NLRB finds Internet/blogging policy violated NLRA, as did Facebook-related firings

A restaurant/bar unlawfully discharged employees caught venting on Facebook because they had to pay extra state income taxes due to the employer’s withholding mistakes, a three-member NLRB panel held. The Board rejected the notion that the Facebook conversation, in which the manager who performs the restaurant’s accounting duties was called “an asshole,” was so disparaging or defamatory that the conduct lost the protection of the Act. (In so ruling, it also rejected the use of the Atlantic Steel framework, generally speaking, in cases involving the off-site use of social media.) Among other additional findings, a majority found that the restaurant’s internet/blogging policy was unlawful; Member Miscimarra dissented on this point (Three D, LLC dba Triple Play Sports Bar and Grille, August 22, 2014).

Facebook griping. It all started when several restaurant employees (not represented by a labor union, in this case) discovered that they owed more in state income taxes than they anticipated. The employees discussed the problem with coworkers at the restaurant, and some complained about it to the employer, who planned a staff meeting with its payroll provider to discuss the employees’ concerns. Meanwhile, a former employee took to Facebook to air her grievance: “Maybe someone should do the owners of Triple Play a favor and buy it from them. They can’t even do the tax paperwork correctly!!! Now I OWE money…Wtf!!!!” She added, in another comment: “It’s all Ralph’s fault. He didn’t do the paperwork right. I’m calling the labor board to look into it bc he still owes me about 2000 in paychecks.” At this point, one of the restaurant’s current employees “liked” his former colleague’s initial status update.

Then another employee chimed in: “I owe too. [He’s] Such an asshole.” The employee’s Facebook privacy settings permit only her Facebook friends to view her posts. But the restaurant owner’s sister, also an employee, was a Facebook-friend of-a-friend, and she told her brother about the social media conversation. She was fired when she came into work. When asked why, the boss said she was “not loyal enough” to work for the restaurant given her Facebook comment.

The employee who hit “like” in response to the initial comment met a similar fate. He was brought into the office, asked whether he “had a problem with them, or the company,” was interrogated about the Facebook discussion, the identity of other people who took part in the conversation, whether he had written anything negative about management, and the meaning of his “like.” He was also told that because he “’liked’ the disparaging and defamatory comments,” he obviously wanted to work elsewhere. After showing the employee the door, the restaurant owner told him that his attorney instructed the restaurant to fire anyone involved in the Facebook conversation — and that “you’ll be hearing from our lawyers.” He never did hear from the restaurant’s attorneys, although they contacted the other discharged employee raising the threat of legal action against her; they also contacted the former employee, who deleted the entire conversation and posted a retraction.

Unfair labor practice findings. The law judge found the Facebook discussion was concerted activity under the standard set forth in Meyers Industries because it involved four current employees and was “part of an ongoing sequence” of discussions that began in the workplace about the restaurant’s calculation of employees’ tax withholding. She also found the employee who hit the “like” button was expressing his support for the others, and thus was a participant in the protected, concerted activity. Also, the activity did not lose the protection of the Act under Atlantic Steel Co or Jefferson Standard, she concluded, so their discharge on this basis was unlawful. The Board panel agreed, but under a slightly different analysis.

The Board also held the restaurant separately violated the NLRA by threatening the employees with discharge, interrogating them about their Facebook activity, informing them that they were being discharged because of their Facebook posts, and threatening them with legal action.

Atlantic Steel does not apply. Unlike the law judge, the Board did not apply the Atlantic Steel framework in deciding whether the Facebook comments at issue here lost the protection of the Act, explaining that this precedent was “not well suited to address issues that arise in cases like this one involving employees’ off-duty, offsite use of social media to communicate with other employees or with third parties.” Atlantic Steel factors work well when analyzing whether direct, face-to-face workplace confrontations between employees and management are “so opprobrious” as to lose the Act’s protection, the Board noted, but it hasn’t typically been applied as to third-party or public communications. (Atlantic Steel’s “place of discussion” factor, in particular, is problematic to apply outside the physical workplace.) In those cases, the Board has applied the standards set forth in Jefferson Standard and Linn. Doing the same here, the Board found the comments at issue were statutorily protected.

The Board was careful to note, though, that it did not hold “employees’ off-duty, offsite use of social media can never implicate an employer’s interest in maintaining workplace discipline and order in the same manner that a face-to- face workplace confrontation with a manager or supervisor does.” However, the facts here did not support such a finding. Specifically, the employee’s use of “asshole” to describe the manager, during the course of a protected discussion on social media, didn’t sufficiently implicate the restaurant’s interest in maintaining discipline in the workplace to the extent that an Atlantic Steel analysis was called for.

Comments not unduly disparaging. It was undisputed the Facebook conversation was a concerted activity under the NLRA; the question was whether the conversation lost the protection of the Act. As the restaurant saw it, the Facebook posts were made in a “public” forum accessible to other workers and to customers — adversely affecting management’s authority in the workplace as well as its public image. However, the Board disagreed that the communication here was “so disloyal, reckless, or maliciously untrue” as to be unprotected under Jefferson Standard and its progeny. The comments made no mention of the restaurant’s products or services, let alone disparage them, the Board observed. “Where, as here, the purpose of employee communications is to seek and provide mutual support looking toward group action to encourage the employer to address problems in terms or conditions of employment, not to disparage its product or services or undermine its reputation, the communications are protected.”

Distinguishing the comments at issue with those deemed unprotected in Jefferson Standard, the Board said the Facebook discussion here clearly revealed the existence of an ongoing dispute (over the restaurant’s tax withholding practices). Also, the discussion was not directed to the general public; the comments were posted on an individual’s personal Facebook page, not on a company page providing information about its products or services. Although the record was unclear as to the privacy settings used by the former employee (whose initial status update launched the whole affair) or other “friends” who made comments, “we find that such discussions are clearly more comparable to a conversation that could potentially be overheard by a patron or other third party than the communications at issue in Jefferson Standard, which were clearly directed at the public.”

Grappling with Facebook-specific discourse. What did the employee’s Facebook “like” mean in this instance? While it was ambiguous, the Board concluded the employee was merely endorsing the original status update by the former employee, not the “entire topic as it existed at the time,” as the law judge had found (including the subsequent contention that it was “all Ralph’s fault.”) Had the employee intended to express his approval or agreement with the additional comments that followed the initial status update, he would have “liked” them each individually, the Board reasoned.

The Board also rejected the employer’s effort to pin the blame on the two employees for other comments made during the ongoing Facebook conversation, including comments by the former employee accusing the restaurant of pocketing employees’ money. Assuming that accusation would have been unprotected, it didn’t follow that the employees would have lost the protection of the Act by simply participating in an otherwise-protected Facebook discussion where unprotected statements were made by other participants.

Comments not defamatory. Nor were the employees’ Facebook comments defamatory under the standard set forth in Linn and its progeny. The Board saw no basis to find the employees’ claims that their withholding was insufficient to cover their state tax liability, or that the shortfall resulted from an error on the employer’s part, were maliciously untrue. As for the “asshole” comment about the restaurant manager, made in connection with the complaints about the withholding errors, the Board said the employee was merely voicing a negative personal opinion that could not reasonably be read as a statement of fact. This, too, did not lose the Act’s protection.

Discharges unlawful. Because it was undisputed that the employees’ discharges were motivated by the Facebook comments, it also followed that the finding that such comments were protected meant the discharges violated the Act. Of note here: the law judge applied Burnup & Sims in finding the discharges violated Sec. 8(a)(1), notwithstanding that the restaurant may have mistakenly believed, in good faith, that the Facebook posts were unprotected. But Burnup & Sims applies in cases involving mistakes of fact, not mistakes of law. This was clearly not a “mistake of fact” case, so this precedent did not apply. “Otherwise,” as the Supreme Court explained, “the protected activity would lose some of its immunity, since the example of employees who are discharged on false charges would or might have a deterrent effect on other employees.”

Internet policy unlawful. Here’s where the Board parted ways with the law judge, and Member Miscimarra, with the majority. The judge had dismissed the allegation that the restaurant unlawfully maintained an overly broad internet/blogging policy, but the Board majority reversed, finding that employees would reasonably construe the restaurant’s policy to prohibit the type of protected Facebook posts that led to the unlawful discharges. Specifically, employees could reasonably interpret the policy barring “inappropriate discussions” as prohibiting any discussions about terms and conditions of employment deemed “inappropriate” by the employer. Problematic to the majority was that the rule contained only one other prohibition — against revealing confidential information — and offered no illustrative examples as to what would be deemed inappropriate in the employer’s eyes. As such, the term “inappropriate” was “sufficiently imprecise” to implicate Section 7 concerns. (On this holding, the majority was guided by its precedent in First Transit, Inc, noting that the Board’s approach in this area “has received judicial approval.”)

(Miscimarra disagreed with the notion that the term “inappropriate” was too imprecise to pass muster without also providing illustrative examples. As he observed, most individuals would appreciate that “inappropriate” behavior could have consequences sufficiently serious as to violate the law and result in discipline. “It does not per se violate Federal labor law to use a general phrase to describe the type of conduct that may do so.” If such were the case, he reasoned, then “just cause” provisions found in most bargaining agreements over the last eight decades would be deemed invalid, he argued.)

And while the General Counsel did not contend that the restaurant expressly relied on its internet policy in discharging the employees in this case, by firing them unlawfully for their Facebook discussion, the restaurant conveyed a strong message to its workforce about “the scope of its prohibition against inappropriate discussions and that they should construe its rule against inappropriate discussions to include such protected activity.” True, the policy contained a general savings clause noting that it had no force or effect if state or federal law precluded it, but the unlawful discharges sent a contrary message.

Miscimarra rejects Lutheran Heritage. Finding nothing in the language of the internet policy that employees could reasonably construe to prohibit Section 7 activity, Miscimarra would find the policy was “legitimately aimed to prevent the revelation of proprietary information and statements about the company, its management, and its employees that may be unlawful.” Miscimarra noted too that he disfavored the current Board standard regarding overly broad work rules and policies, set forth as the first prong of Lutheran Heritage (work rules are unlawful even where they do not explicitly restrict protected activity, and are not applied against or promulgated in response to such activity, if “employees would reasonably construe the language to prohibit Section 7 activity”). He urged the Board to reconsider the standard when an appropriate case permits (but, even under Lutheran Heritage, he would find the policy here lawful — phrased “in general commonsense terms that preclude it from reasonably being considered unlawful under any standard.”)

Miscimarra also decried the majority’s “cobbling together” here of prongs one and three of the Lutheran Heritage standard as “contrary to the careful separation of those two theories of violation established in that case.” Under prong one, the question is whether the language of a work rule would, on its fact, reasonably be interpreted to prohibit Section 7 activity; prong three asks whether a rule, regardless of its wording, has been applied to restrict Section 7 rights. “The majority continues down the path of this hybrid category of violation, under which a rule that is not unlawful on its face and has not been applied to restrict the exercise of Section 7 rights nevertheless is found unlawful based on a mixture of the rule’s language and the employer’s conduct,” he asserted. “In so doing, the majority contributes to the uncertainty employers confront in seeking to square their rules with our Lutheran Heritage prong-one precedent, which, at this point, consists of so many distinctions, qualifications, and factual variations as to preclude any reasonable ‘certainty beforehand’ for most parties ‘as to when [they] may proceed to reach decisions without fear of later evaluations labeling [their] conduct an unfair labor practice.’”

The majority rejected the “cobbling together” accusation, and the notion that its holding would spark greater uncertainty on the part of employers by applying the first prong of Lutheran Heritage to the facts at hand. Rather, its finding that the policy in question here was unlawful was in keeping with “the many Board decisions that have found a rule unlawful if employees would reasonably interpret it to prohibit protected activities,” the majority countered.

Threatening litigation: an unresolved issue. Also worth noting: In adopting the law judge’s finding that the restaurant also violated the Act by unlawfully threatening legal action against the employees, the Board relied on the restaurant’s post-discharge statement to one employee that he would “be hearing from [its] lawyers.” This threat was not incidental to a lawsuit; the lawyers never contacted him and the restaurant took no legal action against him. By its threat alone, though, the restaurant violated Sec. 8(a)(1), regardless of whether a lawsuit against the employee would have been unlawful had one been filed.

On this point, the law judge erred in stating that the NLRB had “explicitly declined to apply” the principles of BE & K Construction Co. (2007) to threats to initiate litigation where the threat is incidental to the actual filing of a lawsuit itself. That issue remains undecided, the Board noted, adding that its resolution was unnecessary here.
By Lisa Milam-Perez, J.D.