Free Speech

Laws Protecting Private Employees' Speech and Political Activity Against Employer Retaliation: Introduction

These exist in many states and some counties and cities, but are rarely focused on.


Back in 2012, I published an article in the Texas Review of Law & Politics on this subject, trying to describe such laws.

I'm not sure where I stand on them as a policy matter. On one hand, they restrict private employer choice, and can often undermine private employers' legitimate policy interests, much as other antidiscrimination laws do; I'm generally a supporter of employment-at-will as a legal rule, and am not eager to see further restrictions on that rule. On the other hand, to the extent that one is concerned about speech restrictions undermining democratic deliberation and debate, one might be as concerned by private employer retaliation as by, say, government employer retaliation or some other government-imposed restrictions.

But I thought in 2012 that it was worth describing the laws, so that people can know more about them, and more sensibly consider whether there should be more such laws or fewer. And because such matters are especially in the news these days, I thought I'd basically serialize this article on the blog over the next few weeks. Here is the Introduction; as with future posts, I have omitted most footnotes, but you can see them all here.

[* * *]

About half of Americans live in jurisdictions that protect some private employee speech or political activity from employer retaliation. Some of these jurisdictions protect employee speech generally. Others protect only employee speech on political topics. Still others protect only particular electoral activities such as endorsing or campaigning for a party, signing an initiative or referendum petition, or giving a political contribution.

Moreover, though the matter is not clear, federal law may often protect private employees who speak out in favor of a federal candidate. To my knowledge, these protections have not been systematically cataloged, and some have never been cited in a law review article.

Some employee free speech protections were enacted following the Civil Rights Act of 1964, which banned employment discrimination based on race, religion, sex, and national origin, and are modeled on that statute. But many of the protections long preceded the Act, and similar state civil rights laws. Indeed, the first date back to 1868.

These early protections for private employee speech and political action were likely based on the very first American laws banning employment discrimination by private employers—voter protection laws, which barred employers from discriminating against employees based on how the employees voted. (Recall that this was the era before the secret ballot.) As early as the 1700s, several colonies and states barred any "attempt to overawe, affright, or force, any person qualified to vote, against his inclination or conscience,"[1] and some also barred, "after the … election is over, menac[ing], despitefully us[ing] or abus[ing] any person because he hath not voted as he or they would have had him."[2]

These voter protection laws seem to have covered threats not just of physical violence but also of legal coercion,[3] and they may have covered threats of economic retaliation as well—a similarly general 1854 English statute[4] was applied to threats of economic retaliation and not just those of physical attack.[5] The bans on threats, from 1721 to the 1860s, were included alongside bans on bribery; given that offering to provide a financial benefit in exchange for a vote was forbidden, it makes sense that threatening to deny a financial benefit in exchange for a vote would have been forbidden as well.[6]

And some voter protection laws enacted in the mid-1800s explicitly covered threat of economic retaliation. The proposed federal criminal code drafted in 1828 by Edward Livingston—who had earlier participated in drafting the Louisiana Civil Code, was at the time a Congressman (and soon to be Senator) from Louisiana, and would later become Secretary of State—expressly covered "threats of withdrawing custom or dealing in business or trade … or any other threat of injury" aimed at influencing votes. The 1832 proposed D.C. criminal code would have done the same. Laws using this language were enacted in Mississippi (1839), Iowa (1850), the Nebraska Territory (1855), Illinois (1871), and Delaware (1881).

Likewise, in 1839, Pennsylvania expressly barred threats of "loss of any appointment, employment or pecuniary benefit" aimed at "influenc[ing] any voter." Also in 1839, Ohio made it a crime for "any person [to] … use any threat or coercion to procure any voter in his employ … to vote contrary to the inclination of such [employee]." Several years later, Connecticut (1846) and Massachusetts (1852) barred "threatening to discharge [an elector] from … employment" in order to influence a vote.

By the 1860s, some states also barred discrimination based on past votes rather than just threats aimed at future votes. This was especially visible in a burst of such lawmaking in the Reconstruction-era South, triggered by the Republican concern that southern employers were pressuring their employees to vote against the Republicans.[7] (In some instances, Union generals administering the military occupation of the South issued such rules as military orders, violations of which were triable before military commissions.)

It is this post-Reconstruction batch of voter protection laws that led to the first protections that went beyond voting to speech. In 1868, Louisiana and South Carolina banned discrimination against most private employees based on "political opinion." And several decades later, both the voter protection laws (which I will not focus on in this Article) and the statutes protecting political opinion and political activity began to spread to other states.

I am not sure such restrictions on private employers are a good idea. First, employers may have a legitimate interest in not associating themselves with people whose views they despise. Second, employees are hired to advance the employer's interests, not to undermine it. When an employee's speech or political activity sufficiently alienates coworkers, customers, or political figures, an employer may reasonably claim a right to sever his connection to the employee. Perhaps such statutes should not be copied by other states, and perhaps they should even be repealed, which is what happened in 1929 when Ohio repealed its "political activities" statute.

But whether the statutes are sound or not, they strike me as worth investigating. I therefore thought it would be useful to publish a list of the statutes that I could find and a summary of some of the key court decisions interpreting those statutes.

[1]. An Act to Ascertain the Manner and Form of Electing Members to Represent the Inhabitants of this Province, § 9, 1761 Ga. Laws 109; see also An Act to Ascertain the Manner and Form of Electing Members to Represent Inhabitants of this Province, § 14, 1721 S.C. Acts 115 (prohibiting the use of certain threats to influence elections); An Act to Regulate the General Elections of this Commonwealth, § 27, 1785 Pa. Laws 351 (same); An Act to Regulate Elections, ch. 50, § 17, 1800 Md. Laws 30 (same). Other states had similar though slightly differently worded statutes, which banned attempts to "directly or indirectly" influence votes by "bribery[,] menace or other corrupt means or device." An Act to Regulate Elections Within this State, ch. 16, 1778 N.Y. Laws 36; see also An Act Dividing the State into Districts for Electing Representatives, § 12, 1793 Vt. Acts & Resolves 13 (prohibiting bribes and threats made to influence elections); An Act Regulating the General Elections of the Indiana Territory, § 14, 1811 Ind. Acts 234 (same); An Act to Support the Privilege of Free Suffrage in Election, §§ 4–5, 1814 La. Acts 98 (same). The New York and Vermont statutes expressly provided for enforcement by the victim, with half the penalty to be given to the victim. The other statutes were cast as normal criminal statutes, but at the time the norm for criminal law generally was that victims would act as prosecutors….

[2]. An Act to Ascertain the Manner and Form of Electing Members to Represent Inhabitants of this Province, § 14, 1721 S.C. Acts 115; An Act to Ascertain the Manner and Form of Electing Members to Represent the Inhabitants of this Province, § 9, 1761 Ga. Laws 109; see also An Act to Regulate General Elections,  § 17, 1837 Mich. Pub. Acts 206–07 (making it a crime to "on the day of election give any public threat … with a view to obtain any … votes for … [any] candidate"); An Act to Preserve the Purity of Elections,  § 5, 1849 Iowa Acts 133 (making it a crime to threaten or compel any elector to vote against his inclination); An Act to Preserve the Purity of Elections, § 11, 1857 Wis. Sess. Laws 105 (likewise); An Act to Regulate Elections in this State, § 57, 1859 Minn. Laws 161 (likewise).

[3]. Consider Fargues McDowell's prosecution and conviction, described in Right of Suffrage, Niles' Weekly Register, Nov. 25, 1815, at 213–14. McDowell operated a jail in which Jacob Parker was detained before trial. Though Parker had been unable to make bail, McDowell had given Parker a bail-like release (something that a jailer was apparently allowed to do), but then threatened to revoke it if Parker voted for a candidate of whom McDowell disapproved. McDowell was prosecuted under the South Carolina statute and convicted.

[4]. Corrupt Practices Prevention Act, 1854, 17 & 18 Vict., c. 102, § 5 (Eng.), reprinted in Henry Jeffreys Bushby, A Manual of the Practice of Elections in the United Kingdom app. at 28–29 (2d ed. 1865) (barring, in relevant part, "mak[ing] use of, or threaten[ing] to make use of any force, violence, or restraint, or inflict[ing], or threaten[ing], the infliction … of any injury, damage, harm, or loss, or in any other manner practis[ing] intimidation upon, or against, any person, in order to induce or compel such person to vote or refrain from voting, or on account of such person having voted or refrain[ing] from voting, at any election" or "by abduction, duress, or any fraudulent device or contrivance, imped[ing], prevent[ing], or otherwise interfer[ing] with the free exercise of the franchise of any voter").

[5]. Regina v. Barnwell, 5 Weekly Rep. 557 (1857); see also Francis James Newman Rogers, Rogers' Law and Practice of Elections and Registration 368 (8th ed. 1857) (likewise concluding that the statute covered "dismissal of a person employed," a "notice to quit given to a tenant," or "withdrawal of custom from a tradesman" based on the targets' votes); 1 Reports of the Decisions of Committees of the House of Commons in the Trial of Controverted Elections, During the Seventeenth Parliament of the United Kingdom 90–91 (F.S.P. Wolferstan & Edward L'Estrange Dew eds., London V & R Stevens & G.S. Norton 1859) (reporting that a vote was disallowed on the grounds of "undue influence" because the voter was pressured by threat of loss of employment).

[6]. See Message from His Excellency, Isaac Toucey to the Legislature of Connecticut (May 1846), in Journal of the House of Representatives of the State of Connecticut, May 1846, at 25–26 (New Haven, Osborn & Baldwin 1846) (justifying the proposed Connecticut law banning threats of retaliation by employers on the grounds that such threats are "a compound of bribery, undue influence and intimidation").

[7]. See An Act to Regulate Elections in this State, § 89, 1868 Ala. Acts 286 (prohibiting an employer from "disturb[ing] or hinder[ing]" an employee exercising the right of suffrage); An Act Extending Protection to Laborers in the Exercise of Their Privilege of Free Suffrage, 1868 La. Acts 64 (making it a crime for employers to discharge their employees because of their political opinions, or to attempt to control the way they vote); Intimidation of Voters, N.C. Code § 2715 (1883) (enacted 1868), reprinted in 2 William T. Dortch et Al., The Code of North Carolina 195 (New York, Banks & Bros. 1883) (prohibiting employers from threatening their employees on account of their votes); An Act Providing for the Next General Election and the Manner of Conducting the Same, §11, 1868 S.C. Acts 135, 137 (special session) (same); An Act to Regulate the Conduct and to Maintain the Freedom and Purity of Elections, § 67, 1870 La. Acts 158 (same); An Act to Provide for the Mode and Manner of Conducting Elections, § 46, 1870 Tex. Gen. Laws 137 (same). All these expressly barred "threats of discharge from employment" aimed at influencing a person's vote; all except Alabama also banned discharge based on past votes. Mississippi already had a law banning threats of discharge from employment for votes, Of Bribery and Undue Influence, § 4, 1839 Miss. Laws 152; a proposal to specify in the state constitution that dismissal from employment based on one's past or future vote shall be a crime, Journal of the Proceedings of the Constitutional Convention of the State of Mississippi 352 (Jackson, E. Stafford 1868), apparently wasn't enacted.

NEXT: Federalism and Law and Order

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. “About half of Americans live in jurisdictions …” is an interesting measure that says something about the breadth of a particular rule. However, I’m not sure it tells the whole story. Do “half of Americans” live under that rule because it’s a common rule or because it’s an extremely uncommon one that happens to be followed in a few high-density jurisdictions?

    Prof Volokh, do you have easy access to the data that could answer that question?

    1. Second this. I could see a nice ALR emerging from such an analysis.

    2. The article lists the relevant jurisdictions in the Table of Contents (though Utah has enacted such a law since the article was published); by my count, 23 states offer some level of protection here, though it varies from quite narrow to quite broad.

  2. Legalities aside, efforts to control speech unrelated to actual employment duties are counter-productive. It is a bad use of line management and HR time AND it dilutes control over workplace speech.

    I personally know executives who waste far too much time patrolling employee social media accounts. Aside from the waste of time doing so creates an impediment to the development of a quality culture.

    HOWEVER, as the professor correctly points out (presumably in response to complaints):

    “When an employee’s speech or political activity sufficiently alienates coworkers, customers, or political figures, an employer may reasonably claim a right to sever his connection to the employee.”

    (See my note below on progressive discipline) The bottom line, however, is that I am not going to lose revenues or productivity because some idiot is incapable of reasonable self-censorship.

    Much of my training is from the late W. Edwards Deming. Deming methodology requires VERY strong control over how employees relate to customers; what they say and how they say it. Attempting to control speech outside of customer (and manager-employee) relations creates a loss of focus. The exception is outlined above.

    While not the core topic of this post, just a note on the at-will doctrine. I agree with the professor. I have never, in my history, extended a contract to any mid-level manager or below. However, employment recruiting is a competitive enterprise and a commitment to at-will cedes some good people to one’s competition.

    In the alternative to at-will, extending a contract to an executive includes benefits to the employer including a NDA when appropriate. Moreover, there is a productivity benefit when executives feel more secure.

    I have been on the “evil” side of some NLRB elections (I used to be pretty good at it). Obviously the presence of a bargaining unit effectively eliminates many of the employer benefits of at-will employment.

    The key to maintaining that environment is neither wages nor benefits (IMO). Rather it is a written policy of progressive discipline: Warning -> unpaid leave -> termination.

    Alas, I have become a SJW. My old self would be pissed at my current self. It effects an interesting form of neurosis.

    1. >“When an employee’s speech or political activity sufficiently alienates coworkers, customers, or political figures, an employer may reasonably claim a right to sever his connection to the employee.”
      At bottom, this qualifier is a red herring. In a healthy company, this situation should never arise out of the blue. When it does, it represents a failure in the hiring process. If all employees are screened such that they (1) align with company values in the workplace, (2) fit in decently well with current company culture including employees and customers, and (3) are respectful, then such an outcome cannot occur. In a healthy hiring process, management and future coworkers have a say. Future coworkers won’t recommend hiring someone who will alienate them. And managers won’t hire someone who alienates customers. Therefore, this issue should be exceedingly rare in healthy companies. If an online mob suddenly demands an employee be fired for private speech, those are likely not the firm’s customers, and bowing to their demands is not covered by the Professor’s qualifier. Because if the mob were a significant portion of the firm’s customers, management would have never hired the employee in question after screening him appropriately.

      1. I can’t say I’ve been involved in a hiring process to know whether it truly and necessarily is involved enough to catch all instances. So I’m not going to dispute it, though I will admit I am doubtful. And in no place I’ve interviewed at has anyone but the manager and the person responsible for interviewing (if different) been involved. Definitely none of the “future coworkers”

        However, anti-discrimination rules apply to hiring just as much as firing. So your claim that it would show up in the hiring process isn’t much help. If a state gives speech the same anti-discrimination protection it does to race or sex, for example, then not hiring someone because of their speech is just as much a violation. In fact many (most?) anti-discrimination lawsuits claim it happened in a decision not to hire.

      2. That assumes culture never changes. Having watched companies following a change in leadership (new CEO, merger, etc), such a static view of the company culture is unrealistic. It is quite possible to be a good fit to the company’s culture today but a quite poor fit a decade later. Culture change, by the way, is not necessarily a sign of an unhealthy company. It is merely change – could be to the good, could be to the bad, could be merely neutral – yet change it is.

        I’d also note that your analysis assumes that culture fit is the dominant consideration during hiring. While culture fit is important, so are critical skills and expertise. Sometimes you hire a moderate fit because you think the benefits of their expertise outweigh the detriments of culture. Often, that works – and sometimes it doesn’t. The times it doesn’t work out are not automatic evidence of a hiring failure.

      3. II am a well trained interviewer using situational techniques. Nevertheless, I have had to terminate people because you really do not know what you are getting until they are on the job.

        Furthermore, people change. They are often influenced by others and those “others” change.

        Companies also change. The quality cycle DEMANDS change and some people simply refuse to adapt and there is no interviewing process that can identify all of the future rebels.

        It used to be a great deal easier. These days everyone is afraid of getting sued and that produces neutral references.

        Companies that I have led have been sued. The plaintiff and a contingency fee lawyer know that many frivolous employment suits are settled because that is less expensive than winning the suit in court.

        One of these days I could write a book about adventures with the NYC Labor Board.

  3. “When an employee’s speech or political activity sufficiently alienates coworkers, customers, or political figures, an employer may reasonably claim a right to sever his connection to the employee.”

    This *can’t* be right. Part of the point of the original nondiscrimination laws was that employers shouldn’t be able to just say, “but my customers/employees/etc. WANT me to discriminate, so I really really need to…” And of course, once it’s the law in some reasonably uniform way throughout most of society, any such pressure loses its impact because 1) a business can just shrug and say it’s the law, and 2) competitor businesses are in the same boat, so there isn’t much threat-power.

    Note that saying that is not getting into whether this is a similar sort of civil rights issue, nor where/how to draw the line if it is. But *if* it is, then to that extent, the same reasoning should apply.

Please to post comments