Compelled Speech at Work: Employer Mobilization as a Threat to Employee Speech Rights

Employer mobilization has a long history in the United States. Throughout the nineteenth century, it was common for employers to try to sway the votes of their employees. Employers “often publicly announced their political preferences before an election and sometimes offered inducements to their men to vote for a particular ticket” (Bensel 2004, p. 78). And many employers actively coerced their workers to vote in whatever way the company preferred. Especially prior to the adoption of the secret ballot, it was common for employers to watch how their employees voted and to threaten to fire any employees who voted the “wrong” way (Cohn-Postar 2021; Hill 2005, p. 122). Some employers would even march their employees to the polls and give them pre-marked ballots to cast (Cohn-Postar 2021). The prevalence of company towns also meant that many polling places were on land owned by companies. Some employers used this to restrict access to the polls, ensuring that only those who would vote as the company wanted were allowed to vote at all (Cohn-Postar 2021; Keyssar 2001, pp. 129–130).

Thanks to the adoption of the secret ballot and other electoral reforms of the late nineteenth and early twentieth century, employers can no longer exert such direct control over workers’ ballots (Cohn-Postar 2021; Keyssar 2001). But employers still can, and frequently do, monitor the political activity of their workers and attempt to shape that activity through a variety of means. Using a series of systematic surveys, Hertel-Fernandez found that nearly half of all corporate managers in the U.S. tried to politically mobilize their employees in some way, and over a quarter of managers “actively encouraged workers to back particular politicians” (Hertel-Fernandez 2018, p. 45). About a quarter of employees say their boss has contacted them about a political issue, and significant numbers of employees report that their bosses have asked them to donate to particular candidates, contact their elected representative about an issue facing the company, volunteer for a candidate’s political campaign, or engage in some other political activity (Hertel-Fernandez 2018, p. 71–76). Moreover, employer mobilization appears to be highly effective. Corporate managers rank it as one of their most effective tools for shaping law and policy—more effective than political donations, purchasing ads, or participating in the U.S. Chamber of Commerce (Hertel-Fernandez 2018, pp. 47–49).

Contemporary employer mobilization efforts are enabled in large part by the ability of employers to track employees’ political activities. This is easy to do because of digital technologies that enable employers to closely monitor nearly everything that employees do at work, as well as what employees post on social media and other online platforms outside of work (Abril et al. 2012; Grisold et al. 2024). Employers can readily monitor employees’ social media to see whether employees post political content that is in line with what the company wants. Widely available software platforms enable employers to target political mobilization campaigns toward particular employees who will be most susceptible, and to track whether employees engage in a wide range of requested behaviors, such as emailing elected representatives or watching company-approved political videos (Hertel-Fernandez 2018, pp. 51–57). For example, “when an employer sends an email inviting employees to a political event, the employer is able to see whether an employee has read the email, clicked through to the invitation, and whether they have RSVP’d to the event” (Hertel-Fernandez and Secunda 2016, p. 8). Employers can also determine whether or not an employee voted (but not how they voted) by pulling public voting records (Hertel-Fernandez 2018, p. 53). And employers can track whether employees donate money to the company P.A.C. (Conlin & Lozada 2015).

Employer mobilization practices pose a threat to employees’ right against compelled speech because, at least in the United States, employers have the power not only to monitor employees’ political activities, but also to compel them. As a Note in Harvard Law Review explains, “under current federal law, corporations … may compel rank-and-file employees to carry out the entity’s political advocacy just as they may compel employees to carry out other activities for the organization” (Citizens United at Work 2014, p. 674; see also Hertel-Fernandez and Secunda 2016). And U.S. law also gives employers broad freedom to sanction employees for their political activities or lack thereof. In the U.S., most private-sector employees are employed “at will.” That means that they can have their wages reduced, their hours cut, be demoted, or be fired for almost any reason—or even no reason—so long as it is not on the basis of characteristics protected by anti-discrimination laws or for other specific reasons forbidden by law (Anderson 2017; Radin and Werhane 2003). Political identity is not among those specific reasons forbidden by federal law, so under federal law it is legal for employers to sanction employees for their political views or activities (Cox 2015; Hertel-Fernandez 2018, pp. 216–217). Some states have laws that forbid sanctioning employees on the basis of their political expression, but most only apply to political expression off the job, and they do not protect workers against being sanctioned for refusing to engage in political expression as directed by their employer (Citizens United at Work 2014; Hertel-Fernandez and Secunda 2016). Employers are thus empowered to reward employees that participate in mobilization efforts and to sanction those who refuse. Employees may thereby be compelled to voice or support political messages that they do not want to voice or support. In other words, employees may be compelled to speak.

Forcing someone to speak is not always wrong. For example, witnesses can be required to testify in judicial proceedings; food and beverage companies can be required to disclose nutrition information about their products; and citizens can be required to report their income for tax collection purposes. The fact that a speaker has been required to say something—even when that requirement is backed by the threat of sanctions—is not sufficient to show that the speaker’s right against compelled speech has been infringed (Shiffrin 2014). Thus, the mere fact that an employer encourages or requires an employee to say something is not sufficient to constitute a violation of the employee’s right against compelled speech.

In genuine cases of compelled speech, the victim is not only forced to speak, but also suffers an attendant moral wrong. Sepinwall (2020) distinguishes three different wrongs that have been articulated in the literature: misattribution, thought control, and commandeering. My aim in the next two sections is to articulate a pair of conditions that distinguish employer mobilization practices that are liable to inflict those wrongs on employees from practices that are not. But first, I will say more about each of the wrongs in turn.

Misattribution occurs when support for a message is attributed to a speaker who does not in fact support the message and has only expressed it because they were compelled to do so. In such cases, the beliefs and values conveyed by the message are liable to be misattributed to the speaker who does not in fact hold those beliefs and values (Shiffrin 2014, p. 505; Sacharoff 2008, p. 367). For example, if someone pledges allegiance to a political party only because they are coerced to do so, then others who are unaware of the coercion may misattribute genuine support for the party, along with its beliefs and values, to the speaker. Such misattribution involves a wrong to both the speaker and the listener. The speaker is wronged because such misattribution infringes the speaker’s “freedom not to endorse—or be perceived as endorsing—messages [they] do not believe” (Greene 2018, p. 1491). And the listener is wronged insofar as compelled speech “misleads [the listener] into concluding that the speaker believes the message he or she has been compelled to utter,” which may in turn mislead the listener as to the source, popularity, and veracity of the message (Sacharoff 2008, p. 385).

Thought Control occurs when someone is forced to express or support a message in a way that is liable “to interfere with the autonomous thought processes of the compelled speaker” (Shiffrin 2011, p. 302). Such interference can occur in different ways. Forcing someone to regularly express a particular message may have the effect of making its content seem intuitively true, or it may make the message feel familiar such that it becomes “a comfort and an internal source of authority for consultation” (Shiffrin 2005, p. 854). A speaker that is forced to express a message may also experience cognitive dissonance rooted in the mismatch between what they have been made to say and what they actually believe. Over time, the speaker may unconsciously resolve that dissonance by coming to believe the content of the message (Shiffrin 2005, p. 859). On the basis of a concern about such thought control, the U.S. Supreme Court long ago ruled that students cannot be required to recite the Pledge of Allegiance or to salute the flag. In the words of the majority ruling, such a requirement wrongfully “invades the sphere of intellect and spirit” (West Virginia State Board of Education v. Barnette, 319 U.S. 624 (1943), p. 642).

Commandeering occurs when someone is conscripted against their will to convey or support another person’s ideological message (Tribe 2001, p. 645; Shiffrin 2014, p. 506). Such commandeering wrongly treats the compelled speaker as a means to an end, hijacking the person’s voice, body, or resources to advance an ideological message that the victim does not endorse. It is an affront to the victim’s dignity and autonomy (Greene 2018, p. 1493; Shiffrin 2014, p 506). For example, it would be wrong to force a religious devotee to foreswear their religion, even if the context made it unlikely that anyone would believe that the devotee had actually renounced their faith, and even if the devotee’s religious beliefs are unlikely to be influenced. It would also be wrong to force an avowed atheist to donate money to the devotee’s church. In the words of Thomas Jefferson, “to compel a man to furnish contributions of money for the propagation of opinions which he disbelieves and abhors, is sinful and tyrannical” (Koch and Peden 1998, p. 289). The U.S. Supreme Court quoted the same line from Jefferson when it ruled that a state bar association—which attorneys in the state are legally required to join—could not use mandatory membership dues to fund political advocacy that some of the members objected to (Keller v. State Bar of California, 496 U.S. 1 (1990)).

I now turn to the task of articulating a set of conditions to identify employer mobilization practices that are liable to infringe on employees’ rights against compelled speech. The first of those conditions is:

This condition is necessary (but not sufficient) for employees to suffer one or more of the wrongs of compelled speech because those wrongs all involve a speaker being forced to express a particular message that does not accord with the speaker’s actual beliefs and values. In cases of misattribution, support for the particular message (and the beliefs and values it conveys) are incorrectly attributed to the speaker by others (Sacharoff 2008, p. 367). In cases of thought control, the compelled expression of a particular message distorts the speaker’s thinking by way of the incongruence between the content of the message and the speaker’s beliefs and values (Shiffrin 2005). And in cases of commandeering, the wrong is that a speaker is forced to express a particular message that does not reflect their own beliefs and values (Tribe 2001, p. 645; Shiffrin 2014, p. 506). So only practices that meet condition (1) are liable to inflict the wrongs of compelled speech on employees.

Many common employer mobilization practices clearly meet condition (1). For example, one common practice is to ask employees to send letters (or emails) to their elected representatives supporting policies favored by the employer. This is often facilitated by online systems that enable employees to sign and send letters that are pre-written for them. For example, in 2009 the telecommunications company AT&T mobilized its employees to write thousands to letters to the Federal Communications Commission (FCC) opposing a proposed rule change that would affect the company (Hertel-Fernandez 2018, p. 59–60). Such practices clearly ask employees to express particular political messages. And given political diversity among employees, those particular messages will often be at odds with some employees’ beliefs and values. So such practices meet condition (1). Another common practice is to ask employees to donate to company-affiliated political action committees (P.A.C.s) that support particular political causes. For example, the energy company BP encourages its salaried employees to donate to the company’s P.A.C. and gives preferred parking spots to employees who donate at least 2.5 percent of their salary (Conlin & Lozada 2015). By doing so, BP encourages their employees to support the particular political messages sponsored by the company P.A.C.—messages that not all BP employees will agree with. So, such practices also meet condition (1).

But not all employer mobilization practices meet condition (1). For example, many companies take measures to encourage and enable their employees to vote. For example, Patagonia, Levi Strauss, and PayPal have co-founded an initiative to increase electoral participation by informing employees about how to vote and offering paid time off on election day (Todd-Ryan 2020). As part of such efforts, employees may receive emails or other communications from their employer that explain the importance and value of voting, or that provide information about how to register to vote and ways of voting. These communications do not tell employees who or what to vote for; they provide “exclusively information about registering or turning out to vote” (Hertel-Fernandez 2018, p. 45). Voting is often understood to be an expressive act (e.g. Brennan and Hamlin 1998; Hamlin and Jennings 2011). So such voter turnout initiatives can reasonably be seen as encouraging employees to engage in political expression. However, such initiatives do not meet condition (1) because they do not encourage employees to express any particular political message. The message that is conveyed by casting a ballot generally depends on who or what one votes for.¹ So, by encouraging employees to vote, but not telling employees who or what to vote for, employers encourage political expression, but they do not ask or encourage employees to express any particular political message.

Some companies not only encourage their employees to vote, they also tell their employees how the company thinks they should vote (Hertel-Fernandez 2018, p. 73–77). For example, Wynn Resorts distributed a voter guide to all of its employees in Nevada that not only provided instructions on how to vote, but also included a list of specific candidates that the company encouraged employees to vote for (Hertel-Fernandez 2018, p. 18). By encouraging employees to vote for particular political candidates, such voter guides encourage employees to express particular political messages that are liable to be at odds with some employees’ beliefs and values. Such voter guides thus meet condition (1).

Another common form of employer mobilization takes the form of “employee education” (Hertel-Fernandez 2018, p. 21). Such initiatives typically ask employees to attend employer-sponsored talks about politics, to participate in reading groups that feature texts on political issues, or to complete a set of online modules comprising a kind of civics education course (Hertel-Fernandez 2018, p. ix, 21). Do such initiatives meet condition (1)?

Not necessarily. Asking someone to read or listen to a message is not the same as asking or encouraging them to express that message. For example, I require some of my students to read Milton Friedman’s (1970) article arguing that “the social responsibility of business is to increase its profits.” But by requiring my students to read Friedman’s text, I do not thereby ask or encourage them to express agreement with Friedman’s message. And notice that, requiring students to read Friedman’s text does not inflict any of the wrongs of compelled speech upon them. Friedman’s message is not liable to be misattributed to my students simply because they have read Friedman’s article. Requiring my students to read the article is not a form of thought control because it does not try to distort the students’ thinking “in a way that bypasses [their] deliberative faculties” (Shiffrin 2005, p. 302). Indeed, my aim is precisely to engage the students in rational deliberation. And requiring students to read Friedman’s article does not commandeer their voices, bodies, or resources to express a message that they do not want to express. Also note that all of the above holds even if some students are ultimately convinced by Friedman’s argument and choose to express agreement with Friedman’s message.

However, the situation changes if I not only require students to read Friedman’s text, but also communicate to them (explicitly or implicitly) that I want them to agree with Friedman. I may do this by, for example, praising every student that expresses agreement with Friedman while pushing back sharply against any student that criticizes Friedman’s view. Or I may give better grades to students who express agreement with Friedman, or frame an exam question in a way that signals that I want the students to express support for Friedman’s view. Because I would thereby be asking or encouraging my students to express a particular political message, I would be liable to inflict the wrongs of compelled speech on my students insofar as some students are forced to express agreement with Friedman’s view despite holding contrary beliefs and values.

Similarly, an employer that asks or encourages employees to complete some form of civic education does not necessarily thereby ask or encourage employees to express any particular political message. For example, many companies encourage their employees to attend company-sponsored seminars about how democracy works, ways to recognize political misinformation, and other civics issues (Eddy 2023). Such initiatives do not meet condition (1) insofar as they do not ask or encourage employees to express any particular political messages that are at odds with employees’ beliefs and values.

However, many employee education initiatives are significantly skewed toward particular political views and intended to get employees to endorse those views. And the context in which employees are asked to complete such programs may often signal to employees that their employer expects them to express particular political messages. Indeed, Hertel-Fernandez provides evidence that many employers develop such employee education initiatives with the aim of getting employees to advocate for or against particular policies and politicians (Hertel-Fernandez 2018, pp. 58–59). In such cases, employee education campaigns do meet condition (1).

For example, Menards, a chain of large home improvement stores, asked their employees to complete a company-sponsored civics class that emphasized the importance of limited government and low taxes. The materials instructed employees that taxes are inimical to freedom, that taxes on business are too high, and that it is best for the country to support political candidates who favor lowering business taxes. Completing the class was officially optional, but managers closely tracked which employees passed the course, and employees that passed were recognized in company publications (Hertel-Fernandez 2018, p. 21). Menards’ employee education program meets condition (1) because it goes beyond simply asking employees to read or review materials about politics. Both the content of the program, and the communications from management about the program, signal to employees that their employer wants them to express particular political views and support particular kinds of political candidates.

While Menards’ employee education program and many other employer mobilization practices meet condition (1), that, by itself, does not entail that those practices violate employees’ speech rights. That is because, in cases of compelled speech, a speaker is forced to express some particular message that they do not want to express (Greene 2018, p. 1528). This is what makes compelled speech compelled. So, for an employer mobilization practice to infringe on employees’ right against compelled speech, it is not sufficient that the practice meets condition (1). The practice must also meet a second necessary condition:

In other words, employers must do more than merely ask or encourage their employees to express a particular message; they must force their employees to do it.

Adopting a definition from Michael Kates (2015), I take it that a person is forced to do something if and only if the person chooses to do it because they have no acceptable alternative. For example, Peter is forced to hand over his wallet to an armed robber if and only if Peter chooses to do so because he has no acceptable alternative. This is not to say that Peter has no alternatives. Peter could defy the armed robber and risk serious injury or death; but doing so is not an acceptable alternative to handing over his wallet. Applied to speech, this means that a person is forced to express a message if and only if refusing to do so is not an acceptable alternative for that person. What makes an alternative unacceptable? Following Kates, I take an alternative to be unacceptable “just in case it falls below a certain minimal threshold of basic human need” (Kates 2015, p. 374). Exactly where to set the threshold is difficult to determine, but for present purposes I take an alternative to be unacceptable if (but not only if) it involves the prospect of serious financial hardship—hardship of a kind that makes it difficult, if not impossible, for a person to live a decent life. This means that employees are forced to express a message whenever they are made to choose between either expressing the message or else facing the prospect of serious financial hardship.

The most obvious and direct way that employers can force employees to express a message is by threatening to sanction employees that refuse. As explained above in the “Employer Mobilization” section, most private-sector employees in the U.S. are employed “at will.” That means that employers can generally cut employees’ wages, reduce employees’ hours, demote employees, and even fire employees for refusing to participate in employer mobilization initiatives (Citizens United at Work 2014; Hertel-Fernandez and Secunda 2016). And, for many employees, suffering such a sanction is liable to lead to financial hardship, thereby making refusing to express their employer’s message an unacceptable alternative.

Consider, for example, the case of mining company Murray Energy. In 2012, the company hosted a campaign rally for then U.S. presidential candidate Mitt Romney. Employees were told that attendance was mandatory, and many workers “told reporters that they showed up the to rally even if they opposed Romney out of fear of losing their jobs or being disciplined in some other way” (Hertel-Fernandez 2018, p. 69). Subsequent investigative journalism revealed that, for years, Murray Energy had often called out, demoted, denied bonuses to, and even fired, employees who did not donate to the company’s political action committee or regularly participate in political fundraising events (MacGillis 2012). In the words of one employee “if you don’t contribute, your job’s at stake. You’re compelled to do this whether you want to or not” (MacGillis 2012). Murray Energy’s employer mobilization practices clearly meet condition (2) because the company threatened to (and did) impose severe sanctions on employees who refused to participate—sanctions that were liable to impose serious financial hardship on affected employees. In other words, Murray Energy forced their employees to express the company’s political messages.

At this point one might object that, even when employers threaten employees with sanctions, employees are still not genuinely compelled to speak because employees can always simply quit and find a different job. In other words, one might insist that employees always have an acceptable alternative to complying with employer mobilization demands—they can always just quit. The problem with this objection is that, for many workers, quitting and finding another job is not a viable option. This may be so for several different reasons. First, the majority of Americans live paycheck to paycheck (Morris 2023), and most Americans do not have sufficient cash reserves to cover $1000 worth of emergency expenses (Thier 2024). That means that many Americans are not in a position to quit their current job unless they can immediately start a new job that pays at least as much—something that many employees will not be able to do, especially in times when the job market is poor. Second, about 20 percent of U.S. employees are subject to noncompete clauses that forbid them from working for any other company in the same industry (or even related industries) for months, or even years, after leaving their job (Anderson 2017; Coy 2021; Dougherty 2017).² Third, many employees in a wide range of industries, from nurses, to electricians, to bank tellers are subject to so called “stay-or-pay” clauses—also known as Training Repayment Agreement Provisions, or T.R.A.P.s—that require employees to pay their employer the cost of the on-the-job training they received (often tens of thousands of dollars) if they quit or are fired before working some allotted amount of time (usually months, or even years) for the company (Kaiser-Schatzlein 2023). Thus, when an employer threatens to sanction an employee for refusing to express or support a particular political message, quitting to find another job is simply not an acceptable alternative for many workers. And many workers cannot afford to ignore the threat of employer-imposed sanctions for refusing to participate in the company’s political advocacy efforts because getting fired, or even having their hours and pay reduced, will quickly lead to serious financial hardship.

The forgoing points are reinforced by Hertel-Fernandez’s studies. He finds that “political mobilization of workers by managers is heavily structured by the imbalance of economic power between them,” (Hertel-Fernandez 2018, p. 25). Employers tend to avoid pushing employees to support particular politicians or causes when doing so is likely to spark employee backlash. To avoid such backlash, employers typically focus their mobilization efforts on employees who lack the power to meaningfully resist. Hertel-Fernandez found that lower income workers tend to be significantly more fearful of losing their job or suffering other workplace sanctions, and that those same workers are also more likely to be targeted by, and participate in, employer mobilization initiatives (Hertel-Fernandez 2018, pp. 62–66, 91–98). So, in practice, when employers threaten to punish employees for failing to express or support the company’s political message, the employees are frequently not in a position where they can resist their employer’s demands without suffering the risk of serious financial hardship. Hence, for many workers, refusing to participate in their employer’s mobilization efforts is not an acceptable alternative.

Hertel-Fernandez suggests that the fact that most employees are employees “at will” means that employees are almost always faced with the threat of being sanctioned if they refused to participate in an employer mobilization initiative. “As a result of American employment law,” he says, “political messages and mobilization efforts from top managers to workers carry a potential threat of retaliation—whether managers articulate that threat directly or not” (Hertel-Fernandez 2018, p. 25). But having the legal latitude to impose sanctions is different from actually threatening employees with sanctions. The mere fact that an employer could legally punish workers who refuse to support the company’s political message does not imply that the employer is likely to mete out such punishment, nor that employees believe that there is a threat of such punishment. In other words, the fact that an employer could, legally, force its employees to express a particular message does not imply that it does in fact force them to express that message. Even if employees are employed “at will,” their employer could still run mobilization initiatives in a way that makes participation genuinely voluntary.

Consider, for example, BP’s practice of giving preferred parking spots to employees who donate at least 2.5 percent of their salary to the company’s P.A.C. (Conlin & Lozada 2015). BP clearly incentivizes employees to support its political messages. But insofar as the loss of a preferred parking spot is the only cost imposed on employees who decline to participate, BP’s incentive system does not force their employees to donate. That is because forgoing a preferred parking spot is an acceptable alternative to contributing to the company P.A.C.. Similarly, GSK (formerly GlaxoSmithKline) encourages employees to express its political messages using a point system that rewards employees for performing various actions such as calling or meeting with elected officials, or volunteering for the political campaign of one of the company’s preferred candidates. At the end of the year, the employee with the most points in each sales region wins an all-expenses-paid trip to Washington D.C. (Hertel-Fernandez 2018, p. 55). This system incentivizes employees to participate, but it does not force employees to participate since losing out on the chance of a free trip to D.C. is an acceptable alternative. And note that the above points hold even if the relevant BP and GSK employees are employees “at will”. So to meet condition (2), it is not sufficient that employees be employees “at will”. Nor is it sufficient that employers incentivize speech so long as forgoing the promised rewards is an acceptable alternative for employees.

That does not mean that an incentive system cannot force employees to express an employer’s message. There are cases where forgoing a promised reward is not an acceptable option. For example, suppose than an employee is already experiencing severe financial hardship. And suppose that the employee’s employer offers the employee a significant cash bonus on the condition that the employee express a particular political message for the employer. Insofar as the employee’s only means of escaping financial hardship is to express their employer’s message to receive the bonus, the employee is forced to express the message. That is because not expressing it, and thereby forgoing the bonus, is not an acceptable alternative. Returning to the example of Murray Energy, even apart from the threat of being fired, the fact that the company made bonuses conditional on employees participating in the company’s political activities would, on its own, be sufficient to meet condition (2) if some employees needed those bonuses to avoid severe financial hardship.

Another way that employers may motivate employees to engage in politics, beyond using incentives or sanctions, is to warn employees about adverse consequences, including job losses, wage reductions, and benefit cuts, that may result if certain political candidates are elected or specific policies are implemented. For example, in 2020, the president of Daniels Manufacturing Corporation warned the company’s employees that the company might have to lay off employees if Donald Trump lost his campaign for re-election (Gillespie 2020). Hertel-Fernandez and Secunda suggest that such warnings infringe employees’ speech rights because workers might feel “undue pressure to support employer positions with which they otherwise disagree… in order to keep their jobs or prevent wage loss” (Hertel-Fernandez and Secunda 2016, p. 9–10). And the empirical evidence shows that workers who receive such warnings are significantly more likely to fear for their jobs and to participate in employer mobilization initiatives (Hertel-Fernandez 2018, p. 77–81). However, issuing such warnings does not necessarily meet condition (2). That is because warning employees that layoffs, wage cuts, or other adverse consequences are likely to result from a certain electoral or policy outcome does not necessarily constitute forcing employees to express the company’s political message.

For one thing, it may be true that a particular electoral or policy outcome is likely to force a company to lay off employees or engage in other cost-cutting measures. For example, a coal mine might be forced shut down and lay off all of its employees if new environmental regulations reduce demand for coal to the point where it is impossible to profitably operate the mine. Hertel-Fernandez suggests that the accuracy of such warnings is “somewhat beside the point” because “what matters is the fact that workers recognize that employers are uniquely positioned to act on those predictions” (Hertel-Fernandez 2018, p. 81). But warning an employee that they may lose their job if a particular policy is enacted is not the same as threatening to sanction that employee if they do not speak out against the policy.

Suppose, for example, that Kenji works for an electric vehicle company. A major tax rebate for electric vehicles is about to expire. Kenji’s manger warns all the employees that if the tax rebate expires, demand for electric vehicles will fall sharply, which may lead to layoffs at the company. The manager urges Kenji and his fellow employees to contact their elected representatives to advocate for the renewal of the tax rebate. The manager does not suggest that there will be any specific sanctions targeted at employees who do not advocate for renewing the tax rebate. What is communicated is just a warning that many employees’ jobs will be at risk if the rebate is not renewed. Fearing for his job, Kenji emails his elected representative to advocate for the tax rebate. Did Kenji’s manager force Kenji to advocate for the tax rebate? I contend that the answer is no. The reason is that the threat to Kenji’s job is not created by his employer. In other words, the threat to Kenji’s job is independent of his employer’s warning about that threat. Of course, if Kenji is laid off, it will be his employer that does it. But the reason that Kenji’s job is at risk is not that his employer has threatened to sanction him if he refuses to participate in the company’s mobilization initiative. Rather, it is that a pending public policy decision may necessitate cost-cutting at the company. Thus, Kenji’s employer does not meet condition (2) because the threat to Kenji’s job is not employer-imposed. Kenji’s advocacy for the tax rebates is thus not compelled by his employer.

That said, in some contexts, warnings do meet condition (2). That is because warnings about potential layoffs may often act as coded threats that employees will be sanctioned if they fail to participate in mobilization efforts. When such a warning is issued by an employer that has a history of sanctioning employees for refusing to support the company’s political messaging, then the warning is likely to be interpreted as not just a warning, but a threat. Kenji’s situation would be very different if he reasonably believed that the most immediate threat to his job was not the pending expiry of the tax rebate, but rather the risk of sanctions imposed by his employer if he does not advocate for renewing the rebate. In that case, Kenji’s employer would meet condition (2), and Kenji’s advocacy would be compelled if he acts out of fear of being punished by his employer for failing to express his employer’s message.

In other cases, it may be very unclear whether a warning meets condition (2) because it will often be unclear to what extent the threat to employees’ jobs is within the employer’s control. Laying off employees or engaging in other forms of cost cutting is often not the only viable way that employers can respond to changes in policy that increase their costs. In many cases, employers can offset those increased costs by passing them on to consumers in the form of higher prices, or by other means. So employers often exercise some degree of choice as to whether a particular political outcome results in layoffs or other costs to employees. In such cases, it may be difficult to determine the extent to which an employer is responsible for the threat to employees’ jobs, and thus difficult to determine whether the employer is forcing employees to express a particular message.

I have articulated two conditions, (1) and (2), that aim to identify employer mobilization practices that are liable to inflict the wrongs of compelled speech on employees. According to those conditions, employees’ rights against compelled speech are infringed when employers force employees to express particular political messages that are at odds with some employees’ beliefs and values. However, some jobs are essentially political. By that I mean that there are some jobs, such as lobbyist, press secretary, and campaign manager, in which expressing the employer’s political messages is essential to the job itself. Employees in these roles will often be required to express particular political messages, some of which may be at odds with their own beliefs and values. To avoid the erroneous conclusion that it is unethical to fire such employees for refusing to do their jobs, it is necessary to add a third condition to my analysis:

This condition makes it permissible for an employer to sanction an employee for refusing to express a political message when doing so constitutes refusing to do an essential part of the employee’s job.

Suppose, for example, that Sara is employed by Planned Parenthood, an organization that advocates for abortion rights. Part of Sara’s job is to lobby legislators to create legal protections for abortion providers. Now imagine that Sara changes her view and comes to see abortion as deeply wrong. She thereafter refuses to lobby legislators to protect abortion access. I contend that it would be permissible for Sara’s employer to fire her (if she refuses to quit), and that in doing so the organization would not be infringing Sara’s right against compelled speech. The reason for this is that, in taking a job at Planned Parenthood (a political advocacy organization) Sara agreed to participate in that organization’s advocacy. And Sara has no cause for complaint if she is fired for refusing to do what she agreed to do.

Similarly, suppose that Ryan is employed as a lobbyist for an oil company. Ryan’s job is to advocate against regulations that will reduce demand for fossil fuels. Now imagine that Ryan becomes convinced of the need to reduce fossil fuel use in order to avert catastrophic climate change. So Ryan refuses to express his employer’s anti-regulation messages. Since expressing an employer’s political messages is essential to the job of being a lobbyist, the company would not violate Ryan’s right against compelled speech if it sanctioned him for his refusal to express the company’s messages. Again, this is because that is what Ryan agreed to do when he took a job a lobbyist for the company.

Most employees are not like Sara and Ryan; their jobs are not essentially political. So most employees meet condition (3). And managers cannot simply insist that it is part of the job of every employee to support the company’s political messages. Expressing such messages is simply not an essential part of the job of nurses, firefighters, truck drivers, high school teachers, financial analysts, factory workers, miners, airline pilots, and most other employees. That said, I grant that determining whether employees meet condition (3) may not always be easy. For example, it may be unclear whether expressing a company’s political messages is an essential part of the job of mid-level managers. But it will be sufficiently clear in most cases for condition (3) to provide practicable guidance.

Employer mobilization is one of the most common and effective ways that employers seek to influence law and policy (Hertel-Fernandez 2018). I have argued that many employer mobilization practices are unethical because they infringe on employees’ right against compelled speech. More specifically, I have argued that such practices are liable to infringe that right when three conditions are met:

The available empirical evidence suggests that these three conditions are widely met. Hertel-Fernandez’s studies indicate that more than a quarter of American managers encourage their employees to support particular politicians (Hertel-Fernandez 2018, p. 45). And that figure does not include managers that encourage employees to express or support particular political messages without encouraging them to support particular politicians. So the evidence suggests that a significant portion of mangers meet condition (1). Over a quarter of American employees say that their employer is at least “somewhat likely” to retaliate against them for their political views or activities, with over fifteen percent saying they have seen such retaliation (Hertel-Fernandez 2018, p. 89). It is not clear how often such threats rise to the level of being fired or facing other such sanctions that could jeopardize employees’ livelihoods or lead them to experience serious financial hardship. But anecdotal evidence suggests such threats are not rare. So condition (2) also appears to be met by many employers. And, as I said in the last section, condition (3) is met by most employees. The evidence thus indicates that a significant portion of employees—not the majority, but certainly a sizable minority—are subject to employer mobilization practices that infringe on the employees’ rights against compelled speech.

In closing, it is important to note three limits of my analysis. First, as noted in the introduction, my discussion has focused on the American context. While the conditions I have articulated should apply beyond that context, more work is needed to understand the prevalence, uses, and efficacy of employer mobilization in other countries.

Second, it is important to note that my analysis says nothing about the conditions under which employees can be sanctioned for expressing (as opposed to refusing to express) particular political messages. My analysis thus leaves it open to employers to fire (or otherwise sanction) employees who express particular political positions that the employer opposes or deems unacceptable. The analysis does, however, prohibit employers from requiring employees to explicitly repudiate or denounce particular positions. So, while my analysis leaves it open to an employer to fire an employee for expressing a particular political view, it prohibits the employer from forcing the employee to repudiate that view.

Third, I have argued that practices that meet conditions (1) – (3) are unethical because they violate employees’ right against compelled speech. And, conversely, I have argued that practices that do not meet one of more of those conditions are unlikely to infringe on employees’ rights against compelled speech. But the fact that a practice does not infringe on employees’ rights against compelled speech does not imply that the practice is ethically permissible; it may still be unethical for other reasons. So while my argument shows that practices that meet conditions (1) – (3) are unethical, it does not imply that practices that do not meet those conditions are ethical.

Given the prevalence and impact of employer mobilization practices, much more work is needed to determine their ethical limits. This essay contributes just one small piece to that much larger project.