking candidates about their protected class status in applications, interviews, negotiations, and more. Sometimes decisionmakers intentionally seek information about candidates’ protected characteristics to inform their discriminatory decision-making. For example, a Congressional committee report on the Americans with Disabilities Act (ADA) explained:

Historically, employment application forms and employment interviews requested information concerning an applicant’s physical or mental condition. This information was often used to exclude applicants with disabilities—particularly those with so-called hidden disabilities such as epilepsy, diabetes, emotional illness, heart disease, and cancer—before their ability to perform the job was even evaluated.11

Even when gatekeepers seek this information for benign rather than nefarious purposes, that information, once obtained, remains available for the gatekeeper’s later use, consciously or unconsciously, in screening, selecting, or compensating applicants.12

And once discrimination does occur, efforts to identify and rectify it after the fact are notoriously slow, costly, and difficult. Complaint-driven enforcement—that is, an enforcement regime that relies on individuals to file claims after they believe they have suffered illegal discrimination—is poorly-equipped to redress discriminatory selection practices and other front-end discrimination. In part, this is because an applicant denied a job or an apartment seldom receives a reason for her rejection from a potential employer or landlord and is unlikely ever to learn the successful candidate’s identity, much less his comparative qualifications or other relevant attributes.13 Other factors that contribute to the ineffectiveness of after-the-fact enforcement include the limitations of overworked and underfunded enforcement agencies, challengers’ difficulties in securing legal representation, and a wide range of procedural, evidentiary, and doctrinal barriers to proving a decisionmaker’s discriminatory intent.14 For these reasons, legal scholar Cynthia Estlund describes antidiscrimination law’s dependence on after-the-fact enforcement as its “Achilles’ heel.”15 The greater the barriers to effective after-the-fact enforcement of civil rights protections, the greater the value in preventing discrimination before the fact by denying gatekeepers the information that enables them to discriminate. As law professor Ignacio Cofone observes, “[d]iscrimination is better avoided than compensated.”16

Moreover, because these inquiries are generally made of a less powerful applicant by a more powerful gatekeeper, a candidate’s response may be coerced: either she gives the requested information and risks discrimination if the gatekeeper relies on that information to withhold opportunities, or she refuses to provide the information only to be rejected for the opportunity altogether.17 For instance, one employer declined to hire an applicant after she refused to answer an interview question about her plans to have a family; one of the interviewers responded to her reticence by stating that he “did not want to hire a woman who would get pregnant and quit.”18 Another employer fired a worker when she refused to answer questions about her reproductive choices, questions that included “whether she was pregnant, had ever been pregnant, or was planning to become pregnant; whether she had ever had an abortion, miscarriage, or live birth, and if so, how many times; and whether she was on birth control and, if so, what type.”19

Inquiries of this sort can also deter applicants from pursuing important opportunities by signaling the decisionmakers’ discriminatory preferences.20 Think, for instance, of an applicant with a disability: confronted by an employer’s questions about her medical status or use of prescription drugs, she may well conclude that the job is unavailable to those with certain medical conditions.21

B.    How Legislatures Regulate Gatekeepers’ Inquiries That Enable Illegal Discrimination

Legislatures often seek to prevent illegal discrimination before it happens by not only forbidding decisionmakers from relying on certain characteristics (that is, from using information about protected class status) in their decision-making, but also by forbidding decisionmakers from eliciting that information.22 Thus, antidiscrimination laws often regulate both gatekeepers’ conduct—that is, their decisions about how and to whom to distribute opportunities and services—as well as the speech that enables them to engage in discriminatory conduct.23

Many of these antidiscrimination laws include provisions that prohibit decisionmakers from making certain inquiries altogether.24 For example, Pennsylvania’s state law bars employers from relying on a variety of protected characteristics in their employment decisions and also forbids them from “[e]licit[ing] any information . . . concerning the race, color, religious creed, ancestry, age, sex, national origin, past handicap or disability” of any applicant.25 After the Pregnancy Discrimination Act26 amended Title VII of the Civil Rights Act to make clear that illegal job discrimination “on the basis of sex” includes discrimination based on “pregnancy, childbirth, or related medical conditions,”27 the Equal Employment Opportunity Commission interpreted the statute to prohibit most employer inquiries about applicants’ pregnancy status.28 As another illustration, the decades-old Equal Credit Opportunity Act regulations forbid lenders from asking about applicants’ race, national origin, sex, religion, marital status, and reproductive decisions to prevent illegally discriminatory credit decisions.29

Some antidiscrimination laws instead regulate inquiries about protected characteristics at certain key junctures in the decision-making process. The Americans with Disabilities Act (ADA), for instance, prohibits employers from relying on disability status in their decision-making, and prohibits certain disability-related inquiries at various stages in the employment process to prevent discrimination from infecting employers’ ultimate decision-making. More specifically, the ADA starts by forbidding employers from asking “whether such applicant is an individual with a disability or as to the nature or severity of such disability” before extending any job offer; instead, an employer “may make preemployment inquiries into the ability of an applicant to perform job-related functions.”30 After an applicant receives a conditional job offer but before she begins work, her employer may pose disability-related inquiries regardless of their job-relatedness, so long as the employer makes the same inquiries of all new employees in the same job category.31 Finally, after an employee has started work, an employer may ask only those disability-related questions that are “job-related and consistent with business necessity.”32

Some antidiscrimination laws prohibit not only gatekeepers’ direct inquiries of applicants, but also their efforts to learn about applicants’ protected characteristics from other sources. For instance, the federal Genetic Information Nondiscrimination Act (GINA)33 —which bars health insurers and employers from relying on, and asking about, genetic information in making insurance and employment decisions—“generally prohibits employers from seeking to obtain genetic information at any time during employment and, notably, the GINA’s implementation regulations explicitly apply to the Internet.”34 Similarly, some states forbid employers from “seek[ing] [or] obtain[ing]” applicants’ protected class information “from any source.”35

Although many of these antidiscrimination laws prohibit decisionmakers’ reliance on, and often their inquiries about, characteristics long thought immutable (like race or national origin),36 newer measures reflect legislatures’ expanding understanding of the wide variety of barriers to equality. A growing number of state and local jurisdictions now prohibit employers from relying on, and asking about, applicants’ sexual orientation and gender identity.37 Commentators hail GINA—enacted by Congress in 2008 with a near-unanimous vote—as particularly innovative in its determination to stamp out genetic discrimination before a discriminatory culture had the time to develop by prohibiting decisionmakers’ reliance on, and inquiries about, applicants’ genetic information.38 And many other recent antidiscrimination laws prohibit gatekeepers’ reliance on, and often inquiries about, certain life experiences like applicants’ marital or reproductive choices,39 current unemployment status,40 credit histories,41 status as domestic violence victims,42 certain arrest records,43 and veteran status.44 Some bar reliance on, or inquiries about, these sorts of characteristics to achieve public policy objectives in addition to equality goals. For example, “ban-the-box” laws limit employers’ inquiries about applicants’ criminal record at various points in the employment process in part because of the evidence that ex-offenders’ unemployment strongly predicts their risk of recidivism.45

As another illustration, state and local legislatures have recently begun to deploy related strategies when wrestling with tenacious gender- and race-based pay disparities.46 More specifically, a growing number of jurisdictions now prohibit employers from relying on, and inquiring about, applicants’ salary history in making decisions about hiring and pay. Concluding that candidates’ prior pay too often reflects race or gender discrimination or other factors unrelated to merit,47 these policymakers challenge many employers’ reliance on the (often inaccurate) assumption that prior pay is an accurate measure of a candidate’s skill, experience, and responsibility to reject applicants whose past salaries are perceived as too low.48 These policymakers also seek to address the even more common practice in which employers base workers’ starting pay on how much those workers earned at their last job49 —a practice that ensures that pay disparities continue to follow women and people of color from job to job.50 For all these reasons, these policymakers reject the assumption that a worker’s salary history necessarily reflects an accurate assessment of, and reward for, her job performance.51

In sum, all of these antidiscrimination laws reflect legislatures’ conclusions that relying on (and thus asking about) certain characteristics or experiences when distributing important opportunities is morally wrong, instrumentally unwise, or both.52

C.    New First Amendment Challenges to These Laws

Hundreds of federal, state, and local laws now protect certain characteristics from discrimination by prohibiting gatekeepers from both relying on, and also asking about, those characteristics. Sometimes these measures generate heated political opposition from regulated entities who resist regulation they characterize as disruptive.53 This is nothing new. As one of many examples, some employers opposed the enactment of Title VII of the Civil Rights Act of 1964, the federal law barring job discrimination based on race, sex, color, national origin, and religion.54 And some business owners and associations opposed enactment of the Americans with Disabilities Act, including its protections for HIV-positive workers.55 Yet disrupting gatekeepers’ practices that legislatures have identified as harmful is precisely the point of these efforts. Recall Justice Brandeis’s memorable explanation of the power and value of legislative experimentation in responding to pressing problems:

[T]here must be power in the states and the nation to remold, through experimentation, our economic practices and institutions to meet changing social and economic needs. I cannot believe that the framers of the Fourteenth Amendment, or the states which ratified it, intended to deprive us of the power to correct the evils of technological unemployment and excess productive capacity which have attended progress in the useful arts. To stay experimentation in things social and economic is a grave responsibility. Denial of the right to experiment may be fraught with serious consequences to the nation. It is one of the happy incidents of the federal system that a single courageous state may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.56

Indeed, the Supreme Court has consistently recognized legislatures’ constitutional power to challenge and change longstanding practices judged to be unjust, inefficient, or both. This includes state and local jurisdictions’ constitutional power to regulate the terms and conditions of employment and other transactions (subject only to rational basis scrutiny),57 as well as Congress’s Article I interstate commerce clause power to regulate these matters through federal legislation (again, subject only to rational basis review).58 When a legislature bars gatekeepers from relying on certain characteristics in distributing opportunities and services, it requires those gatekeepers to use what it believes to be better indicia of candidates’ ability and merit. Regardless of whether one agrees with a specific legislature’s conclusions, whether and when legislatures should choose to regulate employers’, lenders’, insurers’, and housing providers’ decision-making is a policy question rather than a constitutional question. In other words, legislatures’ constitutional power to regulate decisionmakers’ reliance on credit history, salary history, or certain other experiences and histories is no different from its constitutional power to regulate decisionmakers’ reliance on characteristics like race, religion, or gender.

Again, antidiscrimination law prohibits gatekeepers from relying on information about certain characteristics in their decision-making when the legislature concludes that such reliance is unfair, unwise, or both. And once legislatures so regulate, it then makes sense for them to restrict gatekeepers’ inquiries eliciting the information that enables what is now illegal discrimination.59

Until very recently these laws generated little, if any, constitutional controversy. But the contemporary antiregulatory turn in First Amendment law and litigation has emboldened new attacks on governmental efforts to address sticky problems of inequality through the sorts of antidiscrimination laws described above. This turn—characterized by some as the “weaponization” of the First Amendment60 —has been described at length elsewhere,61 and includes corporate and other commercial entities’ increasingly successful efforts to resist regulation in a variety of settings.

Most relevant to this Article, the Greater Philadelphia Chamber of Commerce [hereinafter “Philadelphia Chamber”] recently challenged Philadelphia’s salary history law that prohibits employers from relying on, and asking about, applicants’ prior pay when making hiring and compensation decisions.62 In so doing, the Philadelphia Chamber and other industry associations made several sweeping arguments inspired by the Court’s antiregulatory turn, arguing that the First Amendment protects gatekeepers’ ability both to rely on, and ask about, salary history when choosing among and compensating applicants for available job opportunities.63 If accepted, these arguments would also threaten many other antidiscrimination laws, both longstanding and new.

Most aggressively, the Philadelphia Chamber claimed that Philadelphia’s law unconstitutionally restricted employers’ ability to express their view—through their actual employment decisions—that salary history is relevant to workers’ merit. As its brief argued, “[a]n employer who relies on an applicant’s wage history when formulating a proposed salary is communicating a message about how much that applicant’s labor is worth to the employer: the higher the proposed salary, the more valuable the applicant is to the employer.”64 The Philadelphia Chamber thus characterized the government’s regulation of employers’ reliance on information in its hiring and pay decisions as the regulation of speech that should trigger, and fail, heightened scrutiny.65 Indeed, the lawsuit described the entire statute not as a regulation of commercial conduct that triggers only rational basis review, but instead as a regulation of speech based on “disagreement with employers’ message that an employer’s assessment of a prospective employee’s appropriate salary, as reflected in the employer’s salary offer, can be informed by the prospective employee’s salary history.”66 In other words, the Philadelphia Chamber’s lawsuit attacked the government’s constitutional power to regulate discriminatory conduct by restricting gatekeepers’ use of certain information in distributing important opportunities.

Some businesses and employers in 1964 similarly resisted enactment and enforcement of the Civil Rights Act because they felt that requiring them not to discriminate on the basis of race interfered with their ability to communicate their views about race.67 And some employers argued that their assessment of an applicant’s suitability or value is, and should be, informed by sexual orientation or other characteristics now increasingly protected from discrimination by law.68 Many employers believed the same about pregnancy or disability or age because they felt that those characteristics predict workers’ cost or ability; some continue to believe it.69 And some employers no doubt think the same about credit history or arrest record or salary history—i.e., they believe that those characteristics are important to hiring and compensation decisions because they might predict candidates’ ability. As we’ve seen, however, the Court has long made clear that legislatures have the constitutional power to prohibit gatekeepers’ reliance on such characteristics in distributing opportunities and services once those legislatures conclude that such characteristics are not—or should not be—relevant to decision-making.70 The Third Circuit recognized this when it denied the Philadelphia Chamber’s request to preliminarily enjoin the provision of Philadelphia’s law that forbids employers from relying on applicants’ salary history in hiring and compensation decisions.71

The Philadelphia Chamber also specifically challenged legislatures’ power to restrict gatekeepers’ inquiries that enable illegal discrimination. Although deployed so far to challenge laws regulating employers’ inquiries about salary history, these arguments would apply with equal force to the wide range of federal, state, and local statutes described above that prohibit gatekeepers’ questions about religion, national origin, disability, pregnancy, sexual orientation, and many other protected characteristics. We may anticipate similar challenges to other statutes, perhaps starting with laws of relatively recent vintage like statutes prohibiting employers from asking about, and relying upon, applicants’ genetic information, credit history, and reproductive decisions.72

As the next Part explains, these challenges should not succeed. Once a legislature prohibits certain transactions as illegally discriminatory, First Amendment theory and doctrine support the legislature’s choice also to restrict the speech that enables this now-illegal conduct, including but not limited to gatekeepers’ inquiries about applicants’ protected class status.

II.   Why the First Amendment Does Not Protect Speech That Enables Illegally Discriminatory Transactions

This Part starts by examining why the First Amendment does not protect speech that initiates or accomplishes conduct that the government has regulated—in other words, speech that does something and not just says something. It then explains how the speech that enables illegal conduct more generally—as well as the speech that enables illegal discrimination more specifically—exemplifies speech that does something and not just says something. Next, this Part demonstrates how the Supreme Court’s commercial speech doctrine has long captured this insight by holding that the First Amendment does not protect commercial speech related to illegal activity such that the government’s regulation of such speech triggers no First Amendment scrutiny. It closes by describing this doctrine’s application to the laws described in Part I, concluding that the First Amendment does not protect gatekeepers’ inquiries about applicants’ protected class status because those inquiries constitute commercial speech related to the illegal activity of discriminatory employment, housing, and other transactions.

A.    Why the First Amendment Does Not Protect Speech That Enables Regulated Conduct More Generally

The government routinely, and in a variety of settings, restricts speech that enables regulated conduct without triggering any First Amendment scrutiny. Antitrust law, for instance, “restricts the exchange of accurate market, pricing, and production information, as well as limits the advocacy of concerted action in most contexts; yet it remains almost wholly untouched by the First Amendment.”73 Nor does the First Amendment protect solicitations of, and conspiracies to engage in, illegal activity.74

A number of thoughtful commentators have considered this dynamic, explaining it as involving a sufficiently close relationship between speech and regulated conduct that leaves us confident that the government has targeted conduct rather than ideas. Kent Greenawalt, for instance, identifies a universe of what he calls “situation-altering” speech that falls outside of the First Amendment’s protection because it does something rather than just says something. In other words, this speech “dominantly represent[s] commitments to action” rather than “assertions of facts or values or expressions of feeling” that have First Amendment value.75 Under this view, “communications whose dominant purpose is to accomplish something rather than to say something are not reached by a principle of free speech or are reached much less strongly than are ordinary claims of fact and value.”76 This approach explains why, for example, offers and agreements to commit a crime receive no First Amendment protection.77

Expression’s capacity to do something rather than just say something can increase with the power of the speaker. This is the case, for example, of comparatively powerful speakers’ threats and orders: “[a]nother kind of situation-altering utterance is when a boss gives a direct order of behavior to a subordinate. That is effectively a way for the boss to get done what he has ordered.”78 “Such situation-altering utterances,” Greenawalt concludes, “are not the sort of speech that warrants protection under a guarantee of free speech.”79 Targeting actions rather than ideas, the government’s restriction of such threats and orders triggers no First Amendment scrutiny.80

The government’s routine regulation of contractual and other transaction-related speech offers another illustration of this broader dynamic where the government restricts speech because it does something, and not just says something. Indeed, contract law regularly regulates transactional speech without raising First Amendment discussion, much less litigation.81 As law professor Rod Smolla explains, “[A] statement of transaction is the use of language to propose or conclude some form of transaction[,] [such as] ‘I will rent to you this apartment if you will pay me $300 per month.’ . . . . Because virtually all transactions are effectuated through language, freedom of speech never has been thought to encompass all use of language.”82 In other words, once the government exercises its constitutional power to regulate certain transactions, this inevitably requires the regulation of the speech that makes those transactions possible: “To regulate the language is to regulate the transaction.”83

Legal scholar Daniel Farber makes a similar point about speech that serves a contractual function, observing that “[c]ontract law consists almost entirely of rules attaching liability to various uses of language.”84 To help us determine whether the government’s regulation of transactional speech impermissibly targets ideas or instead permissibly targets conduct, Professor Farber proposes the following test:

A justification for regulating the seller’s speech relates to the contractual [as opposed to informational, and thus constitutionally protected] function of the speech if, and only if, the state interest disappears when the same statements are made by a third person with no relation to the transaction. If the same interest is implicated by the third party’s speech, the interest obviously cannot relate to any contractual aspect of the speech, since the third party is not involved in the contract.85

Law professor Jane Bambauer suggests a related approach for parsing the government’s permissible regulation of speech that does something from its impermissible regulation of speech because it says something, observing that “[w]hen the state has a legitimate, non-speech-related reason to manage a relationship, it will typically manage many non-speech aspects of the relationship as well.”86 And that’s what we see with respect to the government’s regulation of gatekeepers’ speech, as the government regularly regulates the conduct of employers, lenders, housing providers, and other commercial actors to prevent discrimination and promote fairness and efficiency.87

B.    Why the First Amendment Does Not Protect Speech That Enables Illegal Discrimination More Specifically

So far, we’ve seen that the First Amendment does not protect speech that accomplishes illegal conduct, nor does it protect speech that performs a contractual function: both involve speech that does something, not just says something. The speech that enables illegally discriminatory transactions thus involves two sets of circumstances “where the regulation of expressive activities seems incontrovertibly outside the ambit of First Amendment concerns: speech in the formation of contracts and speech solicitaing [sic] [illegal] activity.”88

For example, courts and commentators have long recognized (without constitutional controversy) that quid pro quo harassment—in which an employer threatens on-the-job punishment or offers an on-the-job reward based on a worker’s response to unwelcome sexual advances—is unprotected by the First Amendment.89 As Greenawalt explains more generally, “Since someone who orders another is not engaging in expression, but is attempting to have his way through power or authority, a political principle of freedom of speech is no impediment to forbidding undesirable orders.”90 In other words, the First Amendment permits the government to bar quid pro quo threats and promises because they seek to change the terms and conditions of employment through the speaker’s power over the employment relationship.

For decades the Court has also recognized that harassing workplace speech warrants the government’s constraint when sufficiently severe or pervasive to create a hostile work environment based on protected class status.91 Think, for instance, of workers regularly forced to endure an onslaught of racial or sexual slurs that alter the terms and conditions of their employment and signal certain job opportunities as off-limits to targeted individuals based on protected class status.92 For these reasons, the Court has stated that the First Amendment permits the content-based regulation of such verbal harassment as “incidental” to the government’s permissible regulation of discriminatory conduct:

[S]ince words can in some circumstances violate laws directed not against speech but against conduct (a law against treason, for example, is violated by telling the enemy the Nation’s defense secrets), a particular content-based subcategory of a proscribable class of speech can be swept up incidentally within the reach of a statute directed at conduct rather than speech. Thus, for example, sexually derogatory ‘fighting words,’ among other words, may produce a violation of Title VII’s general prohibition against sexual discrimination in employment practices.93

Along the same lines, on multiple occasions the Court has made clear that the First Amendment poses no bar to laws that forbid gatekeepers’ statements of discriminatory preference like “White Applicants Only”94 or “Jobs Of Interest to Men.”95 In so doing, the Court has identified these laws as exemplifying the government’s constitutionally permissible regulation of speech that enables the doing of something that the government has legitimately regulated.96

Consider, for instance, Rumsfeld v. Forum for Acad. & Inst. Rights.97 There the Court held that the First Amendment permits Congress to regulate certain conduct by requiring universities to provide military recruiters with the same access to campus facilities as they provide other employers—even though this law also regulated speech by requiring universities to send emails or post notices on recruiters’ behalf:

As a general matter, the Solomon Amendment regulates conduct, not speech. It affects what law schools must do—afford equal access to military recruiters—not what they may or may not say. . . . The compelled speech to which the law schools point is plainly incidental to the Solomon Amendment’s regulation of conduct, and “it has never been deemed an abridgment of freedom of speech or press to make a course of conduct illegal merely because the conduct was in part initiated, evidenced, or carried out by means of language, either spoken, written, or printed.”98

As an illustration of speech that is unprotected because it “initiates” or “carries out” illegal conduct (in other words, speech that does something and not just says something), the Rumsfeld Court pointed to gatekeepers’ speech that enables illegal discrimination: “Congress, for example, can prohibit employers from discriminating in hiring on the basis of race. The fact that this will require an employer to take down a sign reading ‘White Applicants Only’ hardly means that the law should be analyzed as one regulating the employer’s speech rather than conduct.”99

In Sorrell v. IMS Health, Inc., the Court again offered “White Applicants Only” as an example of speech unprotected by the First Amendment because it does something and not just says something.100 There, a 5-4 Court held that Vermont violated the First Amendment when it restrained the exchange of information (about doctors’ prescribing practices) that would inform disfavored but legal marketing practices (pharmaceutical companies’ marketing of brand-name drugs directly to doctors).101 Yet in so holding, the majority distinguished unprotected speech that the government may restrict free from First Amendment scrutiny because of its close relationship to illegal conduct:

It is true that restrictions on protected expression are distinct from restrictions on economic activity or, more generally, on nonexpressive conduct. It is also true that the First Amendment does not prevent restrictions directed at commerce or conduct from imposing incidental burdens on speech. That is why a ban on race-based hiring may require employers to remove “White Applicants Only” signs . . . .102

In other words, “White Applicants Only” is unprotected because it declares certain transactions and opportunities as off limits to protected class members. Precisely because of gatekeepers’ power, their speech in these transactional settings thus does something and not just says something.103 By deterring applicants from pursuing available opportunities based on protected class status, gatekeepers’ statements of discriminatory preference like “White Applicants Only” enable illegal discrimination and thus can be regulated without triggering First Amendment scrutiny.104

Once we understand why the First Amendment does not protect gatekeepers’ statements of discriminatory preference like “White Applicants Only,” we can see the implications for other speech that enables illegal discrimination. Gatekeepers’ inquiries about applicants’ protected class status, like their statements of discriminatory preference, take place in an environment in which their speech does something and not just says something precisely because of their power in that transactional setting. More specifically, these inquiries can both deter certain candidates from pursuing available opportunities and also elicit the information that makes illegal discrimination possible. First, because the gatekeeper’s query signals a preference for a term of the proposed transaction where the speaker has the functional power to insist on that term, a gatekeeper’s inquiries about candidates’ protected class status deters certain listeners from pursuing important opportunities. Think, for example, of an employer’s questions about an applicant’s religion, HIV-status, or pregnancy. Just as is the case when a decisionmaker announces its preference for “White Applicants Only,” these inquiries communicate certain opportunities as off limits to protected class members and are made by decisionmakers who have the power to enforce those limits. Second, gatekeepers’ inquiries about applicants’ protected class status also make illegal discrimination possible by eliciting information that remains available, consciously or unconsciously, for later use in their decision-making about available opportunities. These inquiries do something rather than just say something because they enable the speakers to limit their targets’ opportunities through their power over the transaction, rather than through the power of their ideas.

The next section explains how the Court’s modern commercial speech doctrine captures these insights by holding that the First Amendment does not protect commercial speech related to illegal activity, including commercial speech that enables illegal discrimination.

C.    Why the First Amendment Does Not Protect Commercial Speech Related to Illegal Activity, Including Commercial Speech Related to Illegal Discrimination

By prohibiting employers, insurers, housing providers, lenders, and other gatekeepers from denying opportunities and services based on protected class status, antidiscrimination law regulates the use of certain information in determining the terms and conditions of commercial activity (i.e., the exchange of money for labor, credit, housing, insurance, and more). And when legislatures forbid commercial actors from relying on certain characteristics in their decision-making, those actors’ inquiries about candidates’ protected characteristics then constitute commercial speech that is unprotected by the First Amendment because it facilitates illegal commercial activity.105

1.     Commercial speech related to illegal activity more generally

In Virginia State Bd. of Pharmacy v. Virginia Citizens Consumer Council, Inc.,106 the Court held for the first time that the Free Speech Clause provides some protection for commercial speech, striking down Virginia’s law that forbade pharmacists from advertising their prescription drug prices.107 The majority underscored the expression’s value to vulnerable prescription drug consumers like “the poor, the sick and particularly the aged,” observing that those consumers share an “interest in the free flow of commercial information[] that . . . may be as keen, if not keener by far, than [their] interest in the day’s most urgent political debate.”108 In so holding, the Court explained that free speech protections are “enjoyed by the appellees [i.e., the consumers] as recipients of the information, and not solely, if at all, by the advertisers themselves.”109

Shortly thereafter, the Court again described commercial expression’s First Amendment value (and thus its protection from the government’s regulation) as turning primarily on its ability to facilitate listeners’ informed decision-making about legal activities. Under this framework, commercial speech that is false, misleading, or related to illegal activity offers no constitutional value to listeners and is thus unprotected from the government’s regulation, subject only to rational-basis review.110 As the Court explained in Central Hudson Gas & Electric Corp. v. Public Service Commission:

The First Amendment’s concern for commercial speech is based on the informational function of advertising. Consequently, there can be no constitutional objection to the suppression of commercial messages that do not accurately inform the public about lawful activity. The government may ban forms of communication more likely to deceive the public than to inform it, or commercial speech related to illegal activity.111

In contrast, accurate commercial speech about legal activity (like accurate speech about prescription drug prices) is valuable to listeners, and thus the government’s regulation of such speech triggers First Amendment suspicion in the form of heightened—that is, intermediate—scrutiny.112

Although the Court has yet to offer a precise definition of commercial speech, the term includes commercial advertising and other speech that does “no more than propose a commercial transaction.”113 Because the speech that proposes a commercial transaction includes the speech involved in communicating and negotiating the terms and conditions of that transaction,114 the Court has recognized speech other than advertisements as commercial for First Amendment purposes, like speech that communicates the price of goods and services.115 As legal scholar Felix Wu explains, “[w]hat makes speech commercial is the extent to which the speech should be understood to be part of a commercial transaction. Pricing information is quintessential commercial speech, because pricing is a key component of any commercial transaction.”116

Recognizing that the employment relationship is a type of commercial relationship in which a worker exchanges her labor and talent for pay, the Supreme Court has identified job advertisements as “classic examples” of commercial speech.117 Lower courts regularly apply this reasoning to conclude that employers’ recruitment efforts, interviews, and negotiations about the terms and conditions of employment also constitute commercial speech that initiates and completes commercial transactions.118 Along the same lines, gatekeepers’ inquiries about applicants’ protected characteristics—along with gatekeepers’ statements like “White Applicants Only”—take place in the context of communicating and negotiating about potential commercial transactions.119

Again, the Court’s commercial speech doctrine provides that commercial speech is unprotected by the First Amendment when it is false, misleading, or “related to illegal activity.”120 In that case, such speech is entirely open to the government’s regulation subject only to rational-basis scrutiny—as recounted in Part I, the Court has long recognized legislatures’ constitutional power to regulate commercial transactions, which includes their power to prohibit decisionmakers from enforcing discriminatory terms or conditions in providing opportunities and services.121 And when a legislature exercises its constitutional power to prohibit certain commercial activity, speech that facilitates that now-illegal activity loses its First Amendment value to listeners, and thus its constitutional protection. Examples include speech that advertises or inquires about the availability of goods and services that legislatures have prohibited, like drugs and drug paraphernalia, prostitution, and income tax evasion services.122

As an illustration, the production and sale of any particular substance remains legal commercial activity unless and until a legislature chooses to make it illegal. Until that time, advertisements for, inquiries about, and negotiations over the price and availability of that substance constitute commercial speech related to legal activity, with the government’s regulation of such speech subject to intermediate scrutiny. But once a legislature chooses to prohibit the production and sale of that substance (and recall that its regulation of such commercial activities generally triggers only rational-basis scrutiny123 ), advertisements for, inquiries about, and negotiations over the availability of that product then constitute commercial speech that is unprotected by the First Amendment because of its relationship to what is now illegal activity. (To be sure, some listeners very much want to receive such information as potential purchasers of illegal drugs or illegal services—but once the legislature makes that activity illegal, that interest is no longer protected by the Constitution.)

2.     Commercial speech related to illegal discrimination more specifically

Along the same lines, a characteristic does not become “protected” from private parties’ discrimination as a legal matter unless and until a legislature passes a statute prohibiting gatekeepers from relying on that characteristic in their decision-making. For example, gatekeepers’ discriminatory reliance on pregnancy (or disability or religion or salary history, etc.) in their decision-making does not become illegal unless and until a legislature enacts a statute to that effect.124 Upon such a statute’s enactment, however, gatekeepers’ inquiries about candidates’ protected class status then constitute commercial speech that is unprotected by the First Amendment because they relate to—that is, enable—the now-illegal activity of relying on those characteristics when making key decisions.

Indeed, in Central Hudson itself, the Court offered gatekeepers’ speech that enables illegal job discrimination as an illustration of commercial speech unprotected by the First Amendment because of its relationship to illegal commercial activity.125 More specifically, it cited its holding in Pittsburgh Press Co. v. Pittsburgh Comm’n on Human Relations, a decision rejecting a First Amendment challenge to a local antidiscrimination law that not only prohibited sex-based employment decisions, but also prohibited gatekeepers’ publication of “any notice or advertisement relating to ‘employment’ or membership which indicates any discrimination because of . . . sex.”126 The Pittsburgh Press Court held that sex-segregated job advertisements constituted unprotected commercial speech because they proposed the illegal commercial transaction of discriminatory hiring. In so holding, the Court analogized the contested job listings (which consisted of columns headed “Jobs—Male Interest” and “Jobs—Female Interest”) to constitutionally unprotected advertisements for illegal drugs or prostitution.127 As it explained, “Discrimination in employment is not only commercial activity, it is illegal commercial activity under the Ordinance . . . . The advertisements, as embroidered by their placement, signaled that the advertisers were likely to show an illegal sex preference in their hiring decisions. Any First Amendment interest which might be served by advertising an ordinary commercial proposal and which might arguably outweigh the governmental interest supporting the regulation is altogether absent when the commercial activity itself is illegal and the restriction on advertising is incidental to a valid limitation on economic activity.”128 In other words, advertising that “I’ve got a job for a man” or stating that “Only whites need apply” is just as related to illegal activity for commercial speech purposes as advertising that “I’ve got cocaine for sale.” All of these statements facilitate illegal commercial transactions. All thus do something, and not just say something.

So too is the case of gatekeepers’ inquiries about candidates’ religion, sexual orientation, pregnancy, prior pay, credit history, or other characteristics protected from discrimination by the relevant jurisdiction. Asking an applicant if she’s pregnant (or HIV-positive, or Muslim) is not meaningfully distinguishable for these purposes from saying “No pregnant [or HIV-positive, or Muslim] people need apply,” as the query deters applicants based on protected class status and elicits information that facilitates illegal decision-making.

The doctrinal recognition that the First Amendment does not protect commercial speech related to illegal activity thus separates the government’s constitutionally permissible interest in regulating commercial transactions from the government’s constitutionally impermissible interest in censoring a message it disfavors. This insight also explains why laws regulating gatekeepers’ speech that enables illegal discrimination (like laws regulating commercial speech related to illegal activity more broadly) do not trigger First Amendment scrutiny even though they target certain speech by certain speakers.129 As the Court has recognized, the First Amendment permits these distinctions because only certain speakers have the power to engage in the conduct that the legislature has regulated. In other words, only employers and other gatekeepers have the power to make illegally discriminatory decisions, and only some of their inquiries and statements enable that illegal conduct.130

For related reasons, the Court’s commercial speech framework also explains why its decision in Sorrell is inapposite to antidiscrimination laws that restrict gatekeepers’ inquiries about candidates’ protected class status. Recall that Sorrell held unconstitutional a Vermont law that restricted the transmission of specific information (individual doctors’ prescribing practices) to prevent that information’s use in disfavored but legal choices (marketing brand-name pharmaceuticals to individual doctors).131 Contrast antidiscrimination laws that instead restrict gatekeepers’ questions that elicit specific information about individual candidates’ now-protected characteristics to prevent that information’s use in illegally discriminatory conduct.132

Recall too Daniel Farber’s proposal for parsing the government’s permissible targeting of speech for its contractual functions from its impermissible targeting of speech because of the ideas expressed. We can be confident that the former is at work if the government’s regulatory interest in those statements or inquiries disappears when made by those who are not parties to a potential transaction.133 The antidiscrimination provisions discussed herein apply only to speech by one party to a potential job, housing, or other transaction about the terms of that transaction because only that party has the power to engage in the regulated conduct. In other words, the government regulates these inquiries because they do something and not just say something.134

Contrast inquiries by a speaker who does not hold power over the listener: such inquiries that “do not accomplish a significant change in normative relations or other aspects of the listener’s environment” because they are “not accompanied by inducements or threats or made in circumstances where a positive response is obligatory.”135 Think, for instance, of how the government’s antidiscrimination interest in questions about an applicant’s pregnancy (or disability or religion or salary history or other protected characteristic) evaporates when the question is asked by a friend or neighbor rather than by an employer or other transacting party. For these reasons, gatekeepers’ statements or inquiries that are not “in connection with” or “with respect to” a regulated transaction do not implicate the government’s interest in the enforcement of antidiscrimination law, and thus these laws appropriately do not extend to communications outside of the transactional context.136 For instance, the Equal Credit Opportunity Act’s regulations limit only inquiries into protected class status made “in connection with a credit transaction,”137 Title VII regulations address similar inquiries only “in connection with prospective employment,”138 and the Fair Housing Act prohibits discriminatory statements “with respect to the sale or rental of a dwelling.”139 Gatekeepers (and everybody else) remain free to express any political, moral, religious, or other opinion outside the transactional context through letters to the editor, testimony, lobbying, and more. As the Court emphasized in Pittsburgh Press, “Nothing in our holding allows government at any level to forbid Pittsburgh Press to publish and distribute advertisements commenting on the Ordinance, the enforcement practices of the Commission, or the propriety of sex preferences in employment.”140

Some may contest the closeness of the relationship between gatekeepers’ inquiries about applicants’ protected characteristics and gatekeepers’ illegal reliance on those characteristics.141 For instance, some may argue that asking an applicant about her religion or whether she has a disability does not carry the same deterrent effect as saying “No Jews” or “No folks with disabilities need apply”—or that asking an applicant about her age or salary history does not mean that the gatekeeper will rely on her answer to make hiring and compensation decisions.142 (Note, however, that the challengers to Philadelphia’s salary history law acknowledged that they sought to rely on those answers to make hiring and compensation decisions.143 ) But the Court has never required that commercial speech related to illegal activity lead inevitably and only to that activity to lose First Amendment protection. Consider Pittsburgh Press, where the defendant argued that because sex-segregated advertisements did not expressly deny employment to women, they were not inevitably, and thus sufficiently, related to illegal discrimination to lose First Amendment protection.144 The Court rejected this argument, emphasizing that listing job openings in sex-segregated columns signaled that employers were “likely” to discriminate and thus would deter at least some women from applying for male-designated jobs (and vice versa).145 So too do gatekeepers’ inquiries about candidates’ protected class status signal that the answers are likely to influence gatekeepers’ choices and deter some applicants—especially when we recall that once a jurisdiction has prohibited reliance on pregnancy or other characteristics in commercial transactions, there’s no constitutional value in commercial actors’ inquiries about those characteristics.

Indeed, both theory and doctrine have long recognized that the First Amendment provides no protection to the speech that enables illegal conduct even if that speech does not always accomplish such conduct. Speech that solicits, or conspires to engage in, illegal conduct is not protected by the First Amendment even though it doesn’t always lead to illegal conduct, as the solicitation may be rejected or the conspiracy may not succeed.146 For instance, the First Amendment does not protect A’s inquiry as to whether B has cocaine for sale or if B would be willing to eliminate A’s enemy for a certain price—even if B declines A’s offer or fails to deliver on a promised exchange. What matters is that those inquiries are likely to accomplish illegal conduct. For the same reason, the First Amendment does not protect gatekeepers’ statement “White Applicants Only;” it is likely to deter nonwhite applicants even though it may not always succeed in so doing.

Gatekeepers’ inquiries about applicants’ protected class status are especially likely to enable illegal discrimination (and thus lose First Amendment protection) when they do not elicit information that is valuable apart from its ability to inform illegal discrimination, or when that information is available through other means or in other settings that do not threaten to infect gatekeepers’ decision-making about specific candidates on illegal bases. Recall, for, example, that the challengers to Philadelphia’s law argued that salary history inquiries not only informed their hiring and compensation decisions, but also permitted them to identify applicants with unaffordable salary expectations and to learn about prevailing pay scales for certain jobs.147 But employers can and do obtain more accurate information about the market for wages through other, aggregate sources outside of negotiations with a specific applicant for a specific transaction.148 And employers can learn whether they can afford a specific applicant simply by telling her the job’s salary or by asking for her salary expectations—just as the ADA permits employers to ask applicants if they can perform a job’s functions with or without a reasonable accommodation while forbidding employers from asking applicants whether they have a disability.149

In sum, legislatures regulate commercial activity when they prohibit commercial actors from relying on certain characteristics in their decision-making, and those actors’ inquiries about candidates’ protected characteristics then constitute commercial speech that is unprotected by the First Amendment because it facilitates illegal commercial activity—in other words, because it does something and not just says something.

III. The First Amendment Implications of Other Antidiscrimination Provisions That Regulate Commercial Parties’ Speech

This Part briefly considers the commercial speech framework’s application to antidiscrimination provisions that regulate commercial speech in ways other than those discussed in Parts I and II—in other words, in ways apart from forbidding gatekeepers’ discriminatory statements of preference and inquiries about candidates’ protected class status when reliance on the answer is illegal. As we’ll see, some statutes prohibit decisionmakers’ inquiries about applicants’ characteristics without forbidding decisionmakers from relying on those characteristics in their decision-making. Other statutes require decisionmakers to disclose certain accurate information about the terms and conditions of available opportunities. Finally, some statutes forbid gatekeepers’ reliance on certain protected characteristics for some reasons and not others, and thus forbid gatekeepers’ inquiries for some purposes and not others.

An exhaustive treatment of these statutes is beyond the scope of this Article.150 Here, I simply show how the Court’s longstanding commercial speech doctrine again provides the relevant analysis. Recall that this doctrine exemplifies a listener-centered approach to certain First Amendment problems by protecting commercial speech that furthers listeners’ interests (like accurate commercial speech about lawful activity) while permitting the regulation of commercial speech that frustrates those interests (like false or misleading commercial speech, or commercial speech related to illegal activity)—in other words, by privileging listeners’ interests over commercial actors’ interests as speakers when their interests collide.151 The Court’s commercial speech doctrine itself thus relies on speaker- and content-based distinctions precisely because those distinctions are relevant to commercial expression’s potential for First Amendment harm and First Amendment value.152

A.    Antidiscrimination Laws That Regulate Decisionmakers’ Inquiries About Certain Characteristics Without Prohibiting Reliance On Those Characteristics

First, some laws prohibit or delay gatekeepers’ inquiries about certain characteristics without ultimately prohibiting gatekeepers’ reliance on those characteristics. In other words, sometimes legislatures block (or delay) gatekeepers’ inquiries to candidates about certain characteristics when gatekeepers’ use of that information is not illegal. For example, some states and localities have enacted “ban-the-box” laws that “generally prohibit employers from inquiring about a job applicant’s criminal record until later in the hiring process, such as after an initial interview or once a conditional employment offer is made” in hopes that employers will be more likely to hire qualified ex-offenders if they assess candidates before learning of any criminal record.153

Because gatekeepers’ inquiries about characteristics that are not protected from discrimination do not enable illegal activity, they do not fall within Central Hudson’s categories of commercial speech that are entirely unprotected by the First Amendment. This means that the government’s restrictions of such inquiries must satisfy intermediate scrutiny. Recall Central Hudson’s holding that the government’s regulation of accurate commercial speech related to legal activity triggers a form of intermediate scrutiny because that speech has constitutional value for listeners:

The State must assert a substantial interest to be achieved by restrictions on commercial speech. Moreover, the regulatory technique must be in proportion to that interest. The limitation on expression must be designed carefully to achieve the State’s goal. Compliance with this requirement may be measured by two criteria. First, the restriction must directly advance the state interest involved; the regulation may not be sustained if it provides only ineffective or remote support for the government’s purpose. Second, if the governmental interest could be served as well by a more limited restriction on commercial speech, the excessive restrictions cannot survive.154

In assessing whether the government’s means directly advances its ends, the Court has applied Central Hudson intermediate scrutiny to permit the government to rely on “studies[,] anecdotes[,] . . . history, consensus, and ‘simple common sense’” to justify its choice, emphasizing that the standard requires “a reasonable,” rather than a perfect, fit.155 Relatedly, the Court has also declined to require the government’s regulation to be the “least restrictive” alternative, but instead requires “a fit that is not necessarily perfect, but reasonable; that represents not necessarily the single best disposition but one whose scope is ‘in proportion to the interest served,’ that employs not necessarily the least restrictive means but . . . a means narrowly tailored to achieve the desired objective.”156 In other words, in these settings the government’s regulation does trigger First Amendment scrutiny. Nevertheless, appropriately designed antidiscrimination provisions that delay or block gatekeepers’ access to certain information about candidates where reliance on that information is not directly prohibited may survive such scrutiny.157

B.    Antidiscrimination Laws That Require or Permit Certain Disclosures

Next, antidiscrimination laws sometimes require employers, housing providers, lenders, insurers, and other commercial actors to make certain accurate disclosures to expose or deter discrimination, or to achieve other equality goals. For example, as part of their efforts to ameliorate stubborn and unjustified pay disparities, some legislatures have enacted laws that require employers to disclose their pay scales and practices.158 These measures seek to address asymmetries in information about pay, where employers know what they pay their own workers but workers generally don’t know what their colleagues are paid.159 As legal scholar Sylvia Law observes, “[e]very story of a successful challenge to the gender wage gap begins with a woman discovering that she is earning less than a male colleague who does similar, or less demanding, work.”160 Other examples of required disclosures for antidiscrimination purposes include laws that require employers and other gatekeepers to disclose truthful information about applicants’ legal rights.161

These sorts of disclosures have a long pedigree throughout the commercial speech context more broadly, where the government routinely requires commercial actors to make certain accurate disclosures to inform and further listeners’ decision-making.162 Consumer protection law and securities law, for example, rely on an array of information-forcing mechanisms to address informational asymmetries between speakers and their listeners.163

Again, the Court’s commercial speech doctrine supplies the relevant First Amendment analysis. As it explained in Zauderer v. Office of Disciplinary Counsel, “the extension of First Amendment protection to commercial speech is justified principally by the value to consumers of the information such speech provides.”164 For this reason, the Court has applied only deferential review to laws requiring commercial speakers to make accurate disclosures to their listeners to protect those listeners from deception, upholding such requirements when they are “reasonably related to the State’s interest in preventing deception of consumers.”165 Lower courts have also often applied this deferential review to disclosure requirements intended to inform consumers even when the regulated commercial speakers have not engaged in deception.166 The sorts of antidiscrimination laws described above167 that require truthful disclosures by commercial actors will generally survive this review.

Despite its more recent antiregulatory turn, the Court has yet to repudiate Zauderer’s deferential review as applied to required commercial disclosures.168 In any event, the disclosures described above can also satisfy Central Hudson’s intermediate scrutiny standard. As I’ve written elsewhere, “[G]overnment requirements that employers disclose truthful information about workers’ rights and other working conditions can provide considerable value to workers as listeners while imposing little, if any, expressive costs. They thus can readily satisfy not only rational-basis scrutiny but also intermediate or even exacting scrutiny when appropriately drafted to achieve the government’s strong interest in informing and protecting workers.”169

Relatedly, note that some antidiscrimination laws that forbid gatekeepers from asking candidates about their protected class status nevertheless sometimes permit candidates to disclose that status to achieve equal opportunity. Think, for example, of the Americans with Disabilities Act, which forbids employers from inquiring into workers’ disability status while permitting—indeed, encouraging—workers to disclose their disability status to explore possibilities for reasonable accommodations.170 Think too of laws that protect workers from their employers’ punishment for sharing their salary information with other workers.171 For purposes of First Amendment analysis, gatekeepers’ inquiries about disability or other protected characteristics are distinguishable from candidates’ disclosure of those characteristics when the former are related to illegal discrimination while the latter enable reasonable accommodation and other equality goals.172 These measures change the dynamic from one where gatekeepers have all the information and power to one where applicants have some too. As noted above, these sorts of measures to address informational asymmetries between transactional parties have a long pedigree in the commercial speech context.173

C.    Antidiscrimination Laws That Permit Gatekeepers to Collect (And Sometimes Rely on) Information About Protected Characteristics

Finally, some antidiscrimination laws permit gatekeepers to collect information about applicants’ protected class status in certain circumstances to achieve equality objectives. More specifically, some antidiscrimination laws permit gatekeepers to collect data about applicants’ protected characteristics to assess the success of their equal opportunity efforts or to determine whether their selection practices have an illegally disparate impact. For example, Title VII (unlike some other antidiscrimination statutes) specifically forbids disparate impact discrimination in addition to intentional discrimination,174 and the Supreme Court has interpreted Title VII to permit employers to consider candidates’ race or gender as part of an affirmative action plan so long as the plan’s purpose mirrors that of Title VII and does not unnecessarily trammel the rights of nonbeneficiaries.175 For this reason, the EEOC explains:

Employers may legitimately need information about their employees[’] or applicant[’s] race for affirmative action purposes and/or to track applicant flow [for purposes of complying with Title VII’s disparate impact provisions]. One way to obtain racial information and simultaneously guard against discriminatory selection is for employers to use separate forms or otherwise keep the information about an applicant’s race separate from the application. In that way, the employer can capture the information it needs but ensure that it is not used in the selection decision.176

IV. Conclusion

Antidiscrimination law regulates commercial conduct when it prohibits gatekeepers from relying on certain characteristics in setting the terms and conditions of employment and other transactions. As theory and doctrine both make clear, the First Amendment permits the government to restrict the speech that initiates or accomplishes this conduct—that is, speech that does something and not just says something. More specifically, this includes commercial actors’ speech that enables illegally discriminatory transactions, such as gatekeepers’ statements like “White Applicants Only” as well as inquiries about candidates’ protected class status. Because these inquiries enable illegal discrimination by deterring candidates based on their protected class status and by eliciting the information that facilitates gatekeepers’ discriminatory decisions, the First Amendment poses no bar to the government’s regulation of them.

• 1I explored related issues in earlier work. Helen Norton, You Can’t Ask (or Say) That: The First Amendment and Civil Rights Restrictions on Decisionmaker Speech, 11 Wm. & Mary Bill Rts. J. 727 (2003). As this Article explains, a great deal has since changed. Among other things, legislatures increasingly regulate gatekeepers’ reliance on and inquiries about certain characteristics to achieve equality and other public welfare goals. See infra notes  NOTEREF _Ref30840012 \h 22 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800340030003000310032000000 – NOTEREF _Ref37412492  \* MERGEFORMAT 52 and accompanying text. And the antiregulatory turn in First Amendment law increasingly inspires certain litigants to attack related efforts. See infra notes  NOTEREF _Ref30884185 \h 60 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380034003100380035000000 –72 and accompanying text.
• 2In this Article, I use the terms “gatekeepers” and “decisionmakers” interchangeably to describe those individuals and institutions empowered to select among applicants for important opportunities and services.
• 3In this Article, I use the terms “protected characteristic” and “protected class status” interchangeably to refer to attributes that a legislature has protected from discrimination by forbidding gatekeepers from relying on those attributes when distributing important opportunities and services.
• 4See infra notes  NOTEREF _Ref30884185 \h 60 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380034003100380035000000 – NOTEREF _Ref37414278 \h 72 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330037003400310034003200370038000000 and accompanying text for a discussion of this turn.
• 5See Kent Greenawalt, Fighting Words: Individuals, Communities, and Liberties of Speech 6 (1995); Kent Greenawalt, Speech, Crime, and the Uses of Language 59 (1989).
• 6Rumsfeld v. Forum for Acad. & Inst. Rights, 547 U.S. 47, 66 (2006); see also Sorrell v. IMS Health Inc., 564 U.S. 552 (2011) (offering “White Applicants Only” as an example of speech that is unprotected by the First Amendment as incidental to illegal conduct).
• 7See infra notes  NOTEREF _Ref30791653 \h 8 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003700390031003600350033000000 – NOTEREF _Ref37412958  \* MERGEFORMAT 21 and accompanying text.
• 8See Ignacio N. Cofone, Antidiscriminatory Privacy, 72 SMU L. Rev. 139, 149–51 (2019) (describing when and how gatekeepers’ access to information about candidates’ protected class status fosters discrimination); Jessica L. Roberts, Protecting Privacy to Prevent Discrimination, 56 Wm. & Mary L. Rev. 2097, 2143–46 (2015) (offering examples of how decisionmakers have used previously unknown information about applicants’ age, religion, national origin, disability, gender at birth, and other protected characteristics to discriminate against those applicants).
• 9Benjamin Edelman, Michael Luca & Dan Svirsky, Racial Discrimination in the Sharing Economy: Evidence From a Field Experiment, 9 Am. Econ. J. Applied Econ. 1, 1–2 (2017) (finding that prospective guests “with distinctively African-American names are 16 percent less likely to be accepted relative to identical guests with distinctively white names”); see also Olivier Sylvain, Discriminatory Designs on User Data 13–14 (2018) (describing how Airbnb elicited information from prospective guests that permitted prospective “hosts” to rely on “illicit biases—against, say, Latinos or blacks—that do not accurately predict a prospective guest’s reliability as a tenant. In this way Airbnb’s service directly reinforces discrimination when it requires users to share information that suggests their own race”).
• 10See Settlement Agreement and Release, Nat’l Fair Hous. All. v. Facebook, Inc., No. 18 Civ. 2689 (S.D.N.Y. 2018), https://nationalfairhousing.org/wp-content/uploads/2019/03/FINAL-S‌IG‌N‌E‌D‌-‌N‌F‌H‌A‌-‌F‌B-Settlement-Agreement-00368652x9CCC2.pdf [https://perma.cc/MRJ7-4LE6]; Summary of Settlements Between Civil Rights Advocates and Facebook, Am. Civil Liberties Union (Mar. 19, 2019), https://www.aclu.org/sites/default/files/field_docum‌e‌n‌t/‌3‌.‌1‌8‌.‌2‌0‌1‌9‌_‌j‌o‌‌in‌t‌_‌s‌t‌a‌t‌e‌m‌e‌n‌t‌_‌‌fi‌nal_0.pdf [https://perma.cc/K93A-XAZH].
• 11H.R. Rep. No. 101-485, pt. 2, at 72 (1990).
• 12See Roberts, supra note  NOTEREF _Ref30791653 \h  \* MERGEFORMAT 8 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003700390031003600350033000000 , at 2122 (“If an employer cannot access a particular kind of information, she cannot discriminate on the basis of that information. However, once an employer acquires the ability to discriminate, the knowledge of an employee’s protected status may influence the employer’s decisions in conscious, as well as unconscious, ways.”).
• 13See Ian Ayres & Peter Siegelman, The Q-Word as Red Herring: Why Disparate Impact Liability Does Not Induce Hiring Quotas, 74 Tex. L. Rev. 1487, 1492 (1996) (“In the absence of an obvious motive or a relevant comparison group, potential plaintiffs have a difficult time recognizing that disparate treatment in hiring has occurred, let alone convincing a court of that fact.”).
• 14See, e.g., Ellen Berrey, Robert L. Nelson, & Laura Beth Nielsen, Rights on Trial: How Workplace Discrimination Law Perpetuates Inequality 13 (2017) (documenting how and why “only a tiny fraction of possible targets of workplace discrimination take formal action [and when they do] they are likely to settle or lose”); Charlotte S. Alexander, #MeToo and the Litigation Funnel, 23 Emp. Rts. & Emp. Pol’y J. 17 (2019) (documenting plaintiffs’ difficulties in winning claims under Title VII); Kevin M. Clermont & Stewart J. Schwab, Employment Discrimination Plaintiffs in Federal Court: From Bad to Worse?, 3 Harv. L. & Pol’y Rev. 103 (2009) (finding that Title VII plaintiffs who file in federal court are less successful than plaintiffs in other types of cases); Katie R. Eyer, That’s Not Discrimination: American Beliefs and the Limits of Anti-Discrimination Law, 96 Minn. L. Rev. 1275, 1282–83 (2012) (“Indeed, of every 100 discrimination plaintiffs who litigate their claims to conclusion (i.e., do not settle or voluntarily dismiss their claims), only 4 achieve any form (de minimis or not) of relief.”); Michael Selmi, Public vs. Private Enforcement of Civil Rights: The Case of Housing and Employment, 45 UCLA L. Rev. 1401, 1409–10 (1998) (explaining “the ironies of a complaint-based [approach to civil rights enforcement], namely that many, perhaps even a majority, of discrimination claims are missed because the discrimination occurs in the contract formation when claims are significantly less likely to be filed”).
• 15Cynthia Estlund, Truth, Lies, and Power at Work, 101 Minn. L. Rev. Headnotes 349, 349 (2016).
• 16Cofone, supra note  NOTEREF _Ref30791653 \h  \* MERGEFORMAT 8 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003700390031003600350033000000 , at 140; see also Lior J. Strahilevitz, Privacy Versus Antidiscrimination, 75 U. Chi. L. Rev. 363, 374 (2008) (“Information-based antidiscrimination policies will be most effective at combating statistical discrimination when traditional enforcement methods are least effective.”).
• 17See Adam M. Samaha & Lior J. Strahilevitz, Don’t Ask, Must Tell—And Other Combinations, 103 Calif. L. Rev. 919, 969 (2015) (“Don’t Ask, May Tell examples are often linked to one-sided worries about the vulnerability of respondents to questioner power.”); id. at 938 (“[O]ne simple reason for Don’t Ask is to prevent unwelcome pressure.”).
• 18Barbano v. Madison Cty., 922 F.2d 139, 141 (2d Cir. 1990) (concluding that the hiring committee’s acquiescence to these questions supported the conclusion that the employer had illegally discriminated on the basis of pregnancy in its hiring decision).
• 19Garlitz v. Alpena Reg’l Med. Ctr., 834 F. Supp. 2d 668, 679 (E.D. Mich. 2011).
• 20See Samaha & Strahilevitz, supra note  NOTEREF _Ref30792716 \h  \* MERGEFORMAT 17 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003700390032003700310036000000 , at 926 (“Questions are themselves telling, in the sense that statements correctly formulated as questions usually reveal something about the questioner’s interests or beliefs.”); id. at 929 (“A question is a special device for information collection: it is an interactive call for information that alerts an audience to the collection effort and that usually reveals something about the questioner . . . . Questions reveal somebody’s interest in and comfort with additional information on a given topic . . . .”).
• 21See Fredenburg v. Contra Costa Cty. Dep’t of Health, 172 F.3d 1176, 1182 (9th Cir. 1999) (explaining that the ADA’s restriction on pre-employment medical inquiries and examinations “prevents employers from using HIV tests to deter HIV-positive applicants from applying”); see also Scott Skinner-Thompson, AIDS and the Law 3–79 (6th ed. 2020) (explaining how the ADA protects applicants from having to disclose private medical information that makes them “vulnerable to discrimination”).
• 22See, e.g., Kimani Paul-Emile, Beyond Title VII: Rethinking Race, Ex-Offender Status, and Employment Discrimination in the Information Age, 100 Va. L. Rev. 893, 936 (2014) (“[The ADA was] designed to prohibit discrimination in employment preemptively. The ADA focuses on regulating the transmission of potentially stigmatizing data during the hiring phase because, as studies have found, the most common form of discrimination against individuals with disabilities is the denial of a job for which the individual is qualified, followed by the refusal of an interview on the basis of a disability.”).
• 23As Part III discusses, legislatures can and do make different choices when drafting antidiscrimination laws. See infra notes  NOTEREF _Ref53245961 \h 151 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003200340035003900360031000000 – NOTEREF _Ref37413193  \* MERGEFORMAT 177 and accompanying text.
• 24Unless and until a statute provides otherwise, the default rule in American jurisdictions permits employers to ask whatever they wish of applicants. Other countries choose different default rules. See Matthew W. Finkin, Pay Privacy in Comparative Context, 22 Emp. Rts. & Emp. Pol’y J. 355, 368 (2018) (“In Germany, out of concern for employee privacy, the employer bears the burden to prove the question is necessary under a strict standard of relatedness to job qualification. In America, out of concern for managerial liberty, the state bears the burden to prove the restriction is necessary to further a specific public end grounded in labor market outcomes.”).
• 2543 Pa. Stat. and Cons. Stat. Ann. § 955(b)(1) (West 2019). For a few of the many other examples, see Alaska Stat. Ann. § 18.80.240(3) (West 2019) (prohibiting landlords and real estate agents from making “a written or oral inquiry or record of the sex, marital status, changes in marital status, race, religion, physical or mental disability, color, or national origin of a person seeking to buy, lease, or rent real property”); Colo. Rev. Stat. § 24-34-502(1)(a) (2019) (same); La. Rev. Stat. Ann. § 46:2254(C)(6) (2019) (prohibiting owners or others engaging in a real estate transaction from making “a record or inquiry in connection with a prospective real estate transaction, which indicates, directly or indirectly, an intent to make a limitation, specification, or discrimination” based on an unrelated disability); Me. Rev. Stat. Ann. tit. 5, § 4581-A(1)(A) (2019) (prohibiting landlords, owners or agents who are renting or selling housing from making “any written or oral inquiry concerning the race or color, sex, sexual orientation, physical or mental disability, religion, ancestry, national origin or familial status of any prospective purchaser, occupant or tenant”); Neb. Rev. Stat. § 20-318(5) (2019) (making it unlawful to “cause to be made any written or oral inquiry or record concerning the race, color, religion, national origin, handicap, familial status, or sex of a person seeking to purchase, rent, or lease any housing”); N.J. Stat. Ann. § 10:5-12(c) (West 2019) (prohibiting employers’ inquiries into “race, creed, color, national origin, ancestry, age, marital status, civil union status, domestic partnership status, affectional or sexual orientation, gender identity or expression, disability, nationality, pregnancy or breastfeeding, or sex” or military status); Or. Rev. Stat. Ann. § 659A.030(1)(d) (West 2019) (same).
• 26Civil Rights Act of 1964, § 701(k), as amended, 42 U.S.C.A. § 2000e(k).
• 2742 U.S.C. § 2000e(k) (2012).
• 2829 C.F.R. § 1604.7 (2019) (“Any pre-employment inquiry in connection with prospective employment which expresses directly or indirectly any limitation, specification, or discrimination as to sex shall be unlawful unless based upon a bona fide occupational qualification.”); see also King v. Trans World Airlines, Inc., 738 F.2d 255, 258 n.2 (8th Cir. 1984) (“[Q]uestions about pregnancy and childbearing would be unlawful per se [under Title VII] in the absence of a bona fide occupational qualification.”); Snyder v. Yellow Transp., Inc. 321 F. Supp. 2d 1127, 1131 (E.D. Mo. 2004) (denying defendant employer’s motion for summary judgment in light of evidence that it had asked questions about the plaintiff’s marital status, parental status, and plans to have children, questions that constituted a per se Title VII violation).
• 2912 C.F.R. § 1002.5(b) (2019) (“A creditor shall not inquire about the race, color, religion, national origin, or sex of an applicant or any other person in connection with a credit transaction.”); 12 C.F.R. § 1002.5(d)(1) (2019) (prohibiting inquiries into the marital status of applicants for certain types of credit); 12 C.F.R. § 1002.5(b)(2) (2019) (prohibiting inquiries into applicants’ sex); 12 C.F.R. § 1002.5(d)(3) (2019) (prohibiting inquiries into applicants’ “birth control practices, intentions concerning the bearing or rearing of children, or capability to bear children”).
• 3042 U.S.C. § 12112(d)(2) (2012).
• 3142 U.S.C. § 12112(d)(3) (2012). An employer may then rescind an individual’s conditional offer only when the exclusionary decision is job-related and consistent with business necessity. 42 U.S.C. § 12112(b)(6) (2012).
• 3242 U.S.C. § 12112(d)(4)(A) (2012).
• 3342 U.S.C. § 2000ff-1(b) (2012) (limiting employers’ ability to “request, require, or purchase genetic information” of potential employees and current employees or their family members); see also id. (defining genetic information to include genetic test results for applicants and their family members as well as family medical history); Conn. Gen. Stat. Ann. § 46a-60(b)(11) (West 2019) (prohibiting employers’ inquiry into applicants’ genetic information); Del. Code Ann. tit. 19, § 711(e) (2019) (same).
• 34Paul-Emile, supra note  NOTEREF _Ref30840012 \h  \* MERGEFORMAT 22 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800340030003000310032000000 , at 937; see also id. at 938 (“This provision includes searches of court records and medical databases. Although the law outlines certain limited exceptions, including inadvertent acquisition, the EEOC regulations emphasize that receipt of genetic information will not generally be considered inadvertent unless the employer instructs the source of the material to exclude genetic information. The law also includes safe harbor language for commercial or publicly available information; however, covered employers are precluded from searching such sources with the intention of acquiring an individual’s genetic information.”).
• 35See Minn. Stat. § 363A.08 subd. 4(2) (2017) (prohibiting an employer from “seek[ing] and obtain[ing] for purposes of making a job decision, information from any source that pertains to” the applicant’s protected characteristics).
• 36See Jessica L. Roberts, Preempting Discrimination: Lessons from the Genetic Information Nondiscrimination Act, 63 Vand. L. Rev. 439, 476 (2010) (“When invoked within antidiscrimination law, immutability stands for the proposition that entities should not discriminate on the basis of traits that a person did not choose and cannot change or control without serious cost.”).
• 37E.g., Cal. Gov’t Code § 12940(a), (d) (West 2019) (prohibiting reliance on, and inquiries into, an applicant’s or employee’s sexual orientation, marital status, and other protected characteristics); Colo. Rev. Stat. § 24-34-402(1)(a), (d) (2017) (prohibiting reliance on and inquiries into an applicant’s sexual orientation); Haw. Rev. Stat. Ann. § 378-2(a)(1)(A), (C) (West 2019) (prohibiting reliance on and inquiries into an applicant’s sexual orientation and gender identity); Mass. Ann. Laws Ann. ch. 151B, § 4(1), (3) (West 2018) (same); Minn. Stat. §§ 363A.03 subd. 44, 363A.08 subd. 2, subd. 4(a)(1) (2017) (same); 28 R.I. Gen. Laws § 28-5-7(1)(i), (ii), (4)(i), (iii) (2017) (same).
• 38See Roberts, supra note  NOTEREF _Ref31138747 \h  \* MERGEFORMAT 36 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330031003100330038003700340037000000 , at 441 (“[GINA’s proponents presented the legislation] as a unique opportunity to stop discrimination before it starts. It is this preemptive nature, basing protection on future—rather than past or even present—discrimination, that truly makes GINA novel.”); see also id. at 472–73 (“[T]he fear of genetic-information discrimination was preventing many potential research subjects from participating in studies, thereby slowing the rate at which genetic technology could progress” and “supporters of genetic antidiscrimination legislation also maintained that the fear of genetic tests was harming the general public—people were not seeking diagnoses and treatments that could improve or sustain their health. For example, one-third of the women offered a genetic test related to breast cancer declined, citing potential discrimination as the reason.”).
• 39E.g., Alaska Stat. Ann. § 18.80.220(a)(3) (West 2019) (forbidding employers’ inquiries into employees’ marital status); Cal. Gov’t Code § 12940(d) (West 2019) (same); Del. Code Ann. tit. 19, § 711(j) (2016) (prohibiting employers from discriminating against applicants or employees based on their reproductive health decisions); St. Louis, Mo., Ordinance 70459 (Feb. 1, 2017) (prohibiting employers’ reliance on and inquiries into applicants’ “reproductive health decisions or pregnancy”).
• 40E.g., D.C. Code § 32-1362 (2012) (prohibiting employers from discriminating against applicants because of their current unemployment); N.Y.C., N.Y., Code § 8-107(21) (2019) (same).
• 41E.g., Or. Rev. Stat. Ann. § 659A.320 (West 2018) (prohibiting employers’ inquiries into applicants’ credit history).
• 42E.g., Md. Code Ann., Ins. § 27-504 (West 2014) (prohibiting insurers from discriminating against applicants because they have been victims of domestic violence); N.Y. Real. Prop. Law § 227-d (McKinney 2016) (prohibiting housing providers from discriminating against applicants because they have been victims of domestic violence).
• 43E.g., Haw. Rev. Stat. Ann. § 378-2(a)(1)(A), (C) (West 2019) (prohibiting reliance on and inquiries into applicants’ arrest records); Or. Rev. Stat. Ann. § 659A.030(1)(a), (b), (d) (West 2017) (prohibiting reliance on and inquiries into an applicant’s expunged juvenile record).
• 44E.g., 38 U.S.C. § 4311 (2012) (prohibiting employers from discriminating against members of the uniformed services by relying on military status when making employment decisions).
• 45See Dallan F. Flake, Do Ban-the-Box Laws Really Work?, 104 Iowa L. Rev. 1079, 1082 (2019) (explaining that “ban-the-box” laws seek to address the “grim situation [that] has emerged in which the very people who most need to work—both for their own benefit and for the benefit of society as a whole—often experience tremendous difficulty finding gainful employment”); see also Genevieve Douglas, ‘Ban the Box’ Laws Finding Inroads in Red States, Too, Bloomberg Law News (Dec. 3, 2019), https://news.bloomberglaw.com/daily-labor-report/ban-the-box-laws-finding-inroads-in-red-states-too [https://perma.cc/TB7T-A5ZY] (“Nearly one in three adults in the U.S. has an arrest or conviction record that can show up on an employment background check, according to the National Employment Law Project. That makes the potential impact on the labor market huge for more widespread ban the box measures.”).
• 46See Orly Lobel, Knowledge Pays: Reversing Information Flows & The Future of Pay Equity, 120 Colum. L. Rev. 547, 553 (2020) (“According to the latest report from the U.S. Census Bureau, American women still earn an average of 80 to 83 cents for every dollar earned by their male counterparts.”). The pay gap is even greater for women of color. Id.
• 47E.g., Cal. Lab. Code § 432.3(a), (b) (West 2019) (prohibiting reliance on and inquiries into an applicant’s “salary history information”); Del. Code Ann. tit. 19, § 709B(b) (West 2017) (prohibiting reliance on and inquiries into an applicant’s “compensation history”); Haw. Rev. Stat. Ann. § 378-2.4(a) (West 2019) (prohibiting reliance and inquiries into an applicant’s “salary history”); Mass. Gen. Laws Ann. ch. 149, § 105A(c)(2) (West 2018) (prohibiting reliance on and inquiries into an applicant’s “wage or salary history”); N.Y. Lab. Law § 194-a(1) (McKinney 2020) (prohibiting all employers from seeking, requesting or relying upon “wage or salary history” from an applicant); Vt. Stat. Ann. tit. 21, § 495m(a) (West 2018) (prohibiting reliance on and inquiries into an applicant’s “current or past compensation”). Similar legislation is currently pending in Congress and several other states and localities. E.g., Paycheck Fairness Act, H.R. 7, 116th Cong. § 10 (2019) (proposing to prohibit employers from relying upon “wage, salary, and benefit history” in their hiring or pay decisions and from seeking prospective employees’ “wage, salary, and benefit history”).
• 48See Brief for Chamber of Commerce of the United States, et al. as Amicus Curiae supporting Appellee/Cross-Appellant at 8, Greater Phila. Chamber of Commerce v. City of Philadelphia, 949 F.3d 116 (3d Cir. 2020) (Nos. 18-2175 & 18-2176) (“By prohibiting employers from inquiring about [or relying on] pay history, [Philadelphia’s salary history law] denies them useful information for evaluating the quality of candidates during the hiring process.”).
• 49See PayScale, The Salary History Question: Alternatives for Recruiters and Hiring Managers 3 (2017) (reporting the study’s results that showed 43 percent of job applicants were asked about prior pay at some point during the application process); Elizabeth Lester-Abdalla, Salary History Should be Her Story: Upholding Regulations of Salary History through a Commercial Speech Analysis, 60 Wm. & Mary L. Rev. 701, 703 (2018) (“When hiring a new employee, Fresno County takes the new hire’s most recent salary and increases it by about 5 percent to place them on a level within the County’s salary classification bracket.”); Valentina Zarya, Amazon Joins Growing List of Employers That Won’t Ask About Your Salary History, Fortune (Jan. 18, 2018), https://fortune.com/2018/01/18/amazon-salary-history-wage-gap/ [‌ht‌t‌‌ps‌:‌/‌/‌p‌e‌r‌m‌a‌.‌c‌c‌/‌9‌N‌K‌L‌-‌C‌D‌Q‌Y‌]‌ (explaining how Google and other large companies are no longer asking about applicants’ salary history, sometimes in response to jurisdictions’ enactment of salary history laws).
• 50See, e.g., Corinne A. Moss-Racusin et al., Science Faculty’s Subtle Gender Biases Favor Male Students, 109 Proceedings of the Nat. Academy of Sci. 16474 (2012) (describing the results of a randomized double-blind study that found that decisionmakers often pay women less than men even from the very beginning of their careers when there are no differences in male and female workers’ experience, education, or family caregiving responsibilities: the study’s participants offered an average starting salary of approximately $30,000 for the male candidate but only about $26,000 for the identically-qualified female candidate).
• 51See Benjamin Harris, The Hamilton Project, Information is Power: Fostering Labor Market Competition through Transparent Wages 9 (2018) (citing research that employers’ initial wage offers were higher by nine percent when those employers could not ask about applicants’ salary history); Lobel, supra note  NOTEREF _Ref30850389 \h  \* MERGEFORMAT 46 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800350030003300380039000000 , at 573 (“The first negotiation difference, which I call the negotiation deficit, is that women negotiate less frequently and ask for less when they do. This deficit can be mitigated, though not erased, with a salary inquiry ban. The salary inquiry ban has the potential to positively shift the process from letting job applicants lead with a starting point figure to employers implementing a practice of more actively suggesting a fair salary.”); id. (“Salary inquiry bans can also counteract the negative assumptions employers may make when women refuse to reveal their prior salary in a regime that allows salary inquiry. This is a separate effect, which I call the negative inference—when employers assume women who refuse to disclose their pay earn less.”).
• 52And the more that legislatures address arbitrary barriers to employment and other important opportunities, the more inclusive their choices become, and the more those choices may appeal to those on both the political right and the left. See Robin L. West, Civil Rights: Rethinking Their Natural Foundation 83 (2019) (urging that we embrace a broader understanding of the civil right to employment as one that should not be denied for any irrational reason unrelated to performance).
• 53See infra notes  NOTEREF _Ref30865844 \h 62 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800360035003800340034000000 – NOTEREF _Ref37414278  \* MERGEFORMAT 72 and accompanying text (describing certain business associations’ opposition to Philadelphia’s salary history law).
• 54See Clay Risen, The Bill of the Century: The Epic Battle for the Civil Rights Act 220 (2014) (describing the U.S. Chamber of Commerce’s warning that Title VII “could be seriously harmful to the conduct of American business” and requesting “that Title VII be stripped from the [Civil Rights Act]; if that was not possible, then it should be limited to a role of conciliation and persuasion”).
• 55See Lennard J. Davis, Enabling Acts: The Hidden Story of How the Americans with Disabilities Act Gave the Largest U.S. Minority Its Rights 161–216 (2015) (describing some business owners’ and associations’ opposition to the ADA); id. at 171–73, 204–16 (describing some business owners’ and associations’ opposition to extending ADA protections to HIV-positive workers).
• 56New State Ice. Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting). The majority opinion’s abrogation was recognized in Wolverine Fireworks Display v. Towne, No. 12-10426, 2012 U.S. Dist. LEXIS 63259 (E.D. Mich. 2012).
• 57See Masterpiece Cakeshop, Ltd. v. Colo. Civil Rights Comm’n, 138 S. Ct. 1719, 1728 (2018) (“It is unexceptional that Colorado law can protect gay persons, just as it can protect other classes of individuals, in acquiring whatever products and services they choose on the same terms and conditions as are offered to other members of the public.”); Energy Reserves Grp., Inc. v. Kansas Power and Light Co., 459 U.S. 400, 416–17 (1983) (rejecting a constitutional challenge to a Kansas law that regulated gas prices as a proper use of police power based on “significant and legitimate state interests . . . to protect consumers”); Colo. Anti-Discrimination Comm’n v. Cont’l Air Lines, Inc., 372 U.S. 714, 722 (1963) (concluding that state antidiscrimination law does not unconstitutionally burden interstate commerce); Day-Brite Lighting Inc. v. Missouri, 342 U.S. 421, 423–25 (1952) (rejecting a constitutional challenge to a Missouri law that prohibited employers from deducting wages from employees for taking time out for voting); Railway Mail Ass’n. v. Corsi, 326 U.S. 88, 94 (1945) (holding that there is “no constitutional basis for the contention that a state cannot protect workers from exclusion solely on the basis of race, color or creed”); West Coast Hotel Co. v. Parrish, 300 U.S. 379, 393 (1937) (“In dealing with the relation of employer and employed, the Legislature has necessarily a wide field of discretion in order that there may be suitable protection of health and safety, and that peace and good order may be promoted through regulations designed to insure wholesome conditions of work and freedom from oppression.”).
• 58See, e.g., Katzenbach v. McClung, 379 U.S. 294 (1964) (upholding Congress’s Article I power to prohibit public accommodations from discriminating on the basis of race); Heart of Atlanta Motel, Inc., v. United States, 379 U.S. 241 (1964) (same); NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1 (1937) (upholding Congress’s Article I power to regulate the terms and conditions of employment through the National Labor Relations Act).
• 59Note that the Supreme Court’s decision in Masterpiece Cakeshop did not resolve the question whether a baker has a Free Speech Clause right to discriminate on the basis of his customers’ sexual orientation in providing certain (arguably expressive) goods and services. See Masterpiece Cakeshop, 138 S. Ct. at 1719 (concluding instead that state agency had demonstrated hostility towards the baker’s religious beliefs in violation of the Free Exercise Clause). If (and only if) some decisionmakers do have a constitutional right to discriminate in some circumstances, then presumably they would then have the constitutional right to speak in related ways, perhaps by asking applicants and customers questions about their protected class status. But, as the Court has repeatedly made clear, gatekeepers generally do not have a constitutional right to discriminate on the basis of protected characteristics. E.g., Hishon v. King & Spalding, 467 U.S. 69, 78 (1984) (rejecting law firm’s claim that Title VII’s requirement that it refrain from sex discrimination in its partnership decisions violated its First Amendment rights); Roberts v. United States Jaycees, 468 U.S. 609, 628 (1984) (rejecting nonprofit organization’s claim that state law prohibiting discriminatory conduct by public accommodations violated its First Amendment rights); Newman v. Piggie Park Enter., Inc., 390 U.S. 400, 402 n.5 (1968) (per curiam) (rejecting business owner’s constitutional challenge to the Civil Rights Act’s bar on racial discrimination in public accommodations based on his view that racial integration “contraven[ed] the will of God”). The Supreme Court has recognized exceptions to this general rule in certain limited circumstances outside of the commercial setting. See Boy Scouts of America v. Dale, 530 U.S. 640, 640 (2000) (holding that the First Amendment’s implied freedom of association permitted the Boy Scouts of America to exclude gay Scoutmasters despite state public accommodations law that prohibited discrimination on the basis of sexual orientation).
• 60See, e.g., Janus v. Am. Fed’n of State, Cty., & Mun. Emps. Council 31, 138 S. Ct. 2448, 2501 (2018) (Kagan, J., dissenting) (describing the majority as “weaponizing the First Amendment, in a way that unleashes judges, now and in the future, to intervene in economic and regulatory policy”); Jedediah Purdy, Beyond the Bosses’ Constitution: The First Amendment and Class Entrenchment, 118 Colum. L. Rev. 2161, 2161 (2018) (“The Supreme Court’s ‘weaponized’ First Amendment has been its strongest antiregulatory tool in recent decades, slashing campaign-finance regulation, public-sector union financing, and pharmaceutical regulation, and threatening a broader remit.”).
• 61See, e.g., Jeremy K. Kessler & David E. Pozen, The Search for an Egalitarian First Amendment, 118 Colum. L. Rev. 1953 (2018); Genevieve Lakier, Imagining an Antisubordinating First Amendment, 118 Colum. L. Rev. 2117 (2018); Helen Norton, Truth and Lies in the Workplace: Employer Speech and the First Amendment, 101 Minn. L. Rev. 31 (2016).
• 62See Principal and Response Brief for Appellee/Cross-Appellant, Greater Phila. Chamber of Commerce v. City of Philadelphia, 949 F.3d 116 (3d Cir. Nov. 21, 2018) (Nos. 18-2175 & 18-2176).
• 63The Third Circuit recently denied the Philadelphia Chamber’s request for a preliminary injunction of the city’s provision forbidding employers from relying on prior pay in hiring and compensation decisions as well as the city’s provision forbidding employers from asking about applicants’ prior pay. Greater Phila. Chamber of Commerce v. City of Philadelphia, 949 F.3d 116, 132–36 (3d Cir. 2020). I note that I served pro bono as co-counsel on behalf of amici civil rights organizations defending Philadelphia’s law. See Brief for Women’s Law Project, et al. as Amicus Curiae supporting Philadelphia, Greater Phila. Chamber of Commerce v. City of Philadelphia, 949 F.3d 116 (3d Cir. 2020) (Nos. 18-2175 & 18-2176).
• 64Principal and Response Brief for Appellee/Cross-Appellant, supra note  NOTEREF _Ref30865844 \h  \* MERGEFORMAT 62 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800360035003800340034000000 , at 29.
• 65Id. at 25–27.
• 66Id. at 16. But as described infra notes  NOTEREF _Ref31814479 \h  \* MERGEFORMAT 94 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330031003800310034003400370039000000 ,  NOTEREF _Ref36850029 \h  \* MERGEFORMAT 99 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003800350030003000320039000000 ,  NOTEREF _Ref36850036 \h  \* MERGEFORMAT 100 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003800350030003000330036000000 ,  NOTEREF _Ref36850100 \h  \* MERGEFORMAT 102 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003800350030003100300030000000 – NOTEREF _Ref36850040 \h  \* MERGEFORMAT 104 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003800350030003000340030000000 and accompanying text, the Supreme Court has repeatedly emphasized that the First Amendment does not protect a gatekeeper’s statement “White Applicants Only,” even though this speech also communicates a message about the value a prospective employer places on certain applicants because of their protected class status.
• 67See Newman, supra note  NOTEREF _Ref36839160 \h  \* MERGEFORMAT 59 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003800330039003100360030000000 (rejecting business owner’s constitutional challenge to the Civil Rights Act’s bar on racial discrimination in public accommodations based on his view that racial integration “contraven[ed] the will of God”).
• 68As an illustration, a 1950 U.S. Senate Subcommittee report argued just this. See Subcomm. on Investigations of the S. Comm. on Expenditures in the Exec. Dep’ts, Employment of Homosexuals and Other Sex Perverts in Government, S. Doc. No. 81-241, at 3–4 (1950) (“In the opinion of this subcommittee homosexuals and other sex perverts are not proper persons to be employed in Government for two reasons; first, they are generally unsuitable, and second, they constitute security risks. . . . [I]t is generally believed that those who engage in overt acts of perversion lack the emotional stability of normal persons. In addition, there is an abundance of evidence to sustain the conclusion that indulgence in acts of sex perversion weakens the moral fiber of an individual to a degree that he is not suitable for a position of responsibility.”).
• 69See supra notes  NOTEREF _Ref53245153 \h 18 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003200340035003100350033000000 –19 and accompanying text (discussing cases in which employers declined to hire women they feared might become pregnant).
• 70See supra notes  NOTEREF _Ref31814041 \h  \* MERGEFORMAT 57 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330031003800310034003000340031000000 – NOTEREF _Ref31814088 \h  \* MERGEFORMAT 58 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330031003800310034003000380038000000 and accompanying text. An employer illegally relies on salary history when it pays a salary that relies on the candidate’s prior salary, not when it communicates this decision to the applicant. If communicating a salary offer that relies on a protected characteristic is protected speech, then the same would be true of communicating a salary offer that relies on other protected characteristics like religion, race, or gender.
• 71Greater Phila. Chamber of Commerce v. City of Philadelphia, 949 F.3d 116, 132–36 (3d Cir. 2020).
• 72As law professor Charlotte Garden has observed, “[A]lthough many of these theories are a stretch for now, individual deregulatory First Amendment cases should not be viewed as outliers: the outward push is occurring simultaneously on multiple fronts, and its standard-bearers include some exceedingly well-respected and influential lawyers.” Charlotte Garden, The Deregulatory First Amendment at Work, 51 Harv. C.R.-C.L. L. Rev. 323, 325 (2016); see also id. at 362 (“[E]ven First Amendment arguments that are unlikely to be accepted can matter; for example, Chicago reportedly considered a minimum wage ordinance modeled on Seattle’s, but abandoned it in light of [an industry group’s unsuccessful First Amendment challenge to Seattle’s increase in its minimum wage, alleging that the increase would leave its members with less resources available to spend on speech activities]. . . . Thus, one problem with the emerging deregulatory First Amendment is that it can accomplish some of its aims without the courts ever adopting it; the increasingly real threat of expensive litigation by high-profile litigators can stay regulators’ hands.”)
• 73Frederick Schauer, The Boundaries of the First Amendment: A Preliminary Exploration of Constitutional Salience, 117 Harv. L. Rev. 1765, 1781 (2004); see also id. at 1770 (“[N]o First Amendment-generated level of scrutiny is used to determine whether the content-based advertising restrictions of the Securities Act of 1933 are constitutional, whether corporate executives may be imprisoned under the Sherman Act for exchanging accurate information about proposed prices with their competitors, whether an organized crime leader may be prosecuted for urging that his subordinates murder a mob rival, or whether a chainsaw manufacturer may be held liable in a products liability action for injuries caused by mistakes in the written instructions accompanying the tool.”).
• 74See United States v. Williams, 553 U.S. 285, 297–98 (2008) (“Offers to engage in illegal transactions are categorically excluded from First Amendment protection. . . . Many long established criminal proscriptions—such as laws against conspiracy, incitement, and solicitation—criminalize speech (commercial or not) that is intended to induce or commence illegal activities.”); New York v. Ferber, 458 U.S. 747, 761 (1982) (holding that the First Amendment does not protect the “advertising and selling of child pornography” because they “provide an economic motive for and are thus an integral part of the production of such materials, an activity illegal throughout the Nation”); California Motor Transp. Co. v. Trucking Unlimited, 404 U.S. 508, 514 (1972) (stating that the First Amendment does not protect speech that is an “integral part” of illegal conduct); Joseph Blocher, Categoricalism and Balancing in First and Second Amendment Analysis, 84 N.Y.U. L. Rev. 375, 388 (2009) (“Fraud and crime-facilitating speech, for example, are thought to be entirely outside the bounds of the Amendment, and no balancing is required to suppress them in a given case.”).
• 75Greenawalt, Fighting Words supra note  NOTEREF _Ref36587566 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380037003500360036000000 ; see also Greenawalt, the Uses of Language, supra note  NOTEREF _Ref36587566 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380037003500360036000000 (“Thus, with some roughness, we can speak of assertions of fact and value as making claims about what already exists in the listener’s world. Situation-altering utterances purport to change that world.”); id. at 239 (describing “communications that I claim fall outside the coverage of the First Amendment” as “too far removed from ordinary statements of fact and value to deserve even moderately stringent constitutional protection”).
• 76Greenawalt, The Uses of Language, supra note  NOTEREF _Ref36587566 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380037003500360036000000 , at 40.
• 77Greenawalt, Fighting Words, supra note  NOTEREF _Ref36587566 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380037003500360036000000 .
• 78See Kent Greenawalt, Speech and Exercise by Private Individuals and Organizations, 72 SMU L. Rev. 397, 400 (2019).
• 79Greenawalt, Fighting Words, supra note  NOTEREF _Ref36587566 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380037003500360036000000 , at 79.
• 80See Erica Goldberg, Free Speech Consequentialism, 116 Colum. L. Rev. 687, 732 (2016) (“[W]hen speech begins to resemble conduct, such as when it impairs discrete, material interests through direct processes and through the fault mostly of the speaker, then courts should consider those conduct-like harms in their consequentialist calculus.”).
• 81See Schauer, supra note  NOTEREF _Ref30868575 \h  \* MERGEFORMAT 73 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800360038003500370035000000 , at 1773 (observing as a descriptive matter that “the speech with which we make contracts is, in general, not within the scope of ‘the freedom of speech’ and thus not covered by the First Amendment”); G. Edward White, Falsity and the First Amendment, 72 SMU L. Rev. 513, 525 (2019) (“No current court would find that the First Amendment shields false or misleading speech affecting the creation of a contract from exposing the speaker to contract damages, or speech asking another to commit a murder from criminal sanctions.”).
• 82Rodney A. Smolla, Rethinking First Amendment Assumptions About Racist and Sexist Speech, 47 Wash. & Lee. L. Rev. 171, 186–87 (1990); see also Ohralik v. Ohio State Bar Ass’n, 436 U.S. 447, 456–57 (1978) (describing the government’s constitutionally permissible regulation of “a business transaction in which speech is an essential but subordinate component”); Greenawalt, Fighting Words, supra note  NOTEREF _Ref36587566 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380037003500360036000000 , at 83 (“Smolla’s core idea of ‘statements of transaction’ is very close to what I have called situation-altering utterances, remarks that do something rather than tell something.”).
• 83Smolla, supra note  NOTEREF _Ref30868862 \h  \* MERGEFORMAT 82 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800360038003800360032000000 , at 187; see also id. (explaining that “the laws governing the language that must appear on a negotiable instrument[] never have been thought to implicate freedom of speech”). Note that transactions may or may not be commercial, depending on whether they involve the exchange of goods and services for compensation. See United States v. Williams, 553 U.S. 285, 298 (2008) (“Offers to provide or requests to obtain unlawful material, whether as part of a commercial exchange or not, are similarly undeserving of First Amendment protection. It would be an odd constitutional principle that permitted the government to prohibit offers to sell illegal drugs, but not offers to give them away for free.”).
• 84Daniel A. Farber, Commercial Speech and First Amendment Theory, 74 Nw. U. L. Rev. 372, 386 (1979).
• 85Id. at 388–89.
• 86Jane R. Bambauer, The Relationships between Speech and Conduct, 49 U.C. Davis L. Rev. 1941, 1948 (2016).
• 87As Kent Greenawalt explains: “The argument against the relevance of a free speech principle is strongest when the information disclosed is so narrowly specific that no significant subject of discussion or learning is involved. [The reasons for free speech protections] apply less strongly if speaker and listener care only about an immediate practical objective and not about any increase in general understanding or expression of personal feelings and attitudes.” Greenawalt, The Uses of Language, supra note  NOTEREF _Ref36587566 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380037003500360036000000 , at 47; see also Schauer, supra note  NOTEREF _Ref30868575 \h  \* MERGEFORMAT 73 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800360038003500370035000000 , at 1801 (interpreting Greenawalt’s argument to distinguish between speech that is “face-to-face, informational, particular, and for private gain” from speech that “is public rather than face-to-face, when it is inspired by the speaker’s desire for social change rather than for private gain, when it relates to something general rather than to a specific transaction, and when it is normative rather than informational in content” and concluding that the First Amendment is “irrelevant” to the former, and “plainly” implicated in the latter); Eugene Volokh, Crime-Facilitating Speech, 57 Stan. L. Rev. 1095, 1102–05 (2005) (suggesting that the First Amendment affords greater protection to “dual-use” information that provides information to a wide public audience even if it enables some listeners to commit illegal acts than it does to single or limited use information that enables the parties in one-on-one conversations to commit illegal acts).
• 88White, supra note  NOTEREF _Ref36589827 \h 81 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380039003800320037000000 , at 525.
• 89See Nadine Strossen, The Tensions between Regulating Workplace Harassment and the First Amendment: No Trump, 71 Chi.-Kent L. Rev. 701, 704 (1995) (“Even the most diehard free speech absolutist recognizes that the speech involved in quid pro quo harassment is tantamount to threats or extortion, expression that has long been punishable without raising substantial free speech concerns in any context.”); Eugene Volokh, Freedom of Speech and Workplace Harassment, 39 UCLA L. Rev. 1791, 1800 (1992) (stating that quid pro quo harassment “would seemingly be as unprotected by the First Amendment as any other form of threat or extortion”).
• 90Greenawalt, The Uses of Language, supra note  NOTEREF _Ref36587566 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380037003500360036000000 , at 85.
• 91See, e.g., Meritor Sav. Bank v. Vinson, 477 U.S. 57 (1986); Harris v. Forklift Sys., Inc. 510 U.S. 17 (1993).
• 92See Meritor, 477 U.S. at 67 (“Sexual harassment which creates a hostile or offensive environment for members of one sex is every bit the arbitrary barrier to sexual equality at the workplace that racial harassment is to racial equality. Surely, a requirement that a man or woman run a gauntlet of sexual abuse in return for the privilege of being allowed to work and make a living can be as demeaning and disconcerting as the harshest of racial epithets.”) (quoting Henson v. City of Dundee, 682 F.2d 897, 902 (11th Cir. 1982)); see also Cynthia Estlund, Freedom of Expression in the Workplace and the Problem of Discriminatory Harassment, 75 Tex. L. Rev. 687, 689–90 (1997) (“A hostile work environment imposes serious discriminatory burdens on female employees and helps to maintain sexual segregation of many segments of the workforce by marking certain workplaces or certain levels of the workplace hierarchy off-limits to women. Similarly, harassment targeting racial minorities, such as persistent racial taunts, ridicule, or threats, retards progress toward racial integration and equality in the workforce and burdens the work lives of minority employees . . . .”); Volokh, Workplace Harassment, supra note  NOTEREF _Ref30870547 \h  \* MERGEFORMAT 89 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800370030003500340037000000 , at 1809 (“When women and minority employees suffer such intolerable abuse, the abuse both interferes with their ability to make a living, and creates barriers for them that others in the workplace do not have to overcome.”).
• 93R.A.V. v. City of St. Paul, 505 U.S. 377, 389 (1992) (citations omitted).
• 94Sorrell v. IMS Health Inc., 564 U.S. 552 (2011); Rumsfeld v. Forum for Acad. & Inst. Rights, 547 U.S. 47 (2006).
• 95Pittsburgh Press Co. v. Pittsburgh Comm’n on Human Relations, 413 U.S. 376, 379 (1973) (holding that the First Amendment did not protect employers’ statements of discriminatory preference in the form of advertisements of “Jobs—Male Interest”); see also Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n, 447 U.S. 557, 564 (1980) (citing the facts in Pittsburgh Press as an example of commercial speech unprotected by the First Amendment because of its relationship to illegal commercial activity).
• 96See Rumsfeld, 547 U.S. at 62–66. Note that laws that prohibit gatekeepers’ discriminatory advertisements or other statements of discriminatory preference are almost as prevalent as laws that prohibit gatekeepers’ inquiries about applicants’ protected class status; e.g., 42 U.S.C. § 3604(c) (2012) (prohibiting housing providers from “indicat[ing] any preference, limitation, or discrimination based on race, color, religion, sex, handicap, familial status, or national origin”); 42 U.S.C. § 2000e-3(b) (2012) (making it unlawful “to print or publish or cause to be printed or published any notice or advertisement relating to employment . . . indicating any preference, limitation, specification, or discrimination, based on race, color, religion, sex, or national origin”); 29 U.S.C. § 623(e) (2012) (“It shall be unlawful for an employer, labor organization, or employment agency to print or publish, or cause to be printed or published, any notice or advertisement relating to employment by such an employer or membership in [such an organization or] classification or referral . . . by such an employment agency, indicating any preference, limitation, specification, or discrimination, based on age.”); see also Norton, You Can’t Ask (Or Say) That, supra note  NOTEREF _Ref30871633 \h  \* MERGEFORMAT 1 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800370031003600330033000000 , at 732–33 (canvassing state and local laws that prohibit gatekeepers’ discriminatory advertisements or other statements of discriminatory preference).
• 97547 U.S. 47 (2006).
• 98Id. at 60–62 (quoting Giboney v. Empire Storage & Ice Co., 336 U.S. 490, 502 (1949)). As the Court notes, speech can “initiate” or “carry out” illegal conduct; such is the case of threats, offers, agreements, statements of discriminatory preference, and other “situation-altering” statements. In this Article, I use the terms “enable” or “facilitate” to describe these connections between certain speech and regulated conduct. The Court also notes the use of speech as “evidence” of a speaker’s illegal motive for its conduct, which describes a slightly different relationship between speech and illegal conduct, and one that is also endemic throughout the law. See Wisconsin v. Mitchell, 508 U.S. 476, 489 (1993) (“The First Amendment, moreover, does not prohibit the evidentiary use of speech to establish the elements of a crime or to prove motive or intent.”); Cass R. Sunstein, Democracy and The Problem of Free Speech 125­–26 (1993) (explaining that a letter saying, “You’re fired, because I won’t let blacks work here” is “simply evidence of what is unlawful, a discharge based on discrimination. Use of the letter to prove discriminatory motive is hardly unconstitutional even if the letter is speech.”) (internal quotation marks omitted). Indeed, a challenger can offer a decisionmaker’s question about a candidate’s protected class status as evidence of the ultimate decision’s discriminatory motive and thus its illegality. See EEOC v. Wal-Mart Stores, Inc., 11 F. Supp. 2d 1313, 1328 (D.N.M. 1998), aff’d, 187 F.3d 1241 (10th Cir. 1999) (“[I]t is reasonable to infer from the testimony presented at trial that the asking of the question [about disability status] set off a chain of events that ultimately led to Wal-Mart’s discriminatory conduct of refusing to hire [the plaintiff].”).
• 99Rumsfeld, 547 U.S. at 62.
• 100564 U.S. 552, 567 (2011).
• 101Id. at 557.
• 102Id. at 567; see also Expressions Hair Design v. Schneiderman, 137 S. Ct. 1144, 1151 (2017) (“[A] law requiring all New York delis to charge $10 for their sandwiches . . . would simply regulate the amount that a store could collect. In other words, it would regulate the sandwich seller’s conduct. To be sure, in order to actually collect that money, a store would likely have to put ‘$10’ on its menus or have its employees tell customers that price. Those written or oral communications would be speech, and the law—by determining the amount charged—would indirectly dictate the content of that speech. But the law’s effect on speech would be only incidental to its primary effect on conduct . . . .”); Nat’l Inst. of Family & Life Advocates v. Becerra, 138 S. Ct. 2361, 2373 (2018) (identifying malpractice and informed consent requirements as examples of the government’s constitutionally permissible “regulations of professional conduct that incidentally burden speech”).
• 103Greenawalt, The Uses of Language, supra note  NOTEREF _Ref36587566 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380037003500360036000000 , at 244 (defining a situation-altering order as a statement “by someone in authority, concerning acts as to which his authority generally extends”).
• 104See Hous. Opportunities Made Equal, Inc. v. Cincinnati Enquirer, Inc., 943 F.2d 644, 652 (6th Cir. 1991) (“Without the regulation of advertisements, realtors could deter certain classes of potential tenants from seeking housing at a particular location, effectively discriminating against these classes without running afoul of the FHA’s prohibition against discriminatory housing practices. Congress obviously recognized the key role housing advertisements play in potential real estate transactions and concluded that the regulation of real estate advertisements is warranted.”); Ragin v. N.Y. Times Co., 923 F.2d 995, 999–1000 (2d Cir. 1991) (“[W]e read [the Fair Housing Act’s bar on discriminatory advertisements] to describe any ad that would discourage an ordinary reader of a particular race from answering it.”); Charles R. Lawrence III, Crossburning and the Sound of Silence: Antisubordination Theory and the First Amendment, 37 Vill. L. Rev. 787, 795 (1992) (explaining that law treats speech like “Whites Only Need Apply” as “‘discriminatory practices’ and outlaw[s] them under federal and state civil rights legislation because they are more than speech”).
• 105Despite the contemporary antiregulatory turn in its Free Speech Clause doctrine, the Court continues to apply this commercial speech framework. See, e.g., Matal v. Tam, 137 S. Ct. 1744, 1763–65 (2017) (discussing commercial speech doctrine with respect to the government’s regulation of trademarks); Expressions Hair Design, 137 S. Ct. 1144, 1151 (2017) (discussing commercial speech doctrine with respect to the government’s regulation of retailers’ communication about prices).
• 106425 U.S. 748 (1976).
• 107Id. (overruling Valentine v. Chrestensen, 316 U.S. 52, 54 (1942)).
• 108Id. at 763.
• 109Id. at 756.
• 110See Schauer, supra note  NOTEREF _Ref30868575 \h  \* MERGEFORMAT 73 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800360038003500370035000000 , at 1776 n.49 (“The Central Hudson approach demands a threshold inquiry into whether the speech is misleading. Thus, misleading commercial advertisements are akin to legally obscene materials in that they are regulable under minimal rational basis scrutiny without regard to First Amendment standards or values. Indeed, misleading commercial speech is arguably simply not covered by the First Amendment.”); White, supra note  NOTEREF _Ref36589827 \h  \* MERGEFORMAT 81 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330036003500380039003800320037000000 , at 527 (“False commercial speech falls outside the coverage of the First Amendment and can be regulated with impunity.”).
• 111Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n, 447 U.S. 557, 563–64 (1980) (striking down governmental ban on electric utilities’ promotion of electricity consumption) (citations omitted).
• 112Id.; see also id. at 562 (noting that the Court’s “decisions have recognized ‘the “commonsense” distinction between speech proposing a commercial transaction, which occurs in an area traditionally subject to government regulation, and other varieties of speech’”) (quoting Ohralik v. Ohio State Bar Ass’n, 436 U.S. 447, 455–56 (1978)).
• 113Pittsburgh Press Co. v. Pittsburgh Comm’n on Human Relations, 413 U.S. 376, 385 (1973). The Court has also characterized commercial speech as “expression related solely to the economic interests of the speaker and its audience.” Central Hudson, 447 U.S. at 561.
• 114See Smolla, supra note  NOTEREF _Ref30868862 \h  \* MERGEFORMAT 82 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800360038003800360032000000 and accompanying text.
• 115E.g., Expressions Hair Design v. Schneiderman, 137 S. Ct. 1144 (2017) (characterizing New York law as a regulation of commercial speech because it regulated retailers’ communication of the price of goods and services); Bd. of Tr. of State Univ. of N.Y. v. Fox, 492 U.S. 469 (1989) (characterizing product demonstrations in campus dormitory rooms as commercial speech); Beeman v. Anthem Prescription Mgmt., LLC, 315 P.3d 71, 74–75 (Cal. 2013) (characterizing a regulation requiring “prescription drug claims processors to compile and summarize information on pharmacy fees and to transmit the information to their clients” as the regulation of commercial speech); Carrico v. City of San Francisco, 656 F.3d 1002 (9th Cir. 2011) (describing a law prohibiting landlords from coercing tenants to vacate their homes through offers of payment, accompanied by threats and intimidation, as the regulation of commercial speech); Nat’l Cable & Telecomms. Ass’n v. FCC, 555 F.3d 996, 996 (D.C. Cir. 2009) (characterizing the government’s regulation requiring carriers “to obtain opt-in consent from a customer before disclosing that customer’s information to a carrier’s joint venture partner” as regulating commercial speech); Trans Union Corp. v. FTC, 267 F.3d 1138 (D.C. Cir. 2001) (characterizing consumer credit reports as commercial speech).
• 116Felix T. Wu, Commercial Speech Protection as Consumer Protection, 90 U. Colo. L. Rev. 631, 644 (2019).
• 117Pittsburgh Press Co., 413 U.S. at 385 (“Each [job advertisement] is no more than a proposal of possible employment. The advertisements are thus classic examples of commercial speech.”).
• 118E.g., Centro De La Comunidad Hispana De Locust Valley v. Town of Oyster Bay, 868 F.3d 104 (2d Cir. 2017) (characterizing potential employers’ solicitation of day laborers as commercial speech because it involves advertisements and negotiations for work); Valle Del Sol, Inc. v. Whiting, 709 F.3d 808 (9th Cir. 2013) (same); Nomi v. Regents for Univ. of Minn., 796 F. Supp. 412, 417 (D. Minn. 1992) (“[Military job] recruiting proposes a commercial transaction; the purpose of recruiting is to reach an agreement under which services will be exchanged for compensation.”), vacated as moot, 5 F.3d 332 (8th Cir. 1993); Amalgamated Acme Affiliates, Inc. v. Minton, 33 S.W.3d 387, 394 (Tex. App. 2000) (concluding that speech asserting that a former employee was subject to, and in violation of, a non-compete agreement was commercial speech).
• 119Greater Phila. Chamber of Commerce v. City of Philadelphia, 949 F.3d 116 (3d Cir. 2020) (concluding that employers’ inquiries about candidates’ salary history in the context of job applications and negotiations constituted commercial speech).
• 120Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n, 447 U.S. 557, 562–64 (1980).
• 121See supra notes  NOTEREF _Ref31814041 \h 57 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330031003800310034003000340031000000 –58 and accompanying text.
• 122See, e.g., United States v. Benson, 561 F.3d 718, 725–26 (7th Cir. 2009) (characterizing the advertising of materials that advocated not filing federal income tax returns as unprotected commercial speech related to illegal activity); Wash. Mercantile Ass’n v. Williams, 733 F.2d 687 (9th Cir. 1984) (characterizing drug paraphernalia advertisements as unprotected commercial speech related to illegal activities); New England Accessories Trade Assocs., Inc. v. City of Nashua, 679 F.2d 1 (1st Cir. 1982) (same); Fla. Businessmen for Free Enter. v. City of Hollywood, 673 F.2d 1213 (11th Cir. 1982) (same); Kan. Retail Trade Coop. v. Stephan, 695 F.2d 1343 (10th Cir. 1982) (same); State v. Johnson, 324 N.W.2d 447 (Wis. Ct. App. 1982) (characterizing advertisements for prostitution as unprotected commercial speech related to an illegal commercial transaction); Washington v. Clark Cty. Liquor and Gaming Licensing Bd., 683 P.2d 31 (Nev. 1984) (same); see also United States v. Williams, 553 U.S. 285, 297–98 (2008) (holding that offers to provide, and requests for, child pornography are unprotected by the First Amendment because the distribution and possession of child pornography is itself illegal).
• 123See supra notes  NOTEREF _Ref31814041 \h 57 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330031003800310034003000340031000000 –58 and accompanying text.
• 124This Article focuses on statutes that protect certain characteristics from discrimination by nongovernmental or governmental employers and other gatekeepers. Of course, apart from any statutory requirements, the Supreme Court has interpreted the Equal Protection Clause to prohibit the government from discriminating based on certain characteristics in its decisions. E.g., Brown v. Bd. of Educ., 347 U.S. 483 (1954) (striking down the government’s race-based segregation of public schools on equal protection grounds); United States v. Virginia, 518 U.S. 515 (1996) (striking down the government’s exclusion of women from the state’s Virginia Military Institute on equal protection grounds).
• 125Central Hudson, 447 U.S. at 564.
• 126413 U.S. 376, 378 (1973).
• 127Id. at 388 (“We have no doubt that a newspaper constitutionally could be forbidden to publish a want ad proposing a sale of narcotics or soliciting prostitutes. Nor would the result be different if the nature of the transaction were indicated by placement under columns captioned ‘Narcotics for Sale’ and ‘Prostitutes Wanted’ rather than stated within the four corners of the advertisement.”).
• 128Id. at 388–89.
• 129See, e.g., Reed v. Town of Gilbert, 576 U.S. 155, 163–64 (2015) (stating that the government’s content-based or speaker-based regulation of speech generally triggers strict scrutiny). But as many thoughtful commentators have observed, the Court’s First Amendment doctrine justifiably includes numerous exceptions (including but not limited to its commercial speech doctrine) in which it upholds the government’s speaker- and content-based distinctions without applying strict scrutiny. E.g., Goldberg, supra note  NOTEREF _Ref30883659 \h  \* MERGEFORMAT 80 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380033003600350039000000 , at 695 (canvassing precedent to conclude that “free speech consequentialism, more than being ubiquitous, is in fact inevitable”); James Weinstein, Speech Categorization and the Limits of First Amendment Formalism: Lessons from Nike v. Kasky, 54 Case W. Res. L. Rev. 1091, 1100 (2004) In summary, the popular view that all content-based restrictions on speech are presumptively unconstitutional unless the speech falls within some unprotected category is not an accurate snapshot of First Amendment doctrine. Speech is too ubiquitous with too many real-world consequences for there to be any such rule. Rather, the strong presumption against content discrimination operates only within a limited (albeit extremely important) domain.
• 130See, e.g., Sorrell v. IMS Health Inc., 564 U.S. 552 (2011) (noting that the First Amendment permits the government to regulate gatekeepers’ statements of discriminatory preference like “White Applicants Only” that initiate or carry out illegal discrimination); Rumsfeld v. Forum for Acad. & Inst. Rights, 547 U.S. 47 (2006) (noting that the First Amendment permits the government to regulate gatekeepers’ statements of discriminatory preference that are incidental to the government’s regulation of “commerce or conduct”); R.A.V. v. City of St. Paul, 505 U.S. 377, 389 (1992) (“[S]ince words can in some circumstances violate laws directed not against speech but against conduct (a law against treason, for example, is violated by telling the enemy the Nation’s defense secrets), a particular content-based subcategory of a proscribable class of speech can be swept up incidentally within the reach of a statute directed at conduct rather than speech. Thus, for example, sexually derogatory ‘fighting words,’ among other words, may produce a violation of Title VII’s general prohibition against sexual discrimination in employment practices.”) (citations omitted).
• 131See Sorrell, 564 U.S. at 578–79 (“The State may not burden the speech of others in order to tilt public debate in a preferred direction.”).
• 132For a more accurate parallel to Sorrell in the antidiscrimination context, consider instead Linmark Associates, Inc. v. Township of Willingboro, where the Court upheld a First Amendment challenge to a law that barred “For Sale” and “Sold” signs on real estate to prevent “panic” selling by whites who feared that the town’s racial integration would drive down property prices; such sales remained legal even though disfavored by the town. 431 U.S. 85, 85 (1977). Note that Linmark predates Central Hudson; under the Central Hudson framework, the law at issue in Linmark would now be understood as a regulation of accurate commercial speech related to legal activity and thus subject to intermediate scrutiny. The Court has generally rejected the government’s paternalistic regulation of speech for fear that listeners will make unwise, yet legal, decisions. But the antidiscrimination laws described in Part I apply to gatekeepers’ inquiries that enable illegally discriminatory transactions, and thus restrict speech that the First Amendment does not protect.
• 133See Farber, supra note  NOTEREF _Ref30882441 \h  \* MERGEFORMAT 84 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380032003400340031000000 , at 400 (describing the government’s regulation of discriminatory job advertisements as “relat[ing] to the contractual function of the ads [as offers of employment], rather than to the suppression of the free flow of information”).
• 134Furthermore, the limitations of after-the-fact enforcement of antidiscrimination laws mean that alternatives—like simply prohibiting reliance on (i.e., use of information about) protected class status in decision-making—will not effectively achieve the government’s objectives. Nor would prohibiting only inquiries made with the intent to inform illegal conduct: not only does advance screening of “innocent” inquiries from those related to illegal decisions pose an unmanageable challenge, but even “innocent” queries can deter applicants from continuing to seek the opportunity at stake and can elicit information about protected class status that remains available, consciously or unconsciously, for later use in decision-making. See supra notes  NOTEREF _Ref53245367 \h 12 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003200340035003300360037000000 –21 and accompanying text.
• 135Greenawalt, The Uses of Language, supra note  NOTEREF _Ref30869069 \h  \* MERGEFORMAT 5 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800360039003000360039000000 , at 68.
• 136See Stewart v. Furton, 774 F.2d 706, 710 n.2 (6th Cir. 1985) (observing the First Amendment problem if the Fair Housing Act prohibited housing providers’ statements of discriminatory preference that did not relate “to a specific discriminatory and illegal transaction”); IMDB.com, Inc. v. Becerra, 962 F.3d 1111 (9th Cir. 2020) (striking down, on First Amendment grounds, California law that prohibited the general publication of truthful age-related information about those in the entertainment industry when the law did not regulate the conduct and speech of parties engaged in a commercial transaction).
• 13712 C.F.R. § 1002.5(b) (2019).
• 13829 C.F.R. § 1604.7 (2002).
• 13942 U.S.C. § 3604(c) (2000).
• 140Pittsburgh Press Co. v. Pittsburgh Comm’n on Human Relations, 413 U.S. 376, 391 (1973).
• 141See Tung Yin, How the Americans with Disabilities Act’s Prohibition on Pre-Employment-Offer Disability-Related Questions Violates the First Amendment, 17 Lab. Law. 107, 118–19 (2001) (accepting Pittsburgh Press’s analysis with respect to discriminatory advertisements while arguing that the ADA’s prohibitions on disability-related inquiries do not “automatically deter” certain applicants in the way that sex-segregated job advertisements do).
• 142The Third Circuit denied the Philadelphia Chamber of Commerce’s request to preliminarily enjoin both the reliance and the inquiry provisions of Philadelphia’s salary history law. Greater Phila. Chamber of Commerce v. City of Philadelphia, 949 F.3d 116 (3d Cir. 2020). Although I agree with the appellate court’s decision to deny the injunctions, I disagree with the portion of its analysis where it declined to describe employer inquiries about prior pay as “related to” illegal activity even though reliance on the answer constituted illegal activity under Philadelphia’s law. There the appellate court mistakenly (in my view) asserted that contested speech must always, and only, be related to illegal conduct to lose First Amendment protection under Central Hudson’s framework. Id. at 141–42. The court instead characterized the provision as regulating commercial speech about legal activity, applied intermediate scrutiny, and then found that the provision survived such scrutiny. Id. at 142–57.
• 143See Brief for Chamber of Commerce of the United States, et al., supra note  NOTEREF _Ref53179316 \h  \* MERGEFORMAT 48 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003100370039003300310036000000 , at 8 (“By prohibiting employers from inquiring about [or relying on] pay history, [Philadelphia’s salary history law] denies them useful information for evaluating the quality of candidates during the hiring process.”).
• 144Pittsburgh Press, 413 U.S. at 388 (“The illegality in this case may be less overt [than advertisements for the sale of illegal drugs] but we see no difference in principle here.”); see also id. at 381 n.7 (recounting the defendant’s argument that sex-segregated advertisements simply reflected men’s and women’s relative interest in certain job categories and that women might find them helpful in their search for employment); Village of Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489, 496–97 & n.9 (1982) (describing ads marketing pipes and other paraphernalia as unprotected commercial speech related to the illegal sale of drugs even though those products could also have been used for lawful activity other than drug use); id. at 497 (“[T]he overbreadth doctrine does not apply to commercial speech.”).
• 145Pittsburgh Press, 413 U.S. at 389 (“The advertisements, as embroidered by their placement, signaled that the advertisers were likely to show an illegal sex preference in their hiring decisions.”); see also Kathleen M. Sullivan, Cheap Spirits, Cigarettes, and Free Speech: The Implications of 44 Liquormart, 1996 Sup. Ct. Rev. 123, 150–52 (1996) (observing that even if the Pittsburgh Press ads did not explicitly exclude women from applying for male-designated jobs, they made such applications substantially less likely).
• 146See Kristina E. Music Biro et al., Solicitation generally, 21 Am. Jur. 2d Criminal Law § 153 (Nov. 2019) (“Solicitation is complete once the request to join in a crime is made and is punishable irrespective of the reaction of the person solicited; therefore, the fortuity that the person solicited does not agree to commit or attempt to commit the incited crime plainly should not relieve the solicitor of liability when otherwise he would be a conspirator or an accomplice.”); John Bourdeau, Nature and extent of liability—Liability of person joining existing conspiracy, 16 Am. Jur. 2d Conspiracy § 21 (Nov. 2019) (“One becomes a member of an existing conspiracy by knowingly cooperating to further the object of the conspiracy. One may join a conspiracy by word or by deed.”).
• 147See Brief for Chamber of Commerce of the United States, et al., supra note  NOTEREF _Ref53179316 \h  \* MERGEFORMAT 48 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003100370039003300310036000000 , at 9.
• 148See Harris, supra note  NOTEREF _Ref30882064 \h  \* MERGEFORMAT 51 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380032003000360034000000 , at 4 (“[M]any employers use compensation surveys to know precisely where their workers fall in the distribution of wages.”); Joanne Sammer, Banning Salary History Questions: A Game Changer?, Soc’y for Human Res. Mgmt. (Oct. 6, 2016), https://www.shrm.org/resourcesandtools/hr-topics/compensation/pages/banning-salary-history.aspx [https://perma.cc/MPL9-YWP9]. When managers base salary offers on a combination of an applicant’s current salary and what the pay budget allows—rather than on what the market is paying for a given position, skills and experience—the hiring process is less likely to yield the best candidate. With no access to applicant salary information, employers have an opportunity to move toward a broader approach to hiring.”).
• 149See PayScale, supra note  NOTEREF _Ref30882388 \h  \* MERGEFORMAT 49 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380032003300380038000000 , at 7 (suggesting alternatives for employers like asking “[w]hat are your salary expectations?” or describing their pay range to applicants). Note that although laws like Philadelphia’s bar employers from relying on an applicant’s prior pay for decision- making purposes, they permit employers to rely on, and ask about, an applicant’s salary expectations in their hiring and compensation decisions. E.g., Del. Code Ann. tit. 19, § 709B(d) (West 2017) (allowing inquiries into “compensation expectations” so long as the employer does not inquire into “compensation history”).
• 150I explored related issues in Truth and Lies in the Workplace, supra note  NOTEREF _Ref53245558 \h 61 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003200340035003500350038000000 (urging that we understand employers’ speech about the terms and conditions of employment as both protected and regulated to the extent that it furthers or frustrates workers’ First Amendment interests as listeners).
• 151See supra notes  NOTEREF _Ref53183968 \h 106 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003100380033003900360038000000 –23 and accompanying text; Wu, supra note  NOTEREF _Ref30884204 \h  \* MERGEFORMAT 116 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380034003200300034000000 , at 631–32 (“Commercial speech doctrine cares primarily about informing consumers, and that is the lens through which courts should determine how much scrutiny to give to a commercial speech restriction. In commercial speech cases, courts should not be applying the kind of speaker-focused approaches they would be using in cases involving noncommercial speech.”).
• 152See Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n, 447 U.S. 557, 562 (1980) (noting that its “decisions have recognized ‘the “commonsense” distinction between speech proposing a commercial transaction, which occurs in an area traditionally subject to government regulation, and other varieties of speech’”) (quoting Ohralik v. Ohio State Bar Ass’n, 436 U.S. 447, 455–56 (1978)); R.A.V. v. City of St. Paul, 505 U.S. 377, 389–90 (1992) (“When the basis for the content discrimination consists entirely of the very reason the entire class of speech at issue is proscribable, no significant danger of idea or viewpoint discrimination exists. Such a reason, having been adjudged neutral enough to support exclusion of the entire class of speech from First Amendment protection, is also neutral enough to form the basis of distinction within the class. . . . [T]o take a final example, a State may choose to regulate price advertising in one industry but not in others, because the risk of fraud (one of the characteristics of commercial speech that justifies depriving it of full First Amendment protection), is in its view greater there.”) (citations and internal references omitted).
• 153Flake, supra note  NOTEREF _Ref30887250 \h  \* MERGEFORMAT 45 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380037003200350030000000 , at 1084. To be sure, some jurisdictions prohibit both reliance on, and inquiries about, certain arrest or other criminal records. See supra note  NOTEREF _Ref53184101 \h 43 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003100380034003100300031000000 and accompanying text. And although, as discussed in Part I, many jurisdictions prohibit both reliance on, and inquiries about, applicants’ salary history, some prohibit inquiries about salary history without prohibiting reliance on such information in employment decisions. E.g., Or. Rev. Stat. § 659A.357 (West 2017).
• 154Cent. Hudson Gas, 447 U.S. at 564 (1980).
• 155Lorillard Tobacco Co. v. Reilly, 533 U.S. 525, 555–56 (2001). There the Court considered a challenge to a state law that restricted the use of billboards to advertise tobacco products within 100 feet of schools and parks to discourage young people from using tobacco. It found that the state had demonstrated a sufficiently direct link between tobacco advertising and minors’ tobacco use. Id. at 561. But it ultimately concluded that the law failed the narrow tailoring requirement because those restrictions operated as essentially a complete ban on advertising a product lawfully used by adults (due to urban density, for example, no space within the city of Boston would be available for tobacco billboards under the statute). Id. at 561, 565.
• 156See Bd. of Trs. of State Univ. of N.Y. v. Fox, 492 U.S. 469, 478, 480 (1989) (“[This Court has] not insisted that there be no conceivable alternative, but only that the regulation not ‘burden substantially more speech than is necessary to further the government’s legitimate interests . . . . and [the Court has] been loath to second-guess the Government’s judgment to that effect.”) (citations omitted).
• 157See, e.g., King v. Gen. Info. Servs., Inc., 903 F. Supp. 2d 303 (E.D. Pa. 2012) (holding that Fair Credit Reporting Act requirement that credit reports exclude outdated arrest record information regulates accurate commercial speech and thus triggers Central Hudson intermediate scrutiny, and then upholding the provision under that scrutiny); see also Greater Phila. Chamber of Commerce v. City of Philadelphia, 949 F.3d 116 (3d Cir. 2020) (holding that the city’s law prohibiting employers’ inquiries about applicants’ salary history survived Central Hudson intermediate scrutiny); Lester-Abdalla, supra note  NOTEREF _Ref30882388 \h  \* MERGEFORMAT 49 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380032003300380038000000 (proposing that salary history laws should trigger, and survive, intermediate scrutiny).
• 158See Cal. Lab. Code § 432.3(c) (West 2019) (requiring employers to provide information about their pay scales upon an applicant’s reasonable request); Rebecca Greenfield, Making Salary Information Public Helps Close the Gender Pay Gap, Bloomberg Law News (Dec. 5, 2018), https://www.bloomberg.com/news/articles/2018-12-05/making-salary-information-public-helps-close-the-gender-pay-gap [https://perma.cc/5W7E-QHJP] (citing a study by Columbia University and University of Copenhagen researchers that found a seven percent reduction in the pay gap between men and women after Danish law required employers to disclose pay data by gender).
• 159See Harris, supra note  NOTEREF _Ref30882064 \h  \* MERGEFORMAT 51 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380032003000360034000000 , at 4 (“In the U.S. labor market, information on wages and compensation is decidedly asymmetric. Employees frequently do not know how their pay compares to comparable workers, either within or outside their firm, and are reluctant to seek this knowledge out of fear of retaliation, social norms, or general inertia. In stark contrast, many employers use compensation surveys to know precisely where their workers fall in the distribution of wages. In other markets characterized by asymmetric information, the entity with more complete information maintains a distinct advantage.”); Lobel, supra note  NOTEREF _Ref30850389 \h  \* MERGEFORMAT 46 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800350030003300380039000000 , at 549 (“[A] central innovation of the new laws is to reverse information flows in the wage market. Efforts to eradicate wage discrimination have failed in large part due to information asymmetries and difficulties in identifying and proving discrimination.”).
• 160Sylvia A. Law, Income Disparity, Gender Equality, and Free Expression, 87 Fordham L. Rev. 2479, 2494 (2019).
• 161See Norton, supra note  NOTEREF _Ref53245558 \h 61 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003200340035003500350038000000 , at 32–33.
• 162See Leslie G. Jacobs, Compelled Commercial Speech as Compelled Consent Speech, 29 J.L. & Pol. 517, 522 (2014) (“This government regulatory power to require the disclosure of facts material to informed consent is not limited to commercial contracts. Consent is a crucial element that renders many types of transactions legal and enforceable. Governments have always had the authority to define the facts that must be communicated and the circumstances that must exist to create this critical element of consent.”); Andrew Tutt, Commoditized Speech, “Bargain Fairness,” and the First Amendment, 2017 BYU L. Rev. 117, 148 (2017) (“Commentators have been puzzled for decades by the fact that some areas of intensely content-based speech regulation remain subject to, at best, modest First Amendment scrutiny. But a judicial concern for ensuring bargain fairness readily explains the lack of rigor. The purpose of the measures in question is to level the bargaining positions of the parties, thereby helping individuals to obtain a better deal in circumstances of significant information asymmetry.”).
• 163See Ryan Calo & Alex Rosenblat, The Taking Economy: Uber, Information, and Power, 117 Colum. L. Rev. 1623, 1631 (2017) (“The law of consumer protection has long concerned itself with information and power asymmetries among market participants.”).
• 164Zauderer v. Office of Disciplinary Counsel, 471 U.S. 626, 651 (1985).
• 165Id.
• 166E.g., CTIA v. City of Berkeley, 928 F.3d 832, 842 (9th Cir. 2019) (applying Zauderer analysis to permit the government to “compel truthful disclosure in commercial speech as long as the compelled disclosure is ‘reasonably related’ to a substantial government interest, and involves ‘purely factual and uncontroversial information’ that relates to the service or product provided”) (citations omitted); Am. Meat Inst. v. U.S. Dep’t of Agric., 760 F.3d 18 (D.C. Cir. 2014); N.Y. State Rest. Ass’n v. N.Y.C. Bd. of Health, 556 F.3d 114, 133 (2d Cir. 2009).
• 167See supra notes  NOTEREF _Ref53245573 \h 158 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003200340035003500370033000000 –61 and accompanying text.
• 168Nat’l Inst. of Family & Life Advocates v. Becerra, 138 S. Ct. 2361, 2372 (2018) (quoting Zauderer as permitting government to require commercial actors to disclose factual and uncontroversial information).
• 169Norton, supra note  NOTEREF _Ref53245558 \h 61 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350033003200340035003500350038000000 , at 75–76; see also Jonathan H. Adler, Compelled Commercial Speech and the Consumer “Right to Know”, 58 Ariz. L. Rev. 421, 438–39 (2016) (urging that compelled commercial disclosures receive heightened scrutiny but concluding that many such disclosures will survive such scrutiny, especially when motivated by government’s substantial interests in consumer protection or regulatory enforcement).
• 170See Jessica L. Roberts, The Genetic Information Nondiscrimination Act as an Antidiscrimination Law, 86 Notre Dame L. Rev. 597, 643 (2011) (“The ADA is, by and large, an antisubordination statute. It seeks to elevate the status of a particular historically disadvantaged group: people with disabilities.”); id. at 646 (explaining that prohibiting employer inquiries about workers’ disability status while permitting workers to disclose their status helps achieve both anticlassification and antisubordination goals). Legal scholar Bradley Areheart has advocated a similar approach to the Genetic Information Nondiscrimination Act in which gatekeepers would be forbidden from inquiring about applicants’ genetic information to prevent discrimination, but applicants could disclose such information when doing so enabled reasonable accommodation or other equality goals. Bradley A. Areheart, GINA, Privacy, and Antisubordination, 46 Ga. L. Rev. 705, 706 (2012).
• 171See Del. Code Ann. tit. 19, § 711(i) (making it unlawful for employers to forbid employees from talking about their pay with other workers); see also Lobel, supra note  NOTEREF _Ref30850389 \h  \* MERGEFORMAT 46 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800350030003300380039000000 , at 590 (“Taken together, the salary history inquiry ban and salary co-worker inquiry protection also correct a long-existing non-gender specific, double standard—employers often demand secrecy from their employees and usually do not reveal the pay scale of their employees when they interview but demand salary history.”).
• 172See Cofone, supra note  NOTEREF _Ref30791653 \h  \* MERGEFORMAT 8 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003700390031003600350033000000 , at 165 (“[A possibility] for making this method compatible with affirmative action and other tools that address diversity concerns under an antisubordination logic . . . would be to condition the information flow instead of banning it directly. When dealing with explicitly diversity-concerned decision-makers, information could be released under the condition of a specific use: if active diversity measures are to be established.”); Roberts, Protecting Privacy to Prevent Discrimination, supra note  NOTEREF _Ref30791653 \h  \* MERGEFORMAT 8 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003700390031003600350033000000 , at 2168–69 (stating that this approach “capture[s] the best of both worlds[:] [i]ndividuals could maintain autonomy by deciding how and when to disclose information related to protected status, and potential discriminators would be unable to ask about protected status unless the inquiry were explicitly designed to accommodate or to cultivate diversity”).
• 173See Calo & Rosenblat, supra note  NOTEREF _Ref30886679 \h 163 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600330030003800380036003600370039000000 , at 1631 (“The law of consumer protection has long concerned itself with information and power asymmetries among market participants.”).
• 174See 42 U.S.C. § 2000e-2(k)(A)(i) (prohibiting an employer from using an employment practice that disproportionately excludes or disadvantages protected class members unless the employer can “validate” the practice — i.e., unless it can show that the practice is “job related for the position in question and consistent with business necessity”).
• 175See, e.g., Johnson v. Transp. Agency, 480 U.S. 616 (1987) (upholding county’s consideration of sex or race as a plus-factor in promotions to remedy substantial underrepresentation of women and people of color in traditionally segregated jobs); United Steelworkers of Am. v. Weber, 443 U.S. 193 (1979) (upholding collective bargaining agreement’s dedication of a certain percentage of openings in training programs to African-American workers to break down longstanding patterns of racial hierarchy within those jobs).
• 176Facts About Race/Color Discrimination, Equal Employment Opportunity Commission, https://www.eeoc.gov/eeoc/publications/fs-race.cfm [https://perma.cc/27C7-5Y47]; see also Enforcement Guidance: Preemployment Disability-Related Questions & Medical Examinations, Equal Employment Opportunity Commission (1995), www.eeoc.gov/policy/docs/preemp.html [‌h‌t‌tp‌s‌:‌/‌/‌p‌e‌r‌m‌‌a.‌cc/U8E5-BN9S] (explaining that the ADA permits federal contractors to invite applicants or employees to voluntarily self-identify as individuals with disabilities for purposes of complying with federal law that requires federal contractors to engage in affirmative action that may require the collection of applicant data).