On Social Media Issues in the Workplace

Home , Like button

“Chapter 2” on Social Media Issues in the Workplace

Friday, May 9, 2014 General Session; 10:30 a.m. – 12:00 p.m.

Traci I. Park, Burke, Williams & Sorensen

League of California Cities® 2014 Spring Conference Renaissance Esmeralda, Indian Wells

Notes:______League of California Cities® 2014 Spring Conference Renaissance Esmeralda, Indian Wells

Status Update: Developments On Social Media In The Workplace

League of California Cities City Attorneys Conference

May 2014

Traci I. Park Partner Burke, Williams & Sorensen, LLP 444 S. Flower Street, 24th Floor Los Angeles, CA 90017 t: 213.236.2833 / e: [email protected] www.bwslaw.com

I. SOCIAL MEDIA & PRIVACY

The advent of social media has taken public things that once were considered private and actively protected as such. Today, however, attitudes have changed and users of social media regularly post personal information such as their names, email addresses, employer, schools attended, photos, events they have attended or plan to attend, home addresses, telephone numbers and birthdates. Some websites call for interests, hobbies, favorite books or films, major life events, and even relationship status.

If you use Facebook, perhaps you’ve noticed that when you post a photo, Facebook recognizes who else is in the photo so you can “tag” them. This is due to a recent feature on Facebook, Tag Suggest, which scans your photos using facial recognition technology. This system detects human faces in photos and then calculates a unique numerical identifier for each face based on characteristics such as the shape of the eyes and the distances between eyes, nose, and ears. It then ties that face to a specific user’s name. Tag Suggest uses this system to search photos you upload of your friends. If it finds one, it suggests that you “tag” the photo with the friend’s name, and the photo will appear on their page.

Perhaps you’ve also noticed ads appearing along the margins of your Facebook page for items you may have shopped for on line or restaurants you’ve evaluated on Yelp. How does Facebook know this information? Because Facebook tracks your internet activity by dropping “cookies” into your device that identifies behavior, habits, preferences, other websites you’ve accessed, geographical places you’ve visited, items you’ve browsed for or purchased, or searches that you’ve run in Google. Advertisers pay Facebook for this information, and voila, targeted advertising appears on your page. For their part, when an advertiser creates an ad on Facebook, they can select specific parameters to reach the right people. A simple example of a parameter could be “Someone engaged to be married, who lives in San Francisco, between the ages of 20-30.” But with the level of detail users of Facebook share, advertisers can narrow it down even more: “Someone engaged to be married, who lives in San Francisco, between the ages of 20-30, who likes swimming and drives a BMW.”

Many social media sites, such as Facebook, offer different levels of privacy settings, with the default being that all content is publically accessible. There are settings and applications that can be used to stop (or at least reduce) on-line tracking of your activity. However, a recent Consumer Reports survey revealed

that nearly 13 million U.S. Facebook users are not aware of the privacy control settings and about 30% of users do not bother to use them.1

Even when privacy settings are in place, there is never a guarantee that anything someone posts online will be private or remain with an intended limited audience. Indeed, a theme being repeated in appellate decisions across the country is that users have no expectation of privacy in any social media website. To the contrary, many courts routinely conclude that using social media is the opposite of expecting privacy. According to one California court, no reasonable person who takes the affirmative act of posting information on a social media website has an expectation of privacy. Moreno v. Hanford Stentinel, Inc., (2009) 172 Cal. App. 4th 1125, 1130. Or, as one New York court put it, “[W]hen Plaintiff created her Facebook and MySpace accounts, she consented to the fact that her personal information would be shared with others, notwithstanding her privacy settings. Indeed, that is the very nature and purpose of these social networking sites else they would cease to exist. Since Plaintiff knew that her information may become publically available, she cannot now claim that she had a reasonable expectation of privacy.” Romano v. Steelcase, Inc., (2010) 907 N.Y.S. 2d 650, 656.

This fundamental lack of privacy in on-line activity raises numerous questions that continue to plague employers. Can employees be disciplined for things they say on line? Can employers base a decision not to hire an applicant based on information learned from social media sites? Do litigants have a right to learn what their opponent says on line? Are there special concerns for use of social media by law enforcement officers or public officials? Although answers to some of these questions are starting to emerge, the impact of social media in the workplace remains a rapidly developing legal area.

II. SOCIAL MEDIA, FREE SPEECH & EMPLOYEE DISCIPLINE

A recurring question facing public employers is whether and to what extent employees can be disciplined for their on-line activities and speech. A recent poll by the Huffington Post revealed that nearly half of Americans think the First Amendment protects them from being fired for what they say, and more than 3 in 10 think it applies to situations like A&E’s now-revoked suspension of “Duck Dynasty” star, Phil Robertson.2 They’re wrong, of course. The First Amendment

1 Facebook & Your Privacy, Consumer Reports Magazine, June 2012. 2 https://today.yougov.com/news/2014/01/13/poll-results-first-amendments/. Interestingly, this poll also revealed that 45 percent of Americans said the First Amendment does not allow people to be fired from a job for expressing their views,

won’t save an employee from his or her own bad judgment on-line, and it won’t necessarily offer job security when an employer doesn’t like what an employee posts.

Although the First Amendment affords greater free speech protections for government employees relative to those in the private sector, speakers do not have carte blanche to insult or deride whomever they see fit. Rather, courts will balance the interest a governmental entity may have in regulating speech versus the constitutional rights of the speaker.

One of the first major employee discipline cases involving questionable judgment and free speech was City of San Diego v. Roe, (2004) 125 S. Ct. 521. There, a San Diego police officer, Roe, made sexually explicit videos himself in his police uniform and sold them on Ebay. He also sold police uniforms and equipment and his profile identified him as a law enforcement officer. Roe's supervisor, a police sergeant, discovered Roe's activities when, while on eBay, he came across an official SDPD police uniform for sale offered by an individual with the username “[email protected].” He searched for other items Code3stud offered and discovered listings for Roe’s sexually explicit videos. Recognizing Roe’s picture, the sergeant printed images of certain of Roe’s offerings and shared them with others in Roe’s chain of command. A subsequent investigation by the SDPD revealed that Roe violated numerous policies, including conduct prejudicial, outside employment, and immoral conduct. Roe was ordered to cease selling sexually explicit materials or engaging in similar conduct on the internet. After Roe failed to comply, his employment was terminated. Noting that governmental employers may impose certain restrictions on the speech of its employees that would be unconstitutional if applied to the general public, the Supreme Court also recognized that when public employees speak or write on their own time on topics unrelated to their employment, the speech can have First Amendment protection, absent a strong government justification in regulating the speech. Id. at 523-524. The Court easily found that Roe’s on-line expression did not qualify as a matter of public concern, as his sex videos did not inform the public about any aspect of the SDPD’s functioning or operations. Id. at 524-525. The Court also found that Roe took deliberate steps to link his videos to his police work, in a way that was injurious to the Department and reflected negatively on the professionalism of its officers. Accordingly, the Court

while only 36 percent said such firings are allowed under the Constitution. Twenty percent said they weren't sure.

concluded that Roe’s expression was not protected by the First Amendment, and his employer could take disciplinary action against him.

In a more recent case, Shepherd v. McGee, 2013 WL 5963076 (D. Ore. 2013), a social worker employed by the Oregon Department of Human Services (“DHS”) was terminated after she posted comments on social media critical of people on public assistance. Shepherd filed a §1983 First Amendment retaliation lawsuit alleging that her speech was constitutionally protected.

Some of Shepherd’s primary functions were to prepare juvenile court cases, make recommendations for juvenile court disposition and to appear in Court. She acknowledged in pretrial proceedings that her job was “to be a neutral appraiser of the settings in which the children live.” In working on dependency cases, she was not supposed to consider the employment status, religious beliefs, or political beliefs of the adults in the home, or concern herself with how they chose to spend money or furnish their home. Many of the families she assisted were on some form of public assistance.

While working for DHS, Shepherd maintained a Facebook page on which she identified herself as a “Child Protective Services Case Worker at Department of Human Services.” Her Facebook page contained no general disclaimer that any content on the page was her opinion and not that of DHS and she never added such a disclaimer to any particular post. She had hundreds of Facebook friends who had access to all of the content posted on her Facebook page. Her Facebook friends included a Polk County Circuit Court Judge, at least three deputy district attorneys, several defense attorneys, and over a dozen law enforcement officers.

During her employment, Shepherd posted the following comment to her Facebook page: “So today I noticed a Self–Sufficiency client getting into a newer BMW. What am I doing wrong here? I think I need to quit my job and get on TANF.”3 Several of her Facebook friends made comments in response to the post and she herself added the following two comments: (1) “Almost every client home I go into has a gigantic flat screen tv. I ask how they paid for it, and it's usually with their tax returns ....yet they don’t pay taxes!”; and (2) “You should let me know when you send people to jail and I can get their ‘benefits’ turned off.”

In another post, Shepherd stated:

3 TANF refers to “Temporary Assistance to Needy Families,” a public welfare program.

I was listening to the radio and they were making up rules for society. Here are my rules: (1) If you are on public assistance, you may not have additional children and must be on reliable birth control (e.g. an IUD), (2) If you've had your parental rights terminated by DHS, you may not have more children ....it's sterilization for you buddy! (3) If you are on public assistance and don't pay taxes, you shouldn't get taxes back from that child tax credit[,] (4) If you are on public assistance, you may not own a big flat screen television, (5) If you receive food stamps, you should be limited on what you may purchase (no more ribeye steaks, candy, soda, chips, etc), (6) If you physically abuse your child, someone should physically abuse you, (7) I should be president so I can make up more rules, (8) If you don't like my rules, too bad. I have a Ph.D. and you don't so I get to make up my own imaginary rules.

Shepherd also added comments in response to comments posted by her Facebook friends including “Oh, and you can add to my rules, but if your rule sucks then I reserve the right to veto it” and “Working at DHS broke me of my liberal ideals from college. Go figure.”

Several of Shepherd’s co-workers saw the posts, which they turned over to the Human Resources Manager, who did not find the comments funny or ironic. He believed that the “rules for society” post showed bias on Shepherd’s part that she was supposed to be putting aside. During an investigation, Shepherd acknowledged that she in fact held some of the beliefs she had written in the Facebook posts and that there was no way of telling from the Facebook posts what she believed and what she did not. She also admitted that she would have a hard time testifying and explaining her objectivity. DHS consulted with the District Attorney, who believed the Facebook posts were subject to discovery by opposing counsel because they were broad statements, not defined to a specific case, and would need to be disclosed in every dependency hearing involving physical abuse, and that it would likely hamper current and future cases. In light of this, DHS terminated Shepherd’s employment.

In analyzing Shepherd’s First Amendment retaliation claim, the court focused on whether DHS had adequate justification for treating her differently from other members of the general public.4 This analysis focuses on whether the

4 In the Ninth Circuit, First amendment retaliation claims are analyzed using a five-factor test first espoused by the Supreme Court in Pickering: (1) whether the plaintiff spoke on a matter of public concern; (2) whether the plaintiff spoke as a private citizen or public employee; (3) whether the plaintiff's protected speech was a substantial or motivating

government's legitimate administrative interests outweigh the employee's First Amendment rights. The question is “whether the relevant government entity had an adequate justification for treating the employee differently from any other member of the general public.” Garcetti v. Ceballos, (2006) 547 U.S. 410. 418. Legitimate governmental interests in the First Amendment retaliation balancing inquiry include interests in “promoting efficiency and integrity in the discharge of official duties and maintaining proper discipline in the public service.” Clairmont v. Sound Mental Health, (9th Cir. 2011) 632 F.3d 1091, 1102-1103. Additionally, the government may show that an employee's statement “impairs discipline by superiors or harmony among co-workers, has a detrimental impact on close working relationships for which personal loyalty and confidence are necessary, or impedes the performance of the speaker's duties or interferes with the regular operation of the enterprise.” Rankin v. McPherson, (1987) 483 U.S. 378, 388.

Examining Shepherd’s duties to interact with families on public assistance and the need for her to testify in court about her cases, the Court agreed with DHS that Shepherd’s Facebook posts irreparably impaired her ability to perform her duties, as her credibility as a witness was severely compromised, would hamper future cases, and reflected adversely on DHS in the relevant local community. Thus, DHS’s interest in maintaining efficient operations outweighed Shepherd’s free speech rights.

Notably, Shepherd argued that her Facebook posts caused no actual, material and substantial disruption, or reasonable predictions of disruption in the workplace. The Court found this argument unavailing because the evidence demonstrated that her Facebook posts caused her supervisors to doubt her ability to be effective. That doubt alone was a disruption to her working relationships, whether or not she was ever actually impeached prior to her termination. According to the Court, “The government employer does not have to compromise its function by allowing the employee to actually cause disruption or fail to perform his or her job duties in order to establish an impairment in efficient operations.” Shepherd at *6.

factor in the adverse employment action; (4) whether the state had an adequate justification for treating the employee differently from other members of the general public; and (5) whether the state would have taken the adverse employment action even absent the protected speech. Eng v. Cooley, 552 F.3d 1062, 1070 (9th Cir. 2009); Pickering v. Board of Education, (1968) 391 U.S. 563, 568. The focus of the Shepherd opinion was almost exclusively on the fourth factor. The court did not need to reach the issues of whether the employee spoke as a private citizen or whether the posts were speech on a matter of public concern because DHS demonstrated as a matter of law that it had adequate justification for the termination.

The Shepherd court also held that in balancing the interests of the government employer and the public employee, the employer must show greater disruption to its provision of services when the speech is at the core of First Amendment protection or is directed at a large audience; correspondingly, the employer is given wider deference when the speech is at the edge of First Amendment protection, a smaller audience is targeted, or a close working relationship is at stake. See Brewster v. Board of Education of Lynwood Unified Sch. Dist., (9th Cir. 1998) 149 F.3d 971, 981 (in weighing interests, it is significant whether the speech at issue was made to the public or the media or instead to a more private audience); and Johnson v. Multnomah County, (9th Cir. 1995) 48 F.3d 420, 426 (“The more tightly the First Amendment embraces the speech the more vigorous a showing of disruption must be made.”)

Here, Shepherd’s Facebook settings allowed only her designated Facebook friends to view her Facebook posts. And, while the subjects of tax credits or the use of government resources or entitlement programs are arguably matters of public concern, the Court did not feel that Shepherd’s comments struck at the heart or core of the First Amendment, given her own characterization of them as “humorous and ironic” and “jokes,” as she had alleged in her lawsuit. As such, the Court concluded, “they are more on the periphery of First Amendment protection because they were banter rather than speech intended to help the public actually evaluate the performance of a public agency.” Shepherd at *9. According to the Court, “[A]ssuming for the purposes of this Opinion that the posts are entitled to First Amendment protection, they were not distributed to the public or the media but were targeting a smaller audience and they are not at the core of First Amendment protection. As a result, the government has more deference in determining whether the speech at issue has interfered with its operations or with the employee's ability to effectively perform his or her duties. Even if Plaintiff had widely distributed her comments to the public and even if they struck at the core of First Amendment protection, the balance would still tip in DHS's favor given the record of potential disruption. DHS has established as a matter of law that it had an adequate justification for treating Plaintiff differently than any other member of the general public.” Id.

In another recent First Amendment retaliation case involving a police officer who criticized a colleague online, the Eleventh Circuit determined that Facebook venting was not protected by the First Amendment. Gresham v. City of Atlanta, (11th Cir. 2013) 542 Fed. Appx. 817.5 Gresham, a police officer, posted the following to her Facebook newsfeed:

5 Although the District Court’s opinion was published and is citable, the Eleventh Circuit opinion has not yet been selected for publication.

“Who would like to hear the story of how I arrested a forgery perp at Best Buy only to find out later at the precinct that he was the nephew of an Atlanta Police Investigator who stuck her ass in my case and obstructed it?? Not to mention the fact that while he was in my custody, she took him into several other rooms alone before I knew they were related. Who thinks this is unethical?”

Gresham’s Facebook page was set to “private,” but her friends could view her posts and could potentially distribute the comment more broadly. Not surprisingly, several co-workers with whom Gresham was friends on Facebook shared the comments with the Department’s Office of Professional Standards, which then launched an investigation.

The post allegedly violated the police department’s official “Criticism” policy, which provided, “Employees will not publicly criticize any employee or any order, action, or policy of the Department except as officially required. Criticism, when required, will be directed only through official Department channels, to correct any deficiency, and will not be used to the disadvantage of the reputation or operation of the Department or any employees.”

Relying on the same multi-factored Pickering test employed by the Court in the Shepherd case, the Eleventh Circuit assumed, arguendo, that the speech involved a matter of public concern, and focused instead on whether Gresham’s interests in speaking outweighed her employer’s legitimate interest in efficient public service. Citing to Rankin, supra, the Court noted, “In conducting this balancing process, the law is well established that maintaining discipline and good working relationships amongst employees is a legitimate governmental interest.” The Court also noted that it is well established that a police department is a quasi–military organization and that “comments concerning co-workers' performance of their duties and superior officers' integrity can directly interfere with the confidentiality, espirit de corps and efficient operation” of the department.” Gresham at 818-819 [internal citations omitted].

Like Shepherd, Gresham argued there was no evidence of actual disruption in the Department due to her Facebook posts. Again, the Court found this lackluster argument unavailing: “The case law is clear that the government's legitimate interest in avoiding disruption does not require proof of actual disruption…When close working relationships are essential to fulfilling public responsibilities, a wide degree of deference to the employer's judgment is appropriate. Furthermore, we do not see the necessity for an employer to allow events to unfold to the extent that the disruption of the office and the destruction

of working relationships is manifest before taking action. Rather, we have held that “the reasonable possibility of adverse harm will generally be enough to invoke the full force of judicial solicitude for a police department's internal morale and discipline.” Gresham at 819 [internal citations omitted]. Here, Gresham had violated a department policy, which led to an investigation, and the Court noted that “common experience teaches that public accusations of unethical conduct against fellow officers would have a natural tendency to endanger the espirit de corps and good working relationships amongst the officers. Thus, we conclude that there is actual evidence in this record of a reasonable possibility of disruption of the legitimate interests of the Department.” Id.

The Eleventh Circuit ultimately agreed that these legitimate interests of the Department outweighed Gresham’s interest in speaking in this matter. “…[W]e note that the context of Plaintiff's speech is not one calculated to bring an issue of public concern to the attention of persons with authority to make corrections, nor was its context one of bringing the matter to the attention of the public to prompt public discussion to generate pressure for such changes. Rather, we agree with the district court that the context was more nearly one of Plaintiff's venting her frustration with her superiors. Thus, we conclude that Plaintiff's speech interest is not a strong one, a factor which the Supreme Court has indicated is appropriate to consider in the balancing process. Moreover, even if Plaintiff's speech interests were somewhat stronger, we conclude that the Department's interest is considerable, and that the balance clearly tilts in favor of the Department.” Gresham at 819-820 [internal citations omitted].

In another very recent First Amendment retaliation case involving a former Deputy Chief of Police who was demoted after posting an image of the Confederate flag accompanied by the comment, “Its time for the second revolution” on Facebook, a federal court in Georgia determined that the employer’s interest in promoting the efficiency of the public services it performs through its employees outweighed the Deputy Chief’s speech rights. Duke v. Hamil, 2014 WL 414222 (N.D. Ga. 2014).

Applying the Pickering multi-factor test, the Court noted that the Deputy Chief posted the image and statement on his personal Facebook page, which did not identify his employment with the Police Department. Nor did the statement refer to any of the Department's policies, practices, or employees. Thus, there was no indication that he spoke pursuant to his official duties in any way. As a result, the Court concluded that the Deputy Chief spoke as a citizen, not as an employee of the Police Department. Duke at *7. Turning to whether the speech was on a matter of public concern, the Court stated, “Plaintiff's speech can be fairly considered to relate to matters of political concern to the community because a

Confederate flag can communicate an array of messages, among them various political or historical points of view. Combine this symbol with a statement calling for a revolution right after an election, and it is plausible that Plaintiff was expressing his dissatisfaction with Washington politicians. Even if Plaintiff had intended to convey a more radical message by using the Confederate flag and the word revolution, that message would also relate to political and social concerns of the community regardless of how unpopular or controversial that point of view may be. Plaintiff's speech was thus a matter of public concern because it expressed disapproval of elected officials, certainly a topic “upon which ‘free and open debate is vital to informed decision-making by the electorate.’” In that regard, the First Amendment protects his speech unless the government's countervailing interests outweigh his interest in speaking. Id.

However, in balancing the Deputy Chief’s free speech interests against the interests of the Police Department to promote efficiency in the public services it provides, the Court noted the employer has wide latitude to take action against employees who disrupt efficient conduct of its operations. Duke at *8. Like the Gresham Court, special consideration was given here to the notion that the government’s interest in efficient public service is particularly acute in the context of police departments, which “have more specialized concerns than a normal government office.” Moreover, police departments have a particular interest in maintaining “a favorable reputation with the public.” Id. Like Shepherd and Gresham, the Deputy Chief argued that his statements did not cause any disruption. That argument was again met with skepticism: “When close working relationships are essential to fulfilling public responsibilities, a wide degree of deference to the employer's judgment is appropriate…Many of these messages are controversial, divisive, and prejudicial to say the least. Because these potentially offensive messages came from the Department's second-in- command, [the employer] did not have to wait to see if the controversy affected the discipline, mutual respect, or trust among the officers Plaintiff supervised before addressing it.” Id.

The Court also considered the time, place, and manner of the Deputy Chief’s Facebook post. At the time he made the post, he was off duty, thus heightening his First Amendment interest. The Deputy Chief argued that the place of speech favored him because his post was not widely disseminated, but simply posted to his private Facebook account, and he intended it to be viewed only by close family and friends who had access to his page. The Court rejected this claim, noting that, “[D]espite his intentions and his quick removal of it, the post became public after someone provided the image to a television station. This illustrates the very gamble individuals take in posting content on the Internet and the frequent lack of control one has over its further dissemination. And even though

there was no social media policy prohibiting political posts on websites like Facebook, the absence of such a policy did not foreclose a response to speech that compromised the Department's interests.” Duke at *9. After carefully weighing all the factors, the court concluded that the Police Department’s interests outweighed the Deputy Chief’s speech rights. Id.

In another landmark First Amendment retaliation case, Bland v. Roberts (4th Cir 2013) 730 F.3d 368, the Fourth Circuit recently held that the mere act of clicking the “Like” button on Facebook was sufficient speech to trigger constitutional protection. Sheriff B.J. Roberts was slated for re-election in 2009. Sheriff Roberts was not pleased when he learned that a group of employees, including both sworn deputy sheriffs and non-sworn civilian personnel, actively supported his rival candidate, Jim Adams, by clicking “Like” on Adams’ campaign page on Facebook and attending a campaign event for Adams (photos of which were also posted on Facebook). According to the plaintiffs, Sheriff Roberts mandated a culture of political loyalty and was intolerant of any opposition. His senior staff was known to track and monitor the levels of political support of employees, as well as any reluctance or signs of opposition. It was also known that his staff monitored Facebook activity related to the campaign. Sheriff Roberts even allegedly said at several staff meetings, “Don’t be getting on Facebook supporting my opponent.”

Sheriff Roberts won the election, and fired the employees who supported Adams (allegedly for poor performance). Not surprisingly, the terminated employees sued Sheriff Roberts in his individual and official capacities, alleging that he fired them in retaliation for exercising their First Amendment rights of free speech and association.

The first issue the Court had to address was whether clicking the “Like” button even constituted “speech” in the first place. The district court held that merely “liking” a Facebook page was insufficient speech to merit constitutional protection and noted that in usual First Amendment cases, there was at least an actual statement made. Assisted in its analysis by persuasive and insightful amicus briefs submitted by the ACLU and Facebook itself, the very conservative Fourth Circuit disagreed with the district court and concluded that the act of “Liking” a Adams’ page on Facebook was pure speech and symbolic expression, essentially the modern-day equivalent of a front yard campaign sign or bumper sticker.

The Fourth Circuit observed:

“Liking" on Facebook is a way for Facebook users to share information with each other. The "like" button, which is represented by a thumbs-up icon, and the word "like" appear next to different types of Facebook content. Liking something on Facebook is an easy way to let someone know that you enjoy it. Liking a Facebook Page means you are connecting to that Page. When you connect to a Page, it will appear in your timeline and you will appear on the Page as a person who likes that Page. The Page will also be able to post content into your News Feed.

Here, Carter visited the Jim Adams's campaign Facebook page (the "Campaign Page"), which was named "Jim Adams for Hampton Sheriff," and he clicked the "like" button on the Campaign Page. When he did so, the Campaign Page's name and a photo of Adams — which an Adams campaign representative had selected as the Page's icon — were added to Carter's profile, which all Facebook users could view. On Carter's profile, the Campaign Page name served as a link to the Campaign Page. Carter's clicking on the "like" button also caused an announcement that Carter liked the Campaign Page to appear in the news feeds of Carter's friends. And it caused Carter's name and his profile photo to be added to the Campaign Page's "People [Who] Like This" list.

Once one understands the nature of what Carter did by liking the Campaign Page, it becomes apparent that his conduct qualifies as speech. On the most basic level, clicking on the "like" button literally causes to be published the statement that the User "likes" something, which is itself a substantive statement. In the context of a political campaign's Facebook page, the meaning that the user approves of the candidacy whose page is being liked is unmistakable. That a user may use a single mouse click to produce that message that he likes the page instead of typing the same message with several individual key strokes is of no constitutional significance.

Aside from the fact that liking the Campaign Page constituted pure speech, it also was symbolic expression. The distribution of the universally understood "thumbs up" symbol in association with Adams's campaign page, like the actual text that liking the page produced, conveyed that Carter supported Adams's candidacy. In sum, liking a political candidate's campaign page communicates the user's approval of the candidate and supports the campaign by

associating the user with it. In this way, it is the Internet equivalent of displaying a political sign in one's front yard, which the Supreme Court has held is substantive speech. Bland at 385-386 [internal citations omitted].

Turning to the multi-factored Pickering analysis used by the Courts in Shepherd and Gresham, the Bland Court observed there was no dispute that the speech was made in the capacity of a private citizen, but even so, it is well established that an employee can speak as a private citizen in his workplace, even if the content of the speech is "related to the speaker's job. Further, the idea expressed in Carter's speech — that he supported Adams in the 2009 election — clearly related to a matter of public concern. Moreover, the employee’s interest in expressing support for his favored candidate outweighed the Sheriff's interest in providing effective and efficient services to the public. The employee’s speech was political speech, which is entitled to the highest level of protection. Indeed, the public's interest in the employee’s opinions regarding the election may have had particular value to the public in light of his status as a Sheriff's Office employee. In contrast, despite the Sheriff's reference to the need for harmony and discipline in the Sheriff's Office, nothing in the record indicated that the employee’s Facebook support of Adams's campaign did anything in particular to disrupt the office or would have made it more difficult for the employee, the Sheriff, or others to perform their work efficiently. Bland at 387-388 [internal citations omitted].

What is clear from the developing case law is that each case involving First Amendment claims will be evaluated on its own facts under the Pickering test. But, seemingly, where the speech has a detrimental impact on the employer or the abilities of the employee to perform their duties, the employer is likely to be given at least some latitude to regulate the speech and take appropriate corrective action, despite the protections of the First Amendment.

III. SOCIAL MEDIA & LABOR RELATIONS

Some of the most significant activity in social media for employers has been with the National Labor Relations Board (“NLRB”) regarding protected labor activities. While the NLRB does not have jurisdiction over public labor unions in California, the Public Employment Relations Board (“PERB”) has a long history of looking to the NLRB for guidance and using their decisions as persuasive authority in addressing public sector labor issues in California.

Between 2011 and 2012, the NLRB issued a series of reports concerning social media cases.6 The reports note a significant increase in cases dealing with the protected and/or concerted nature of union and employee postings on social media and the lawfulness of employer social media policies. Essentially, the NLRB’s position has been that where employees discuss wages, hours or working conditions on-line, they have engaged in “concerted activity” within the meaning of Section 7 of the National Labor Relations Act (“NLRA”), and they cannot be disciplined for that conduct. However, general on-line complaints about the workplace, particularly if not shared with co-workers, fall outside the protections of the NLRA. Moreover, social media policies that are written in a fashion that punish or deter protected or concerted speech are being struck down as overbroad and unlawful.

In one case, the employer, a nonprofit social services provider, unlawfully discharged five employees who had posted comments on Facebook relating to allegations of poor job performance and lack of service by a coworker. The subject of the Facebook comments learned about them and reported them to the employer as cyber-bullying and harassing behavior. The employees who engaged in the on-line discussion were terminated. According to the Board, the Facebook discussion was a textbook example of concerted activity, even though it transpired on a social network platform. The discussion was initiated by one employee in an appeal to her coworkers for assistance. Through Facebook, she surveyed her coworkers on the issue of job performance to prepare for an anticipated meeting with the Executive Director. The resulting conversation among coworkers about job performance and staffing levels was therefore concerted activity. The Board has found employee statements relating to employee staffing levels protected where it was clear from the context of the statements that they implicated working conditions. This finding of protected activity does not change if employee statements were communicated via the internet. See, e.g., Valley Hospital Medical Center, 351 NLRB 1250, 1252-54 (2007), enfd. sub nom. Nevada Service Employees Union, Local 1107 v. NLRB, 358 F. App’x 783 (9th Cir. 2009). Additionally, the discharged employees did not lose the Act’s protection. Although there was swearing and/or sarcasm in a few of the Facebook posts, the conversation was objectively quite innocuous. The discussion did not rise to the level of “opprobrious” under the Atlantic Steel Co. test, 245 NLRB 814, 816-817 (1979), typically applied to employees disciplined for public outbursts against supervisors.

In another case, an employee was unlawfully terminated for violated employer’s internet and blogging standards, which the Board determined to be overbroad.

6 The complete reports can be downloaded from www.nlrb.gov.

The employee posted a negative remark about the supervisor on her personal Facebook page, which drew supportive responses from her coworkers, and led to further negative comments about the supervisor from the employee. Applying Atlantic Steel, the Board concluded that the employee did not lose the Act’s protections when she referred to her supervisor by such terms as “scumbag.” The comments were made during an online employee discussion of supervisory action, which is protected activity. Regarding the nature of the outburst, the name-calling was not accompanied by verbal or physical threats, and the Board has found more egregious name-calling protected. Accordingly, the last Atlantic Steel factor strongly favored a finding that the conduct was protected. Further, portions of the employer’s policy prohibiting employees from posting pictures of themselves depicting the company in any way violated Section 8(a)(1) of the Act because it would prohibit an employee from engaging in protected activity; for example, an employee could not post a picture of employees carrying a picket sign depicting the company’s name, or wear a t-shirt portraying the company’s logo in connection with a protest involving terms and conditions of employment. The portion of the policy prohibiting employees from making disparaging comments when discussing the company or the employee’s superiors, coworkers, and/or competitors was also deemed unlawful because it contained no limiting language to inform employees that it did not apply to Section 7 activity.

In yet another case, employees who posted comments critical of a sales event held by the employer, a luxury car dealer, were improperly disciplined. The comments related to the poor quality food and beverages provided at the event, which the Board felt related directly to the employees’ concerns about their commissions, and was thus concerted activity under Section 7 of the Act. Similarly, employees who complained on-line about their employer’s tax withholding policies, and those who “Liked” the comments, were improperly disciplined under an unlawfully overbroad and vague internet and bogging policy that prohibited “inappropriate discussions.” This policy language could reasonably tend to chill employees in the exercise of their Section 7 rights and could be interpreted to prohibit discussions related to the terms and conditions of employment. Without a provision excepting Section 7 activity, the policy was unlawful.

In several cases where employees posted comments on social media critical of their employer, but no co-workers joined the conversations, the Board determined there was no concerted activity, and thus, the conduct fell outside the protection of the NLRA. And, in numerous other cases, where an employee aired an individual gripe about a particular supervisor or co-worker, the conduct was not concerted when there was no evidence of a group complaint, even

though co-workers expressed their support with “hang in there” type comments. And, in one case from May 2013, the NLRB ruled that an employee’s Facebook comments that her employer was “full of shit,” and challenging the employer to “FIRE ME…Make my day!” were not protected where the employer accepted the challenge and terminated her employment.

In yet another case, the Board held that a provision in Costco's employee handbook prohibiting employees from electronically posting statements that "damage the company, defame any individual or damage any person's reputation or violate the policies outlined in the Costco Employee Agreement" violated the NLRA by chilling Section 7 rights. Section 7 gives employees the right to engage in “concerted activities” for mutual aid and protection. Simply put, it protects the rights of employees to discuss wages and other working conditions. The Board specifically pointed out that the provision did not include "accompanying language that would tend to restrict its application," therefore allowing employees to "reasonably assume" it applied to protected concerted activity. See Costco Wholesale Club, Case No. 34-CA-012421. In April 2013, in the Dish Network decision, the Board affirmed the Costco decision by holding that similar language in Dish's social media policy prohibiting "disparaging or defamatory comments" about the company violated the NLRA. See Dish Network Corporation and Communications Workers of America, Local 6171, Case Nos. 16-CA-062433 and 16-CA-066142.

This series of studies by the NLRB General Counsel’s office provide employers with critical guidance on the appropriate scope of a social media policy and what kind of on-line discussion is protected under the NLRA.

• Be mindful of the NLRA. It is best to avoid overbroad statements and ambiguous words that could be interpreted to "chill" Section 7 rights. The NLRB has disfavored terms and phrases such as "confidential information" and "disparaging comments" – without further explanation or context – as being too vague and likely to violate the NLRA. Employer policies should not be so sweeping that they prohibit the kind of activity protected by federal labor law, such as the discussion of wages or working conditions among employees.

• Provide examples whenever possible. The Board has repeatedly indicated that a particular policy might have been lawful if it had included specific examples of prohibited conduct. Thus, instead of stating that the policy prohibits "inappropriate behavior," consider providing examples such as harassment, bullying, etc. Employers should work with employment law counsel to craft appropriate and workplace-specific policy language.

• Include a savings clause. These clauses exclude protected Section 7 activity from the scope of a social media policy. Although a savings clause will not cure an otherwise unlawful and overbroad policy, it might enable the court to narrowly construe the policy or excise an offending portion and preserve the rest, and there is no downside to including one.

IV. APPLICANT SCREENING AND SOCIAL MEDIA

Another social media concern being raised by employers involves the use of social media to screen applicants. A 2011 survey by the Society for Human Resource Management found that 56% of companies use social media to find candidates, and more than 25% went on-line to check applicants’ profiles on social media platforms. Over 1/3 of employers who use social media to recruit admit they have decided not to hire someone based on what they found on-line. Among the reasons frequently cited by employers for refusing to hire applicants based on information learned from social media were frequent references to drugs and alcohol, provocative or inappropriate photos, bad-mouthing of a previous employer, co-workers or customers, or discovery that the applicant lied about their employment history or qualifications. This practice also often gives employers access to information about an employee’s protected status, such as age, sexual orientation, disability, race, or religious views. Making an employment decision based on such criteria is unlawful in California and under federal law.

For example, in Nieman v. Grange Mutual Casualty Company, 2012 WL 5029875 (C.D. Ill. 2013), an applicant sued for age discrimination under Title VII and the ADEA claiming that the employer learned his age from his LinkedIn profile (based on the date of his graduation from college). He also alleged that he was not hired because the employer learned through an internet search that he had sued a prior employer. The Court concluded that such conduct by an employer was sufficient to trigger the protections of those statutes and the employer was not permitted to make an employment decision based on age information obtained from LinkedIn.

While social media and the internet provide a wealth of information – which can certainly be tempting for employers – relying on such information creates potential legal risks. Although the information may be public, once an employer has it, it can be difficult to prove that it was not used in the hiring decision.

In recent years, employers were increasingly more insistent on accessing the social media pages of applicants, including by asking the applicant to provide

their password, and/or to open their social media pages during the interview for the employer’s review. Given the difficult job market of the last several years, many employees were willing to do so if it meant they had a better chance at getting the job.

Despite the widespread precedent holding that there is no expectation of privacy in any material posted on social media, there has been a significant legislative backlash against the practice of requiring applicants or employees to share information from their social media profiles. Many states have enacted “password protection” laws that take a more protective view of privacy than have the courts. Here in California, Labor Code §980 went into effect on January 1, 2013. The new statute prohibits an employer from requesting a job applicant or employee for access to his or her social media, except in limited circumstances. Section (b) of the new statute provides that an employer may not “require or request” a job applicant or employee to do any of the following:

(1) Disclose a username or password for the purpose of accessing personal social media;

(2) Access personal social media in the presence of the employer; or

(3) Divulge any personal social media.

The meaning of this third prohibition against asking an employee to “divulge any personal social media” is far from clear. Because divulge is used in a very general sense, and not with respect to any specific information, “divulge” apparently means telling the employer which types of accounts the employee has (e.g. Facebook versus Twitter). Yet, Legislative history materials, including the California Senate’s analysis, suggest that “divulge” means to disclose specifically the username and password of an account.

Notably, the new law does not prohibit accessing employer-issued electronic devices. The statute defines “social media” so broadly it includes e-mails and text messages. Legislators carved out an exception for when employers must access Personal Digital Assistants (“PDAs”) such as Blackberries and iPhones, which may be protected by a password known only to the employee. Labor Code §908(d). Generally, employees have no privacy rights to employer-issued PDAs and computers.

Further, the new law does not prohibit employers from requesting the employee to “divulge” personal social media if it is relevant to a formal investigation. The new law “does not alter an employer's existing rights and obligations to request

an employee to divulge personal social media reasonably believed to be relevant to an investigation of allegations of employee misconduct or employee violation of applicable laws and regulations, provided that the social media is used solely for purposes of that investigation or a related proceeding.” Labor Code §980(c). Moreover, there is no prohibition in the statute against an employer using information from social media that it obtains from sources other than the employee himself or information that is publically accessible. The law is also silent as to whether an employer can use private social media information volunteered by an existing employee that is “friends” with the applicant or employee.7 Even so, because of the ambiguities in the statute, until case law develops to offer additional guidance, employers should err on the side of caution when accessing or seeking access to personal social media accounts.8

Labor Code §980 is silent as to its applicability to public employers, and an amendment introduced in December 2012 to specifically apply the statute to public employers was not successful. However, as a practical matter, public employers should still use extreme caution and seek advice of counsel before requiring an applicant or employee to provide user account information or content.

V. DISCOVERY OF SOCIAL MEDIA IN LITIGATION

In civil litigation, typically, any information “reasonably calculated” to lead to the discovery of admissible evidence is appropriate. The test during discovery is not whether the evidence would be admissible at trial, but simply whether it is

7 Cases involving disgruntled co-workers bringing social media posts to the attention of the employer abound. In Lineberry v. Richards, 2013 WL 438689 (E.D. Mich. 2013), employees brought Facebook photos to a supervisor which showed a co-worker who was supposed to be out on FMLA leave with a back injury riding a motorboat with two children on her lap while on vacation in Mexico. That employee was fired for misusing her FMLA leave. She sued the employer for retaliation for exercising her FMLA rights and not reinstating her. Summary judgment was granted for the defense. In another case, Jaszczyszn v. Advantage Health Physician Network, 2012 WL 5416616 (6th Cir. 2012), the court affirmed summary judgment for an employer who fired a worker for fraudulent use of FMLA leave where co-workers provided their supervisor with Facebook photos of the allegedly “completely incapacitated” employee at a festival while she was on leave. 8 The Legislature did not include any specific penalties for violation of the statute. However, existing law under the Labor Code Private Attorneys General Act, would likely allow an aggrieved employee to file a civil lawsuit, receive a specific penalty amount, and obtain an attorney’s fee award.

sufficiently related to issues raised in the litigation. More and more frequently, social media sites, such as Facebook, Twitter and YouTube are becoming repositories of information that may be relevant and discoverable in legal proceedings. Even where the social media user employs privacy settings, courts have almost universally held that the content is discoverable when shown by the requesting party to be relevant. Indeed, in any case where social media might be a source of relevant evidence, it is important for counsel to immediately send a retention and preservation request to the other side that specifically addresses social media.

Although courts have widely held that information from social media websites is discoverable if relevant, actually obtaining the information can be a challenge. In most cases, a litigant is not going to voluntarily grant access to their social media accounts. Moreover, a civil subpoena directed to the service provider will likely not yield results, due to the federal Stored Communications Act (“SCA”), 18 USC §2701 et seq., which prohibits any entity, such as Facebook or Twitter, from disclosing user content (i.e., messages, timeline posts and photos) without the consent of the owner of the account. 18 USC §2702(b)(3).9 There is no exception in the statute for producing this information when sought pursuant to a civil subpoena. 10

So, how then, do litigants go about obtaining evidence from social media sites? This is an area of law that is not yet fully developed, and there does not appear to be any consistent approach utilized by trial courts or endorsed by the appellate courts. However, there are several possible ways to obtain such information. The method most likely to yield valid and thorough results would be a written consent signed by the user, accompanied by a narrowly tailored subpoena, served to the provider, and obtaining the data directly from the source. Of course, obtaining written consent from a party opponent in litigation is likely wishful thinking.

Another approach is to serve written discovery demands seeking social media account and user information, activity log, and relevant content. With respect to

9 The SCA was enacted in 1986, long before the advent of social media and the erosion of privacy resulting from the regular posting of personal information online. It was initially conceived as a privacy protection statute to apply the Fourth Amendment’s protections to the virtual world. The statute is out of date, dense, confusing, and full of gaps in need of legislative attention. Both state and federal courts across the country have struggled with its application to modern circumstances and technology. 10 There are exceptions for law enforcement in the statute, and Facebook does cooperate with police to produce information relevant to criminal investigations and prosecutions.

information from a Facebook page, the discovery requests should specifically direct the responding party to Facebook’s “Download Your Information” option, which is located under the “Settings” drop down menu. This feature allows the user to download their account data, including posts, photos, messages and other information. After the user hits the “Download” button, Facebook will prepare all of the information into a single file and send the user an email with a link when the information is compiled and ready to be downloaded. The file can then be downloaded directly to a computer desktop or removable storage device.11 Discovery requests should also seek the user’s “Activity Log,” which includes a list of posts, stories and photos the user has been tagged in, life events, and connections, organized by the date they happened on Facebook. The discovery requests should direct the responding party to Facebook’s instructions for accessing the activity log feature. The following types of information are available using these Facebook features12 :

About Me Account Status History Active Sessions Ads Clicked Address Ad Topics Alternate Name Apps Birthday Visibility Chat Check-ins Connections Credit Cards Currency

11 This Facebook video explains how the Download Your Information feature works: https://www.facebook.com/video/video.php?v=10150292657680484. Unfortunately, Twitter currently does not have a similar self-collection mechanism. However, it’s Terms of Service clearly state that a Twitter user provides Twitter a license to distribute to anyone at any time whatever the user tweets. In People v. Harris, (2012) 945 N.Y.S.2d 505, a criminal prosecution of an Occupy Wall Street protestor, the prosecution served a subpoena on Twitter. The Court denied motions to quash filed by both the defendant and Twitter holding that the user had no proprietary interest or expectation of privacy in his tweets and that by submitting tweets, he had granted Twitter an unlimited license to use and distribute his tweets. Similar results have occurred in civil cases regarding the Terms of Service for other social media. 12 Instructions for accessing Facebook Data and details about each type of data can be found at: https://www.facebook.com/help/www/405183566203254/

Date of Birth Deleted Friends Education Email Address Events Facial Recognition Data Family Favorite Quotes Followers Following Friend Requests Friends Gender Groups Hidden From Newsfeed Hometown IP Address Last Location Likes On Others’ Posts Likes on Your Posts From Others Linked Accounts Locale Logins Logouts Messages Name Name Changes Networks Notes Notification Settings Pages You Admin Pending Friend Requests Phone Numbers Photos Photos Metadata Pokes Political Views Posts By You Posts By Others Posts To Others Privacy Settings Recent Activities

Registration Date Religious Views Removed Friends Screen Names Searches Shares Spoken Languages Status Updates Work Vanity URL Videos

In light of how much information is potentially available, it becomes evident why courts require social media discovery requests to be narrowly tailored. For example, a request for “all content” would draw overbreadth, relevance and potential privacy objections, which likely would be sustained by a court. Accordingly, when crafting these kinds of discovery requests, the requesting party should be very specific about which of the above types of data is being sought, the pertinent time frame, and the subject matter of the content.

Assuming the responding party refuses to comply with narrowly tailored social media requests, the moving party can seek an order compelling the responding to comply. According to at least one California court, where the party to a communication is a party to the litigation, the trial court has the power to require the party to consent to the disclosure (thus bypassing the Stored Communications Act). See O’Grady v. Superior Court (2006) 139 Cal. App. 4th 1423, 1446. Indeed, the developing case law demonstrates that compelling an authorization by the social media user for release of the information rather than compelling the the social media provider to produce it is a more successful method. See Romano v. Steelcase, Inc., (2010) 907 N.Y.S. 2d 650 (Material on plaintiff’s public Facebook and MySpace pages revealed she traveled and lived an active lifestyle, despite claims of debilitating injury; Court ordered plaintiff to provide authorization to defendant to access her private pages); Ledbetter v. WalMart Stores, Inc., 2009 WL 1067018 (D. Col. 2009)(Court ordered plaintiff to produce contents of Facebook, MySpace and Meetup.com accounts because public information on those accounts contradicted claims of injury).

If the Court is willing to order the disclosure, it would be prudent for the requesting party to request that an electronic discovery expert is appointed to assist with the data collection using the download or data retrieval features offered by the social media provider. Any responding party should request that

the data and content be produced directly to his or her counsel, and then subject to in camera review before production to the other side.13

In one oft cited case where broad discovery was granted, EEOC v. Simply Storage Management, 270 F.R.D. 430 (S.D. Ind. 2010), the EEOC brought a Title VII sexual harassment action on behalf of two employees against their employer. During discovery, the employer sought production of photos, videos and complete profile information, including all updates, changes, or modifications to the profiles, all status updates, messages, wall comments, causes joined, activity streams, blog entries, details, comments and applications, on the grounds that the employees had placed their mental states at issue by alleging emotional distress. The EEOC objected to the production as overlybroad, irrelevant, burdensome, harassing, and an invasion of privacy. The EEOC insisted that production should be limited to content that directly addresses or comments on matters alleged in the complaint; the employer argued that the nature of the alleged emotional injuries potentially implicated all social media communications. In evaluating the proper scope of production, the court held that social media content is not “private” just because privacy settings are utilized. Id. at 434. Moreover, relevance to a claim or defense depends on the substance of the communication, and if relevant, the information should be produced. Id. at 435. Thus, the court determined that the proper scope of discovery included profiles, posts or messages related to any emotion, feeling or mental state, or communications that revealed, referred or related to events that could reasonably be expected to produce a significant emotion, feeling or mental state. With respect to videos and photos, only those that depicted the claimants were ordered produced (on the grounds that the context of the picture and the claimant’s appearance might reveal emotional status). See also Reid v. Ingerman Smith LLP 2012 WL 6720752 (E.D.N.Y. 2012)(legal secretary sued for sexual harassment and sought damages for emotional distress; employer obtained private Facebook postings by showing that her posts contradicted her claims for mental anguish); Thompson v. Autiliv ASP, Inc. 2012 WL 234928 (D. Nev. 2012)(Plaintiff sought damages for emotional distress, depression and physical injuries, but defendant found wall posts and photos on Facebook undermining those claims. Plaintiff belatedly changed her privacy settings and produced redacted versions of the material; Court ordered all Facebook and

13 At least one court has come up with a creative alternative. In Barnes v. CUS Nashville, 2010 WL 2265668 (M.D. Tenn. 2010), a magistrate judge offered to create a Facebook account if two witnesses were willing to accept the judge as a “friend” solely for the purpose of reviewing photos and related comments in camera. After reviewing and disseminating any relevant information to the parties, the magistrate judge would close the account.

MySpace information for more than a 5 year period produced without requiring an in camera review).

In contrast, a discovery request seeking content related to “any emotion, feeling or mental state of the plaintiff” in a workplace discrimination case involving severe emotional distress allegations was held to be impermissibly overbroad in Mailhoit v. Home Depot, 285 F.R.D. 566 (C.D. Cal. 2012). According to the plaintiff, rather than narrowly tailor its requests, the employer sought “to rummage through the entirety of [plaintiff’s] social media profiles and communications in the hope of concocting some inference about her state of mind.” Id. at 569. The Court agreed: “To be sure, anything that a person says or does might in some theoretical sense be reflective of her emotional state. But that is hardly justification for requiring the production of every thought she may have reduced to writing…Arguably, watching a football game or a movie on television is an ‘event’ that may produce some sort of ‘significant emotion,’” but such material would not be relevant. Id. at 571-572. Ultimately, the requests were deemed overbroad and they failed to describe the requested content with “reasonable particularity.” See also Howell v. Buckeye Ranch, 2013 WL 1282518 (S.D. Ohio 2012)(Motion to compel production of user names and passwords for each social media site denied because request was too overbroad and not limited to information relevant to alleged sexual harassment and emotional distress).

Ultimately, whether and to what extent a court will compel disclosure of social media content will likely depend on numerous factors, including the relevance of the information to the claims in the case, the scope of the requests, privacy settings and public access to the information, and the judge’s familiarity with the technology at issue. Even if a party can easily articulate why discovery of social media content is relevant, one can expect a contentious and costly fight to get it, and the outcome of such requests is far from certain.

It is also important to avoid spoliation claims in connection with potential social media evidence, and clients should be advised to maintain any potentially relevant social media posts. In Lester v. Allied Concrete Company, Va. Cir. Ct. CL09-223 (Final Order 2011), the managing partner of one of the state's largest personal injury firms told his client to remove several photos from his Facebook account on fears that they would prejudice the client's wrongful death case brought after his spouse’s fatal automobile accident. One of the photos depicted the allegedly distraught widower holding a beer and wearing a t-shirt emblazoned with “I [heart] hot moms.” Murray instructed his client through his assistant to “clean up” his Facebook account. “We do not want blow ups of other pics at trial,” the assistant’s email to Lester said, “so please, please clean up your

Facebook and MySpace!”' The Court imposed sanctions against the client in the amount of $180,000, and against attorney in the amount of $542,000 for spoliation of evidence. Further, the Court referred ethics-based allegations against the attorney to the Virginia State Bar and perjury charges against the client were referred to the local prosecuting authority.

In another recent case, Gatto v. United Airlines and Allied Aviation Services, Inc.,10-cv-1090-ES-SCM (D. NJ. 2013), a grounds operation supervisor at JFK International Airport, was injured on the job. He sued United Airlines and Allied Aviation Services, claiming they were responsible for the accident which left him permanently disabled and unable to work. During litigation, Gatto was ordered to give Allied and United access to his Facebook account. However, he deactivated the account, which led to its deletion by Facebook two weeks later. Although Gatto claimed he deactivated the account because he thought it had been hacked and that he never intended to destroy the contents, at trial, the judge gave the jury an adverse inference or spoilation instruction, advising the jury it was free to infer that the content of his Facebook page would have harmed his case against United and Allied.

VI. SOCIAL MEDIA, PUBLIC RECORDS & PUBLIC OFFICIALS

Many cities and public agencies utilize social media as a means of communicating with citizens, local businesses and constituents. Official posts by cities or other public agencies on social media sites are public records and should be maintained as such pursuant to the California Public Records Act (“PRA”). This practice also raises questions about establishing a proper posting policy and concerns about the monitoring (and possible deletion) of public comments.

One emerging issue in California relates to the question of whether communications by public officials and employees retained only on personal electronic devices and private accounts are subject to disclosure under the PRA. The Court of Appeal for the Sixth District issued its published decision on March 27, 2014 in the case of City of San Jose v. Superior Court, Case No. H039498, Sixth Appellate District. In the case, a citizen had sought "voicemails, emails or text messages sent or received on private electronic devices" used by certain City officials under the PRA. The City disclosed the responsive, non-exempt records sent from private devices using City accounts, but not records sent using personal accounts, on the grounds that records sent using personal accounts were not public records. Smith sued the City to compel disclosure of the records sent from private electronic devices using personal accounts.

The issue before the Court of Appeal was "whether those private communications, which are not stored on City servers and are not directly accessible by the City, are nonetheless 'public records' within the meaning of the California Public Records Act." The Court held that those communications are not public records. Specifically, the Court found that "the language of the CPRA does not afford a construction that imposes on the City an affirmative duty to produce messages stored on personal electronic devices and accounts that are inaccessible to the agency, or to search those devices and accounts of its employees and officials upon a CPRA request for messages relating to City business."

Section 6252 of the PRA defines a public record as "any writing containing information relating to the conduct of the public's business prepared, owned, used or retained by any state or local agency." In reaching its conclusion, the Court focused on both this definition, as well as principles of statutory construction.

The Court agreed with the City's arguments that under the express terms of Section 6252, only those writings prepared, owned, used or retained by an agency are public records; writings that are prepared, owned, used or retained by an agency's employees and officials but not the agency itself are not public records under the PRA. The Court stated that while it may be true that an agency can only act through its officials, not every act by an official is an act of the agency.

Further, the Court noted that the definition of "local agency" in Section 6252 refers to governmental bodies themselves, not individual members or representatives of the governmental body. "Because it is the agency - here, the City - that must prepare, own, use or retain the writing in order for it to be a public record, those writings that are not accessible by the City cannot be said to fall within the statutory definition. The City cannot, for example, 'use' or 'retain' a text message sent from a council member's smartphone that is not linked to a City server or City account."

This decision has significant implications for local officials' communications related to City business conducted entirely on privately owned and maintained electronic devices. While it may be prudent to wait to see if this case will be reviewed by the California Supreme Court, according to City of San Jose v. Superior Court, local officials' private email and mobile phone records will not be subject to disclosure under the Public Records Act. Officials should also be aware that the decision involved truly "private" devices, it does not address the question of whether communications on "private devices" for which the public

agency provides reimbursement with public funds would be subject to the Act. A reviewing court could reach a different conclusion in that context.

Many elected officials use blogs, Facebook and other social media outlets to connect with constituents and to promote political agendas. This is a reasonable use of social media, but elected officials should not use official city social media platforms for campaigning purposes.

Elected officials should also be mindful of the Brown Act, which requires public notice and access to meetings. A quorum of the governing body, including city councils, commissions, boards or committees may not meet to discuss official business unless the rules of the Brown Act are met. So, for example, if a city council has five members, three of them may not meet for dinner at a local restaurant to discuss City business. This raises the potential pitfalls of council members engaging in business-related conversations or debates on-line. If one member were to post a comment on his personal Facebook page about an on- going City project, and several other members commented (or perhaps even “Liked”?) the comment, it is possible that the on-line conversation would run afoul of the Brown Act.

Additionally, personal social media account names should not be tied to the City. This will help clarify that the individual is not speaking officially on behalf of the City. Elected officials who use social media should reveal that they are an elected official and include a statement that any opinions they post are their own, not those of the City. They should also be mindful of the need to abide by privacy and confidentiality laws and avoid sharing non-public information related to co-workers, personnel data, medical information, claims or lawsuits, or other sensitive, non-public, or confidential information. Elected officials should not post official information about the City or City business on their personal sites in order to avoid mistakes and create unforeseen liability issues. Elected officials also need to be aware that using personal technology to communicate about official City business can create issues if a request for data is made or a hard drive or mobile device is subpoenaed in connection with a lawsuit. Finally, elected officials need to follow existing City policies related to electronic communications among council members and guidelines related to the use of City technology.14

14 Increasingly, Cities are adopting policies that disallow members of the City Council, commissions and boards to like, share, retweet or engage in published postings in serial meetings on any issue within the subject matter jurisdiction of the body.

VII. SOCIAL MEDIA & LAW ENFORCEMENT

In addition to the concerns stemming from conduct on public social media sites such as Facebook and Twitter, there are a number of emerging social media sites specifically geared towards the law enforcement community.

With more than 1.5 million unique visitors per month and over 450,000 registered law enforcement members from 14,000 agencies, the leading law enforcement social media site is PoliceOne.com. By the end of 2013, PoliceOne actually had a Facebook community of more than 350,000 fans and close to 20,000 followers on Twitter. PoliceOne purports to be “a trusted and safe community for the exchange and discussion of key law enforcement issues, including breaking news, top products and important officer safety strategies.” Users create a profile page and have access to news stories, forums, products, articles, and career opportunities. With respect to user’s personal information, PoliceOne states, “At PoliceOne, we take your privacy very seriously…We pledge not to sell or otherwise misuse information from your membership profile.” While PoliceOne is an excellent resource for news, training, and industry specific information, its interactive capabilities are somewhat limited.

In August 2013, 20for25.com launched as the social media site “for cops by cops.”15 According to one of the site’s founders, Josh Olson, a current peace officer, “Law enforcement professionals need a place to network, share training ideas, and socialize in a medium that is safe and secure.”16 Any current local, state, or federal law enforcement officer can join the site. Before the account is activated, 20for25 will confirm that the user is an active peace officer and verify the user’s credentials with the employing agency. 20for25 touts itself for its advanced privacy settings, exclusive community, and greater user control over their content. For example, when a post or photo is deleted, there are no backups kept and the material cannot be recovered. 20for25 stresses that it will not “use, sell or disclose your personal information for any other purpose without your consent unless required by law.” (Emphasis added).

In October 2013, former Police Commissioner Bill Bratton, launched the BlueLine, touted as “the secure professional network built exclusively for law enforcement.” Bratton’s New York-based venture capital backed startup, Bratton Technologies, was responsible for creating and launching the website. Available to accredited members of law enforcement, the website is held out as a place

15 The website’s name comes from common police parlance: 20 refers to location and 25 refers to a place to meet. 16 http://www.prweb.com/releases/2013/8/prweb11038036.htm

where officers can share their expertise, insight and information. The platform offers instant messaging, videoconferencing and screen share capabilities. Members can create databases, have PowerPoint meetings and search for other members via name, topics and interests. Each member creates a profile page, sees a news feed, can comment on posts, join groups and view other member’s profiles. It is also a marketplace for law enforcement gear and merchandize. The website looks and feels like LinkedIn with advertisements by vendors who target law enforcement agencies. “Our focus is to have a walled community where you're verified and authenticated, so you have a safe form of communication with law enforcement, analysts and administrators,” said David Riker, Bratton Technologies' president.17 BlueLine requires multiple verifications for members of law enforcement to join and enter the network, and the platform is housed in a secure data center. Similar to 20for25, Bratton Technologies claims it will not share information without permission - except “when we believe release is appropriate to comply with the law or to protect the rights, property, or safety of Bratton Technologies, our users, or others.” (Emphasis added).

While these industry-specific social media platforms do offer useful new tools, methods of communication and data sharing, they create a false sense of privacy and security. There are no privileges that protect on-line communications between police officers, and the developing law confirms there is little, if any, expectation of privacy in anything posted on a social media website, even where privacy controls are in place. Did you notice the carefully worded privacy disclaimers? Personal information won’t be disclosed “unless compelled by law” and “except when we believe release is appropriate…” Those very phrases arguably undermine any expectation of privacy because they clearly contemplate producing personal content or information when compelled to do so by subpoena, court order or other legal process.

Further, while these websites claim that they are not intended for sharing information pertaining to specific crimes or investigations, the systems are self- policing. If a user posts information about a specific case or personal information about another officer or user, other members can flag it for removal. BlueLine says it is confident this will be enough to deter inappropriate posts, and Bratton has stated, “There aren’t a lot of police officers on [Facebook, LinkedIn, and other social sites]. Police officers tend to be very guarded about their privacy and their personal lives. . . . We make it quite clear this isn’t Facebook. This is not for showing pictures of the kids or your vacation.”18 Even so, one of the very first comments posted on the BlueLine by an officer during the beta testing phase

17 http://news.yahoo.com/social-network-law-enforcement-launch-173004559.html 18 Inside BlueLine, The Social Network For Police, Fast Company, October 2013.

was, “This is amazing! Our agency policy on Facebook is very restrictive so I have been looking for a professional networking tool just for us!” That sentiment perfectly illustrates the concerns set forth in this paper.

In fact, the public comments to various articles on the internet about these new law enforcement social networking sites are themselves telling:

• “Where and when do we serve our FOIA requests?” • “The government says it needs to know our communications for our safety, and they also say they need us to not know their communications for our safety. This is like some modus ponens test, right?” • “‘Our focus is to have a walled community’…There folks, is the entire problem.” • “Every criminal defense attorney who doesn’t serve this site with a subpoena requiring every post related to his client…should be reported to his bar association.” • “This is going to become just another good, easy way to get fired.”

Public posts commenting on active cases can jeopardize investigations and can even end up as grounds for successful appeals. Comments or posts by an officer that could be construed as harassing, bullying, disparaging or that breach confidentiality are potential grounds for internal affairs investigations and workplace discipline. Comments critical of an investigator or particular investigation technique would likely be relevant and discoverable if requested by a criminal defense attorney. Officers should assume that anything they say or post on a law enforcement website (or any other online forum) is potentially subject to discovery.

It is imperative for every law enforcement agency to develop a comprehensive social media policy that specifically addresses appropriate use, both on public and law enforcement social media sites. To the extent the policy will provide for potential officer discipline, notice will need to be given to the union and the department will likely need to meet and confer. Officers should be trained about proper use of social media so they understand the potential personal and agency-wide risks they create when engaging in on-line conduct.

VIII. CONCLUSION

As the law related to social media continues to develop and expand, it is important for public employers to develop policies that are narrowly-tailored and mindful of labor relations, free speech and public records issues. Elected

officials need to be trained about the particular constraints of the Brown Act and their personal use of social media. It is also important for law enforcement agencies to stay abreast of emerging social media trends and develop clear policies that specifically regulate the use of industry-specific websites by officers. Managers and human resource professionals should be familiar with the potential risks associated with using social media in making hiring decisions and for internal investigations and discipline purposes.

Social media technology continues to develop and change at a pace far more rapid than the courts and Legislature can keep up with. The facts and circumstances of each case are different, and bright line rules do not yet exist, so it remains difficult to predict how a court will rule on any given issue. Thus, it is incumbent upon cities and public agencies to be mindful of their practices and policies and to consult with legal counsel as questions arise.

This page left intentionally blank.

League of California Cities® 2014 Spring Conference Renaissance Esmeralda, Indian Wells