Social Media Policies: Implications for Contemporary Notions of Corporate Social Responsibility

Abstract

Three global developments situate the context of this investigation: the increasing use of social media by organizations and their employees, the burgeoning presence of social media policies, and the heightened focus on corporate social responsibility (CSR). In this study the intersection of these trends is examined through a content analysis of 112 publicly available social media policies from the largest corporations in the world. The extent to which social media policies facilitate and/or constrain the communicative sensibilities and values associated with contemporary notions of CSR is considered. Overall, findings indicate that a large majority of policies, regardless of sector or national headquarters, increasingly inhibit communicative tenets of contemporary CSR (i.e., free speech, collective information sharing, and stakeholder engagement/dialogue) and thereby diminish employee negotiation and participation in the social responsibilities of corporations. Moreover, policies generally enact organizational communication practices that are contrary to international CSR guidelines (e.g., the UN Global Compact and other international agreements). Findings suggest that social media policies represent a relatively unrecognized development in the institutionalization of CSR communicative norms and practices that call into question the promising affordances of social media for the inclusion of various voices in the public negotiation of what constitutes corporate social responsibility.

Introduction

We recognize that you will use social media channels outside work time in a private capacity. However, such use can still have an impact on your employment. You are required to act in the Group’s best interest at all times and this extends to your participation in and use of social media channels.

Commonwealth Bank of Australia, Social Media Policy (#226, Fortune Global 500, 2013)

When using social media, one should give ample consideration to basic human rights, personal rights, privacy rights and the treatment of personal information, etc. while also obeying all relevant laws and regulations as well as the Code of Conduct.

Sumitomo Rubber Group, Social Media Policy (#129, Financial Times Global 500, 2013)

From think tank surveys (Pew Internet Research Project 2014) to academic research (e.g., Steinfield et al. 2013), industry initiatives (Proskauer 2012, 2014) to media reports (Bonvanie 2012), the increasing use, opportunities, and communicative challenges of social media use in organizational contexts have been well documented. Social media refer to platforms or “internet-based applications built on the ideological and technological foundations of Web 2.0” (Kaplan and Haenlein 2010, p. 271). Social media allow for the creation and exchange of user-generated content, moving beyond document and data exchange to influencing organizational structures and networks of relations (Linke and Zerfass 2013). Research suggests that between 2013 and 2016 the percentage of internet users worldwide active on social media sites will increase from 67.7 to 76.6 % (Sigsworth 2013). In the corporate world, 80 % of Fortune 500 companies have one or several Facebook accounts; over 80 % use Twitter; and over 30 % of these companies employ blogs (Barnes and Lescaut 2014). And their use will continue to increase as the number of social media platforms multiplies.

Concomitantly, there is escalating and large scale interest in Corporate Social Responsibility (CSR). CSR has become a ubiquitous “buzzword” of corporate speak (Choudhary and Singh 2012), the focus of multiple investigatory and regulatory activities by governments, international agencies, and NGOs (Richter 2001), and a topic of much debate within civil society (May et al. 2007). CSR represents the ongoing global dialogue and activities regarding the financial, social, communicative, and environmental expectations and responsibilities of corporations towards their shareholders, employees, communities, and external stakeholders (Black 2005). Although still a somewhat contested term, global acceptance of the idea that corporations have responsibilities beyond profit-making (i.e., the classical understanding of first-generation CSR; Stohl et al. 2007) is evidenced by the large number of national and international CSR initiatives and mechanisms dealing with issues of sustainability, transparency, community engagement, and human rights (Griffin and Prakash 2014), including “the largest voluntary corporate responsibility initiative in the world”—the UN Global Compact.

These trends are intersecting and represent possibilities for a new generation of CSR. Social media provide dynamic and far-ranging opportunities for CSR communication as a result of their unparalleled affordances (i.e., visibility, persistence, editability, and association; Treem and Leonardi 2013). The platforms enable informal, interactive, and publicly visible CSR communication among employees, management, customers, clients, and the general public in relatively cheap, efficient, and novel ways (Etter 2014; Fieseler et al. 2010). Social media facilitate organizational information sharing about CSR activities, campaigns, brand management, as well as identification of and responsiveness to customer concerns and stakeholder needs (Eberle et al. 2013; Hearn et al. 2009; Lee et al. 2013). Social media can also become a model of CSR communication itself, enabling increased involvement and voice, providing the means to engage, organize, and create dialogue among global stakeholders and corporate actors. These are the discursive processes that constitute contemporary views of CSR (Stohl 2012).

However, as highly public “social media disasters” illustrate, including the 2008 “United Breaks Guitars” viral video and the 2014 real-time tweets by an HMV employee as workers were being summarily terminated, organizations are painfully aware that social media may also expose and generate public discourse about irresponsible business practices. Consequently, there is grave concern that social media may negatively influence organizational branding, reputation, creditability, and trust (Schlinke and Crain 2013) because they enable what organizations deem as employee breaches of confidentiality and “inappropriate” disclosures (Hekkala et al. 2012), as well as unprofessional or irresponsible employee commentary and responses—all of which have the potential to publicize and problematize irresponsible business practices.

Such cascading and viral stakeholder sequences highlight another essential feature of social media that has implications for CSR: the blurring of material and spatio-temporal boundaries (Bimber et al. 2005). Through social media, work-related incidents can become part of personal and public conversations, both online and offline. Further, employees increasingly use work-situated communication technologies for private purposes (e.g., checking personal social media accounts, sharing experiences with friends) and personal technologies for work-related tasks (e.g., accessing work email from a home computer). Today work-related communication can easily take place at home and/or in private spaces both before and after formal work hours, just as private communication can occur in the workplace. The HMV employee referenced above, who used a corporate Twitter account to express her personal outrage, knew that she had blurred the boundaries between personal and professional communication, but felt it was justified since the company she “loved” was being destroyed (Adams 2013). Overall, the permeability of these boundaries raises fundamental questions around what constitutes employee communication, as well as the scope of organizational expectations, responsibilities, and control mechanisms.

Given these global trends and related concerns, the rapid development of social media policies regulating how employers and employees navigate the benefits and mitigate the risks of this new and tumultuous communication landscape is not surprising. Corporations are looking to industry leaders, national and international regulatory environments, and professional consultants and societies to design the policies. Such activities are associated with the mimetic, coercive, and normative mechanisms of institutional isomorphism (DiMaggio and Powell 1983), the processes by which similar taken-for-granted, normative understandings and habitualized patterns and/or rules about social interaction and social practices are established and justified (Lammers and Garcia 2014).

As evidence of the developing institutionalization of social media policies, an entire internet cottage industry has been created to help organizations formulate these policies (Meister 2013). Industry surveys indicate that between 24 and 69 % of corporations from various industries have social media policies (Grant 2012; Protiviti 2013). A Google search of the term “social media policy consultants” yielded more than 78 million results in English, 70 million in Spanish (Consultor de Política Social Media), 82 thousand in Japanese (ソーシャルメディア政策コンサルタント), and 136 thousand in Chinese simplified (社会化媒体策略顾问). Like Codes of Conduct, these policies articulate organizational expectations, responsibilities, and consequences for employees and employers and define what comprises “(in)appropriate” communication. In actively framing, regulating, and sanctioning employee speech and simultaneously transgressing boundaries of private and public spheres, questions of corporations’ responsibilities towards employee communication arise. Focusing on the content and boundary specifications within the social media policies of 112 of the world’s largest corporations, we explore the ways in which social media policies represent a relatively unrecognized development in the emergence of normative communicative expectations and institutionalization of CSR.

Linking Social Media Policies to CSR

Recent scholarship on social media policies has examined: (a) the social and technical rationales for the development and implementation of policies (Vaast and Kaganer 2013); (b) their impact on organizational performance (Weigel 2013); (c) the corporate values embedded in the guidelines (Fuduric and Mandelli 2014); and (d) legal constraints of the policies regarding workplace confidentiality, privacy, and monitoring (e.g., Flynn 2012). Typically, when scholars and practitioners link CSR to social media, it is not through policies, but rather the instrumentality of communication (e.g., Etter 2014). Studies suggest that organizations increasingly use social media to communicate their CSR efforts to diverse stakeholder groups (e.g., Capriotti 2011; Etter 2013), but there is a lively debate over the efficacy of utilizing social media for publicizing an organization’s CSR efforts and enhancing its brand versus the increased danger that stakeholders will use social media to monitor, critique, expose, and voice skepticism regarding an organization’s CSR efforts (Lindgreen and Swaen 2010; Whelan et al. 2013). In the first case, social media are positioned within a reputational and/or business frame wherein communication is conceived as a public relations instrument for achieving corporate advantage and mitigating harm (Linke and Zerfass, 2013). In the latter case, social media are positioned within a risk frame in which communication is viewed as a vehicle for disclosing or exposing information that may be harmful to the organization.

What is missing from this literature is an analysis of the ways in which CSR is constituted and enacted through social media communication. Yet, recent communication scholarship specifies a relationship between the participatory characteristics of social media and the potential of these new information and communication technologies to create an arena where various stakeholders might negotiate the expectations and social responsibilities of organizations (Schultz et al. 2013; Whelan et al. 2013). Castelló et al. (2013), for example, explicate the ways in which the increased speed and connectivity of social media lead to a “dynamization of communication making it more indeterminate as it fosters multi-directional outcomes of communication” (p. 685) and argue that CSR emerges out of this polyphony. However, they do not consider the ways in which corporate regulation of the plurality of voices may constitute, coopt, or prevent particular conceptions of CSR.

Thus, we suggest a third conceptualization of social media in which communication is viewed as a mechanism for transparency, employee dialogue, and employer, employee, and community rights—a co-creation frame for social media. We argue that social media policies themselves are material manifestations of a company’s CSR repertoire, culture, strategy, and underlying organizational belief systems. As such, they provide a powerful text for understanding the communicative tensions and dynamics embedded in contemporary notions of CSR. Accordingly, our overarching research question is: To what extent do organizations’ social media policies embody the sensibilities and communicative values associated with second and third generation CSR (i.e. employee rights and dialogic processes) or represent first-generation CSR ideas (i.e., adherence to laws to facilitate profit-making)? We explore the ways in which these policies are framed for stakeholders, the communicative constraints and opportunities embedded in the guidelines and the boundary conditions under which these regulations apply. Before presenting the empirical study, we explicate the nature and role of employee communication as conceived in different generations of CSR and identify five central communication tenets of contemporary conceptualizations of CSR.

Conceptualizing CSR and the Role of Employee Communication

Although conceptualizing and institutionalizing CSR has evolved over the last century (see Stohl et al. 2007 for a detailed discussion of three generations of CSR), it remains a highly contested and somewhat ambiguous concept (Ihlen et al. 2011). Developed within differentiated economic, political, historical, social, and technological circumstances, each generation embodies distinct and at times contrasting values, communicative expectations, and regulatory mechanisms. However, these generations are neither mutually exclusive nor necessarily temporally discrete. Rather, as in a family where generations reside together under the same roof continually influencing and being influenced by one another, within the multiple and fragmented contemporary CSR discourses, there are persistent traces of the competing as well as complementary normative expectations of each generation.

The first generation of CSR—still quite common in discussions of the “business case for CSR” (Carroll and Shabana 2010)—is epitomized by Milton Friedman’s (1970) classic and narrow formulation: the social responsibility of business is to increase its profit. First-generation CSR is centered on maximizing returns to the general shareholders of a company in ways that are consistent with local laws. Within this conceptualization, employee communication is viewed as an instrument to create economic value. It is heralded as an opportunity to promote the quality of brands and products, create new business relationships, illustrate the responsiveness of the company, develop customer and client trust, and scout new business opportunities by monitoring the marketplace. Employees are regarded as important ambassadors of corporate CSR programs (Morsing et al. 2008) and corporate representatives whose voices and identities are expected to be consistent with organizational values and initiatives (Lee et al. 2013). Employee communication with external stakeholders provides the opportunity for the enhancement of the organization’s reputation as a socially responsible company that is good place to work and an ethical place to conduct business transactions (Fombrun 2005). Such branding is believed to achieve economic benefits, and especially increased sales (Du et al. 2010).

The second generation of CSR was born out of a response to the social upheaval associated with expansion of large industrial enterprises within Western European capitalist democracies in the late nineteenth and early twentieth century. Social visionaries such as Robert Owen and the Cadbury and Lever families argued that it was not only a corporation’s responsibility to make a profitable return on investment for their stockholders, but also to improve employee welfare and protect individual freedoms. Voluntary corporate philanthropic activity, charitable giving, and benefits to improve the lives and working conditions of employees and their families are all part of second generation CSR activities (Stohl et al. 2007). Employees are viewed as critical stakeholders whose individual civil liberties, including freedom of speech and rights to privacy and assembly, must be protected within the organizational context. Articles 17, 18, and 19 of the UN Declaration of Human Rights, a text used today as a referent for virtually all international standards and agreements regarding the institutionalization of CSR, including the UN Global Compact and the OECD corporate guidelines for responsible behavior, refer directly to individual rights of privacy, freedom of thought, expression, conscience, and religion. Importantly, these rights also include “freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers” (Council of Europe 2010). Along with these rights come certain obligations and communicative responsibilities including honesty, transparency, and respect.

The third generation of CSR retains the importance of employee rights but expands financial, social, and communicative responsibilities to larger set of stakeholders within a context of intensifying global connectedness, aspirational discourse, and collaboration and among corporate, non-governmental, and governmental organizations (Ihlen and Roper 2014). Responsibilities are expected to be mutually defined and enacted in an emerging ethical contract that is negotiated and agreed upon through a process of stakeholder dialogue (Black 2005). Communication is no longer merely instrumental, but also relational and dialogic. Undertakings that were formerly regarded as activities of the political system are considered to be part of corporate spheres of influence (Harman and Porter 1997; Scherer and Palazzo 2007, 2011). For instance, in 2006, when Amnesty International first issued a report condemning Yahoo!, Google, and Microsoft for their complicity in internet censorship in China, Google justified the launching of its Chinese search engine, as a socially responsible business decision because China was such a lucrative market (Amnesty International 2006). But today, after public outrage and protests, many corporations—including Google—have signed on to the Global Network Initiative (Lengnick-Hall and Wolff 1999), a multi-stakeholder CSR project through which they not only make a commitment to core principles of free expression and privacy for their employees and the community, but also agree to be accountable and evaluated independently on the extent to which they actually protect the global public from violations of communication rights.

In the third generation of CSR, corporations are no longer depoliticized, private business actors, but instead entities responsible for setting, implementing, and developing the norms and values of contemporary society (Crane et al. 2008; Scherer and Palazzo 2011). Communicative boundaries between public and private are waning—not only in the sense of developing norms of transparency and openness, but also in terms of accountability and upholding societal values that transcend local organizational contexts. Under these conditions, the ethical mandates of corporate activities are no longer taken for granted; social responsibility cannot be defined a priori by the corporation itself, but is instead negotiated through deliberative public discourse (Gilbert and Behnam 2009; Hess 2008; Rasche and Esser 2006; Roloff 2008; Stansbury 2009). From a third-generation CSR perspective, then, communication is not simply a mechanism through which organizations and their employees are expected to convey objectives, intentions, and avowedly good deeds (i.e., including their various CSR activities), nor merely the process by which organizational CSR strategies are enacted. Rather, employee communication is a continuous practice “through which social actors explore, construct, negotiate, and modify what it means to be a socially responsible organization” (Christensen and Cheney 2011, p. 491).

In summary, no matter how CSR is conceived, communication among employers, employees, and other stakeholder groups is a fundamental mechanism through which CSR is constituted. The business case for CSR, embodied in the first generation, relies on employee trust, support, and endorsement of organizational CSR efforts to relevant stakeholder groups (Collier and Esteban 2007), as well as the business-generating aspect of employee communication. The social contract case for CSR (i.e., second generation) is grounded in employees’ rights for free expression, information sharing, and concerted action. Finally, the civil society case for CSR (i.e., third generation) includes dialogue, stakeholder participation, and transparency in developing organizational norms and regulations that shape CSR. Taken together, there are five central communicative tenets embedded in contemporary notions of CSR: (a) freedom of speech; (b) collective information sharing; (c) respecting differences; (d) engagement and stakeholder dialogue; and (e) transparency. Table 1 presents foundational principles, sources, and examples of these five communicative tenets.

Table 1 Five communicative tenets of 2nd and 3rd generation corporate social responsibility

In the following section, we describe the methods used to analyze the ways in which social media guidelines reinforce or delegitimize CSR’s first, second, and third generation’s communicative tenets through (a) framing dynamics; (b) boundary specifications; and (c) speech regulations.

Methods

Sample

To explore how social media policies reflect and potentially shape global trends, attitudes, and communicative values regarding corporate social responsibility, we examined the publicly available policies of the most powerful and influential corporations in the world. These include the 250 companies with the world’s highest market values, obtained from the Financial Times ( 2014) Global 500 list and the 250 companies with the world’s highest revenues, obtained from Fortune (2014) Global 500 list. By virtue of their global reach, network complexity, large numbers of employees, diverse customer base, and a significant public presence, these companies are on the front line confronting the opportunities and challenges associated with the burgeoning use of social media. What these companies do and how they react to the changing global context not only has powerful implications for work life but also toward shaping normative institutional expectations regarding employee communication.

Because there is overlap between the top 250 companies on each list, there were a total of 348 unique companies in the population. For each of the companies, a web search with the key terms “(company name) social media guideline,” “(company name) social media policy,” and “(company name) social media manual” was conducted. Additionally, using the key terms “(company name) code of conduct,” “(company name) code of employee conduct, and “(company name) code of ethics,” a second search allowed us to find social media guidelines which were embedded in other company policies and/or codes of conduct. The searches resulted in 112 publicly available social media guidelines—about one-third of the original 348 companies. It is likely that this sample underestimates the actual number of top global companies that have policies (i.e., there are likely company policies that exist but are not publicly available online), but this percentage is within the range reported by the global surveys cited above. A majority of the guidelines in our sample (58.93 %) are embedded in the company’s code of conduct, and 41.07 % are stand-alone guidelines. The oldest guidelines date back to the year 2008, 45 were published between 2008 and 2012, and 67 came online between 2013 and 2014. Practically all guidelines (95.54 %) apply to all social media channels, 2.68 % apply only to Facebook, and less than 1 % apply only to blogs and micro-blogs (e.g., Twitter). On average, a social media guideline contains 745 words, but varies widely from a short, 18-word policy to a 5089-word policy.

Table 2 contains the breakdown of the sample by region and corporate sector. A majority of the companies are headquartered in the United States (53.57 %), followed by Western Europe (20.54 %) and the Commonwealth countries (17.86 %). Asian companies comprise only 6.25 % of the sample and less than 2 % are from other regions. The regional distribution of guidelines closely resembles the distribution of the population of 348 companies, with the notable exception of our sample having fewer companies that are headquartered in France and Asian locations. Interestingly, where regional representation is disproportionate, the discrepancies parallel the findings of studies of codes of ethics done in the 1990’s and early 2000s. In these studies, 68–90 % of US corporations were found to have codes of ethics, whereas only 30–45 % of French companies and 10–38 % of Japanese companies did (see Stohl et al. 2009). These distributive differences were accounted for in two ways that have relevance for our sample. First, as Kaptein (2004) noted, “codes are often unavailable via the intranet or in English (despite the fact that these companies maintain business relations in English-speaking countries)” (p. 27). Our sample includes only those policies that are available online and in English, so it is not surprising that French and Japanese companies are less represented in the sample. Second, American companies were often found to adopt codes of ethics much earlier than corporations headquartered in other countries, particularly in Asia (Schwartz 2002). We are still in the early stages of social media use and corporate policies and it is likely that over time more and more companies throughout the world’s region will face increasing institutional and public pressures to develop these policies and make them publicly available.

Table 2 Breakdown of Companies by Sector and Region

Table 3 Embedded versus stand-alone policies

Table 4 Frequencies of three primary rationales/frames (RQ 1)

Table 5 Frequencies of five communicative tenets (RQ 2)

Table 6 Frequencies of boundary specifications (RQ 3)

With regard to sectors, whereas the pharmaceutical sector—one that is subject to very strict communication regulations in many parts of the world—is overrepresented in the sample (17.86 % of our sample compared to 8.33 % of the 348 top global companies), the energy sector is underrepresented (9.82 % in sample compared to 19.83 % in the population). Financial services have 22.32 % of the companies in the sample, more than any other industry and healthcare equipment, as it is in the entire population has the fewest number of companies, only 2.63 %. The number of employees in these companies varies widely, ranging from 5000 to 2.2 million employees worldwide. Overall, these 112 guidelines regulate the communication of more than 15.1 million employees worldwide.

Development of Measures and Coding Procedures

Coding categories and measures were developed in a mixed inductive-deductive procedure, based upon CSR theoretical frameworks, a review of the empirical literature, and repeated readings and discussions of a random sample of 20 guidelines. Codes were created to represent the three central constructs of the study: (a) rationalization and frames of social media policies; (b) CSR communication tenets addressed in the policies; and (c) the scope of boundary specifications in the policies. From the literature, three types of rationales/frames for policies were identified and coded: business interests, risk protection, and co-creation. The communication principles identified in the UN Global Compact, the Universal Declaration of Human Rights, the Organization for Economic Co-operation and Development corporate guidelines, and CSR theory formed the basis for the measurement of the positive affirmation or direct denial of CSR communication tenets (i.e., the presence or negation of freedom of speech, engagement of stakeholders, collective information sharing, respecting differences, and communicative transparency). Boundary specifications and scope were operationalized as the degree to which policies indicated permeable as opposed to discrete boundaries. Boundaries were defined in terms of private–public, personal–professional, materiality, space, and time (see Mesch and Talmud 2007). Table 7 in Appendix provides definitions and examples of each code used in the study.

Table 7 Definitions and examples of coding scheme

Coding was refined across several meetings of the authors until the coding procedures were considered robust. In March 2014, four coders were trained using a detailed 19-page coding manual. After the training, coders independently coded five randomly selected guidelines. Given the diversity of policy formats, the unit of analysis for initial coding was a self-contained section, that is, a paragraph or bullet point. Any one section could incorporate multiple codes; initial codes were not mutually exclusive. The policies had on average 13.75 sections, with a range from 1 to 70 sections. Essentially, the coding process involved marking each section for the codes that were present and absent. The first coding round yielded unreliable results; percentage of agreement ranged only from 40 to 80 % agreement. Thus, three more coding rounds were conducted. Each included coding five policies and subsequent discussion and analysis of disagreements as well as refinement of the categories. In the final training session, each analyst coded 10 guidelines and inter-coder reliability of 0.89 was achieved (Hayes and Krippendorff 2007). This was acceptable and each coder then proceeded independently to code a subset of the policies.

Research Questions and Statistical Analyses

To address our overarching research question (i.e., the extent to which organizations’ social media policies embody the sensibilities and communicative values associated with second and third-generation CSR or represent first CSR generation ideas), three specific questions guide the analysis. Given that previous research on the institutionalization of corporate codes of ethics suggests both global content convergence (O’Dwyer and Madden 2006) and regional and sectorial differences (e.g., the history of industrial and union relations, stakeholder activism, and regulatory environments of a region are associated with different foci in codes of ethics; see Stohl et al. 2009), regional and sectorial differences are examined for each question.

• RQ 1. What are the rationales organizations use to develop and frame their social media guidelines?

  • RQ 1a. What is the relationship between industrial sector and rationale?

  • RQ 1b. What is the relationship between regional location and rationale?

• RQ 2. To what extent do social media policies reflect the five communicative tenets of contemporary CSR (i.e., freedom of speech, engagement with stakeholders, collective information sharing, respecting differences, and communicative transparency)?

  • RQ 2a. What is the relationship between industrial sector and communicative tenets?

  • RQ 2b. What is the relationship between regional location and communicative tenets?

• RQ 3. To what extent do social media guidelines indicate permeable and/or distinctive boundary specifications (with regard to time, space, and material along the dimensions of private–public and personal–professional)?

  • RQ 3a. What is the relationship between industrial sector and boundary specifications?

  • RQ 3b. What is the relationship between regional location and boundary specifications?

Chi square analyses were used to assess whether company policies containing each frame, communicative tenet, and boundary specification were similarly distributed across regions and sectors. For these tests, companies served as the units of analysis, and each was categorized in terms of its regional location and sectorial affiliation (respectively). Therefore, the cells within the contingency table were mutually exclusive; companies belonged to one region and one sector and were analyzed based on whether their policies contained or did not contain a particular variable (i.e., no = 0, yes = 1). Analyses were run separately for each of our variables concerned with frames, communicative tenets (including separate positive and negative variables for free speech, collective sharing, and stakeholder engagement), and boundary specifications. In cases wherein small, sparse, or unbalanced data across regions and sectors resulted in violations of the Chi square sample size assumption, exact P values and confidence intervals were computed based on the permutational distribution of the test statistic (i.e., exact inference for categorical data; Mehta and Patel 1998).

We also ran Chi square analyses (using Fisher’s Exact test, when appropriate) to examine two additional dimensions on which policies could systematically and significantly differ: embeddedness (i.e., whether the policy was a “stand-alone” set of guidelines or a subset embedded in a larger corporate code of ethics or employee code of conduct) and time (i.e., when the policy was made public). Table 3 presents the percentages of embedded versus stand-alone policies by region and sector, as well as the percentage of each that include the frames and rationales, communicative tenets, and boundary specifications addressed in this study.

Results

Policy Rationales and Framing Social Media

The first research question addresses the types of frames and rationales companies use for explaining the purpose and goals of their social media policies. Companies may include more than one type of rationale or directive. Thus, percentages indicating the number of companies by category (e.g., business case, risk, and co-creation) may be greater than 100 %. Table 4 presents the distributions across regions and sectors. Not surprisingly, 70.54 % of the policies contain explicit rationale statements explaining the business case for the policy, framing social media use as a business tool. For example, Coca Cola suggests that the purpose of social media is brand and reputation management, and within their policy, they tell employees to “represent our company and share the optimistic and positive spirits of our brands.” Almost 60 % of the rationales (58.93 %) include the need to protect the company from the risks of social media. Roche tells employees to take heed, as “certain risks are associated with these new channels.” In contrast, only one-third (33.04 %) of the policies reference or invoke a third-generation CSR co-creation frame as the rationale behind social media use and policies. For example, Linde, a large global energy company, tells its employees in its social media policy, “we need to be ready to engage actively with the democratic consumer.”

Across regions, no significant differences in the presence of third-generation rationales were found, χ² (3, N = 110) = 3.07, ns. Despite the statistically insignificant differences however, the percentages for each region are notable. Whereas 57.14 % of Asian companies’ policies contained at least one direct reference to a co-creation frame or rationale, only 39.13 % of Western European companies, 30 % of US companies, and 25 % of Commonwealth organizations referenced co-creation. These results stand in contrast to previous studies of corporate codes of ethics, which suggest that regions and nations with a strong history of employee rights and participation (e.g., Western European countries and Japan) are more likely to address second- and third-generation CSR concerns within their regulations (Carasco and Singh 2003; Stohl et al. 2007).

Through Chi square analyses, we investigated whether stand-alone policies were different from those embedded in corporate codes of ethics or employee codes of conduct (see Table 3). Although embeddedness was not related to the presence of business or risk frames, it was associated with the degree to which co-creation rationales and frames were present. Specifically, embedded social media guidelines were less likely to be framed and provide rationales in terms of dialogic and democratic engagement (i.e., tenets of the third generation of CSR), χ² (1, N = 112) = 23.23, p < .001.

Examining when the guidelines were published helped to further unpack the regional findings. A previous study by Vaast and Kaganer (2013) found that the business frame for social media had become “more perceptible over time” (p. 930). Thus, we divided our sample into two groups: (a) those guidelines that were made public between 2008 and 2012 (n = 45); and (b) those guidelines approved and released in 2013–2014 (n = 67). This temporal division was based on the burgeoning global discussions and court rulings regarding social media policies after 2011. In the US, for example, the National Labor Relations Board (NLRB) issued its first ruling on social media policies in late 2011 stating that Hispanics United of Buffalo—a non-profit services agency—violated the National Labor Relations Act (NLRA) when it fired employees for posts on Facebook complaining about their jobs (Robertson 2011). In 2012, the NLRB, after hearing more than 30 related cases, issued an extended summary statement indicating that social media policies that “reasonably chill” employees in the exercise of their rights under the NLRA are unlawful (White et al. 2012).

At the same time, amid increasing debates regarding regulating social media in the European Union and Commonwealth countries, the EU High Representative responded by stating that “Freedom of expression as enshrined in the Universal Declaration of Human Rights entails everyone’s right to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers” (Ashton 2012). However, since 2012, there have also been several employment case decisions across the globe that muddy an already ambiguous context of what employee speech is actually protected in different jurisdictions and the participatory role of social media in the commercial sector. For example, in late 2012, United Kingdom Tribunals indicated that employee remarks on social media sites are not necessarily protected by the provisions on freedom of expression under Article 10 of the European Convention on Human Rights (McGuire Woods Consulting 2012). Thus, we explored whether such a volatile context had any effect on the rationales as well as the content of the social media policies.

Our results show that across regions and time, a majority of companies had both risk and business rationales and frames for social media usage and regulation (see Fig. 1). In the case of the business frame, the percentage of US and Asian companies remained relatively stable over time, while the percentage of Commonwealth and Western European companies that included a business frame decreased. Inclusion of risk frames remained relatively stable for US companies and increased slightly for Commonwealth and Western European companies, with the greatest change in Asian companies where all companies that published guidelines after 2012 addressed risk. At the same time, the percentage of Western European companies that addressed co-creation increased from 24 to 50 %, whereas in the other regions, co-creation frames were far less prevalent after 2012.

Fig. 1

Percentage of companies by rationale (RQ 1) over time

Five Communicative Tenets of Third-Generation CSR

The second research question addresses the extent to which social media policies contain specific behavioral references and directives regarding the five tenets of second and third generation CSR. Table 5 includes the distributions across regions and sectors for all companies.

Freedom of Speech

Results indicate that 83 % of companies in our sample address freedom of speech in some way. However, only 22.32 % of companies positively affirm employees’ rights to express their own opinions and viewpoints, whereas 82.14 % contain at least one guideline prohibiting employee expression. For example, 11.61 % of companies’ guidelines encourage employees to be true to themselves and a mere 4.46 % of the policies explicitly ask employees to say what they believe. Cisco includes a highly unusual directive, noting that an employee’s “internet posting should reflect [his or her] personal point of view, not necessarily the point of view of Cisco.” Similarly, General Motor’s (GM) policy acknowledges that “Employees may wish to participate in various forms of social media on an individual basis and communicate their personal opinions and viewpoints concerning GM and/or GM products.”

On the other hand, 34.82 % of companies require employees to align their communication with the interests of the organization. Altria mandates that employees “be careful not to discuss topics in a way that [is] inconsistent with [the] Company’s positions and actions.” Similarly, Priceline reminds employees to “be aware that taking public positions online that are counter to the Company’s interest may cause conflict and can have disciplinary repercussions.” Furthermore, 35.71 % of policies prohibit employees from criticizing the organization, managers, fellow employees, customers, competitors, and/or suppliers and partners. For instance, National Australian Bank’s (NAB) guidelines feature the following directive: “Ensure your opinion doesn’t cause any damage to NAB, even if it is your personal opinion.” Similarly, Common Wealth Bank instructs employees not to “display or post any information that may damage the group in any way.”

Although no significant regional or sectorial differences appear, there are interesting trends among sectors. In particular, the technology and industrial sectors are more likely to acknowledge employees’ free speech rights than other sectors, accounting for 52 % of the 25 companies that acknowledge such rights. These sectors are unique insofar as companies in the technology sector (e.g., Apple, Cisco, and Microsoft) have been under enormous pressure to protect the privacy rights of their stakeholders, and companies in the industrial sector (e.g., Ford, Daimler, and General Motors) have large segments of unionized employees.

Engagement with Stakeholders

The second communication tenet concerns the value and importance of employee dialogue with diverse stakeholders, which is addressed by half of the companies in our sample. Results indicate that only 10.71 % of companies explicitly encourage employees to engage with stakeholders through their social media policies, although the business case for CSR can be grounded in the importance for employees to exchange information with consumers across the marketplace. For instance, IBM states, “We believe in the importance of open exchange between IBM and its clients, and among the many constituents of the emerging business and societal ecosystem.” In contrast, almost half of the companies (46.43 %) discourage employees from entering into dialogue with stakeholders—especially those connected to news media. United Technology’s policy warns employees not to talk with anyone about the organization in social media and rules: “Do not answer questions or make statements about or on behalf of UTC, its products, services or programs without explicit prior authorization from UTC communications.” Sanofi-Aventis’ policy tells employees not to “comment or discuss content posted by other Sanofi-Aventis employees and partners.” There are no significant regional or sectorial differences with regard to stakeholder engagement.

Collective Information Sharing

Analyses reveal that 39.79 % of companies address collective information sharing in some way. More specifically, 15.18 % of companies encourage employees to share their work experiences on social media. Most notably, these companies’ policies frame social media use as a good opportunity to participate in important conversations and raise their voices. Dow Chemical notes, “Employees have a legally protected right to freely discuss their wages, benefits and terms and conditions of employment.” In contrast, a fourth (25 %) of companies’ guidelines deter employees from collective information sharing. UPS states, “Publicizing your concerns through social media is not the most effective manner to get issues resolved.” Fifteen percent of guidelines tell employees to avoid talking about work at all when using social media. For instance, AbbVie mandates that employees avoid “post[ing] information about any AbbVie product, service, or other AbbVie business activity unless authorized to do so.” There were no significant regional or sectorial differences among companies with regard to collective information sharing.

Respecting Differences

Across sectors and regions, policies remind employees to respect others (48.21 %). Whereas 40.28 % of companies ask employees to “be respectful” in general, only 9.82 % specifically encourage employees to “respect cultural differences.” Apple’s policy is typical: “Respect your audience and coworkers; remember that Apple is a global organization whose employees and customers reflect a diverse set of customs, values, and points of view.” There are no statistically significant sectorial or regional differences.

Communicative Transparency

Analyses reveal that 60.71 % of companies explicitly encourage their employees to communicate transparently, honestly, fairly, and ethically in their social media communication. Specifically, 41.96 % of companies ask employees to “stick to the facts” and always communicate truthful and accurate information. Intel notes, “Honesty—and dishonesty—will be quickly noticed in the social media environment,” and thus encourages its employees to “be truthful.” Furthermore, the importance of transparency was mentioned in 40.18 % of companies’ guidelines. For instance, Honda Motors directs its employees to: “Be transparent. If you have something to say, use your real name.” Moreover, 33.93 % of companies require employees to identify themselves as such when talking on behalf of the company. Again here, there are no significant sectorial or regional differences.

Embedded Guidelines and Development of Communicative Tenets Over Time

Given the results above—especially the relative lack of expected sectorial and regional differences—we explored the degree to which embeddedness of policies (see Table 3) and year of issue made a difference in terms of communication tenets. For embedded versus stand-alone policies, Chi square analyses indicate significant differences across all communicative tenets of contemporary CSR. In all cases, independent policies are more likely to include guidelines related to the communicative tenets than embedded policies: (a) free speech, χ² (1, N = 112) = 15.95, p < 0.001; (b) collective sharing, χ² (1, N = 112) = 5.44, p = 0.02; (c) engaging stakeholders, χ² (1, N = 112) = 15.99, p < 0.001; (d) respecting differences, χ² (1, N = 112) = 32.46, p < 0.001; and (e) transparency, χ² (1, N = 112) = 26.43, p < 0.001. This is not surprising when we consider the powerful regulatory and normative context in which codes of conduct are developed. When social media are embedded within these codes, they are less likely to address the ambiguous social arena in which regulation and normative expectations are still in flux and more likely to explicate legal constraints.

Results for differences between the two time periods (2008–2012 and 2013–2014; see Fig. 2) indicate that newer policies are significantly less likely than older policies to contain directives related to stakeholder engagement, χ² (1, N = 112) = 9.28, p = 0.002. Furthermore, when examining these tenets over time and with respect to regions, we find that newer policies in Western European companies are more likely than those in other regions to encourage collective information sharing, χ² (3, N = 66) = 9.45, p = 0.02. Although statistically insignificant, this trend is also evident in the results for free speech; whereas companies in the US, Commonwealth, and Asia have sharply declined in their encouragement of free speech, the tendency for Western European companies to encourage free speech and expression has increased over time. Overall, however, results for temporal differences suggest that second- and third-generation CSR issues are less pronounced in today’s policies than they were in policies developed in 2008–2012.

Fig. 2

Percentage of companies by communicative tenet (RQ 2) over time

Boundary Specifications

Our third research question concerns boundary specifications (see Table 6 for frequencies), which are addressed by 75.89 % of companies in our sample. Boundaries define the limits of corporate control within the policies and help construct employee identities. Results indicate that over two-thirds (70.54 %) of companies attempt to expand their influence over employees beyond the corporate sphere by explicitly blurring the boundaries between corporate, private, and public communication in their guidelines. For example, Novo Nordisk’s policy states:

In online social networks, the lines between public and private, personal and professional are blurred. Just by identifying yourself as a Novo Nordisk employee, you are creating perceptions about your expertise and about Novo Nordisk with our shareholders, customers, and the general public—and perceptions about you by your colleagues and managers.

Interestingly, before this particular section, the policy states that Novo Nordisk employees are required to identify themselves as such; thus, at all times, their online identities are that of “employee” and their communication is universally covered by the regulations in the policy.

In 41.07 % of policies, boundaries are blurred in terms of personal versus professional contexts. As an example, Merck cautions its employees, “While we respect the rights of our employees to engage in personal online activities, you are still responsible for any damage or harm to our business or reputation that results from your online activities.” About the same number of guidelines (48.21 %) frames the issue in terms of private versus public contexts. Siemen’s acknowledges, “there is often no clear distinction in social media between public and private or between business and personal.” On the other hand, 16.07 % of companies recognize that employees have distinct identities (i.e., separate work, personal, and/or public identities). Zurich’s policy states, “It is not Zurich’s intention to tell you how you should engage in the personal use of Social Media.”

Given the overwhelming tendency across policies to collapse public–private and personal–professional contexts, we examined three specific dimensions through which companies achieve context collapse in their social media policies: spatial, temporal, and material. Spatial boundaries concern where employee social media communications occur (e.g., in or outside of the workplace, public or private places, etc.). Results indicate that one in four (26.79 %) policies apply across all spaces and places, compared to only 2.68 % that explicitly apply only in work spaces. For instance, United Health’s guidelines “apply both to the use of social media as part of [employees’] work on behalf of UnitedHealth Group and to [their] social media activity outside of the company.” Temporal boundaries relate to when employee social media communications occur (e.g., on or off the clock; before, during, or after work hours, etc.). Whereas only 3.57 % of policies state that they only apply during work hours, nearly one in five company policies (19.65 %) apply at all times and regardless of the context. For example, Siemen’s policy includes the following warning: “Depending on the totality of the circumstances, an employee’s social media activities—even outside of work during nonworking time—could violate Siemens’ policies.” Finally, material boundaries concern what mediums or devices employees use to communicate (e.g., work-issued vs. personally owned technologies, etc.). Such boundaries are rendered permeable in 22.32 % of policies, which apply to all materials across all contexts, regardless of the content discussed. For instance, General Electric states that “Any blogging or posting that violates any GE policy, including the Spirit & The Letter and these Guidelines, even when done with personal resources, is prohibited.” In contrast, 11.61 % of companies explicitly state that their guidelines only apply to materials and technologies that are owned and/or issued by the corporation.

Analyses revealed no significant regional or sectorial differences for boundary specifications, so we examined whether policies differed in terms of embeddedness and development over time. Consistent with our earlier findings, results indicated that embedded guidelines were less likely than stand-alone policies to discuss distinct, χ² (1, N = 112) = 4.23, p = 0.04, and/or permeable boundaries, χ² (1, N = 112) = 7.62, p = 0.006. On the other hand, there were no significant differences in boundary specifications over time (see Fig. 3). Notably, however, whereas US and Asian guidelines show practically no change over time in terms of permeable boundaries, the percent of guidelines from these same regions that included distinct boundaries diminished between the two time periods. Changes over time for Western European companies suggest the opposite trend. While the percentage of Western European companies’ guidelines that blur boundaries has decreased since 2012, the percent acknowledging distinct boundaries has increased. For Commonwealth companies, issues related to both distinct and permeable boundaries appear to garner less attention in newer policies than in older ones.

Fig. 3

Percentage of companies by boundary specification (RQ 3) and region over time

Results Across Research Questions

Given the propensity of corporations to stipulate how and what employees should communicate through social media—not only in their professional roles, but also in their personal lives—we ran correlation analyses to better understand the relationships among policy rationales, the communicative tenets of contemporary CSR, and the provocative boundary specifications addressed in this study. Results indicate several interesting trends. First, although companies citing third-generation co-creation frames are generally encouraging in terms of the five communicative tenets, co-creation also correlates positively with the limitation and/or restriction of free speech (r = 0.23, p = 0.02). Complicating matters further, the same companies that discourage free speech and expression are also those that mandate communicative transparency in their policies (r = 0.29, p = 0.002). Such competing directives have the potential to spark tensions for employees, as they simultaneously attempt to “be true to themselves” and “align communication with the organization’s brand, vision, standards, and values.”

Results also suggest, however ironically, that policies appearing to focus on risk (i.e., by invoking a risk frame/rationale) tend not to explicitly discourage free speech (r = 0.18, ns), collective sharing (r = 0.02, ns), or stakeholder engagement (r = 0.12, ns). However, the risk frame has a strong positive correlation with permeable boundary specifications (r = 0.26, p = 0.006), which strongly coincide with negative speech regulations (r = 0.36, p < 0.001) and directives that limit stakeholder engagement (r = 0.33, p < 0.001). Thus, while it is atypical for policies framed in terms of risk to explicitly limit communicative tenets of contemporary CSR, they may do so implicitly through strategically ambiguous boundary specifications. Indeed, companies that promote permeable boundaries most often also acknowledge distinct employee identities (e.g., private, corporate, public, etc.) in their policies (r = 0.25, p = 0.009), suggesting yet another set of competing directives. These vague and contradictory mandates seem especially consequential for employees, given that 20.54 % of policies mention corrective and/or disciplinary action as a punishment for violating social media guidelines.

Discussion

Whereas most CSR communication research focuses primarily on external constituents (e.g., Castelló et al. 2013; Etter 2014; Lyon and Montgomery 2013), we have drawn attention to employees as an important stakeholder group for the development, enactment, and assessment of socially responsible behavior. As Christensen and Cheney (2011) suggest:

Communication is not simply a mechanism through which organizations and their employees are expected to convey their objectives, intentions, and avowedly good deeds, including their various CSR activities, nor merely the process by which organizational CSR strategies are enacted [but rather] employee communication is a continuous practice through which social actors explore, construct, negotiate, and modify what it means to be a socially responsible organization. (p. 249)

Accordingly, we have argued that the increased use of social media and the proliferation of social media policies regulating employee communication instantiate an emergent model of CSR. Our findings suggest that, to a great extent, organizations’ social media guidelines hinder rather than enable employees’ free speech rights, collective information sharing, and stakeholder engagement and dialogue, thereby constraining the central mechanisms through which contemporary CSR may be established and maintained.

Furthermore, the blurring of traditional boundaries among corporate, private, and public spheres in most guidelines suggests that as social media are becoming more dominant, so too is the expansion of organizational control beyond employees’ organizational roles. The disconnect between individual perceptions of social media as a mode of interpersonal and social interaction (e.g., Papacharissi 2012) and employers’ perceptions of social media as a business channel (e.g., Kaplan and Haenlein 2010) plays itself out in conflicting and constraining expectations. Identity and individual value are defined within these policies by organizational membership. Even in their capacities as private individuals, social communication is organization-centric, seen as either creating or diminishing organizational value, protecting or hurting the organization. When individuals fail to comply with the policy mandates in their private lives, negative organizational consequences are not uncommon. Almost a quarter of the guidelines allude to corrective and disciplinary actions as punishment for violating the social media policy and 17 % explicitly threaten to terminate employees whose social media use breaches company policies. Several guidelines also demand that employees monitor fellow employees’ social media use in and outside the work context and inform management about others’ violations of guidelines.

Our results indicate that many social media guidelines exacerbate already existing boundary tensions within the CSR domain. On the one hand, scholars and human rights advocates have long argued that an employee has a right to liberty, and flowing from liberty is the right to lead a life that is separate from work. This freedom is required for private thoughts and development of one’s self apart from his or her work identity (Manning 1997). It is posited that there is “a natural right of humans to have a personal space” (Clark and Roberts 2010, p. 518). Article 19 of Declaration of Human Rights states that “Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers” (para. 19). The United Nations (2015a) Global Compact asks companies “to embrace, support, and enact within their sphere of influence a set of core values in the areas of human rights, labor standards, the environment and anti-corruption” (para. 2).

On the other hand, others have consistently argued that an employer has a right to know whatever it can about a person to protect its property rights in the business and that an employee is a direct representative of the company at all times, both on and off the clock (Myatt 2009). For example, recent legal rulings and precedents point to an emerging global consensus around the acceptability of corporations obtaining information about present and future employees on social media through “background checks” (Clark and Roberts 2010, p. 508). Further, employers have established the right to monitor social media activity and dismiss employees for comments made on social media (Proskauer 2014). The Australian Fair Work Commission also recently maintained that it is not “harsh, unjust or unreasonable” for social media policies to operate outside as well as inside the workplace (Priestley and Guilleaume 2014).

Legal cases and studies about social media affordances have also established that boundaries between employees’ work and private lives are not only more permeable with the development of social media, but also make it more likely that “employees will modify their behavior for fear of being judged by their employer” (Clark and Roberts 2010, p. 518). However, business ethicists and CSR scholars have yet to analyze the impact of these critical boundaries on CSR. As our study indicates, these authoritative texts create tensions within employer–employee–community relations, which counter many of the agreed-upon CSR standards and expectations contained in the UN Global Compact and other international agreements.

Complementing recent calls for research on the public negotiation of responsibilities and expectations toward organizations (Castelló et al. 2013; Christensen et al. 2015) and the burgeoning literature that critically investigates the promising potential of social media for deliberation in the new public sphere (Whelan et al. 2013), we find that social media guidelines not only have the potential to compromise employees’ rights, but also their opportunities to influence and participate in the evolution of CSR. Whereas social media have been heralded as making a plurality of voices and opinions visible (Friedland et al. 2006) and “potentially heard” (Castelló et al. 2013, p. 685), this study supports the increasing number of scholars who question social media as powerful “deliberative tools” (Seele and Locke 2014, p. 13) influencing public discourse in relation to CSR issues (e.g., Etter and Vestergaard 2015).

Increasingly, both research and the popular press have drawn attention to the power of large ICT companies, such as Facebook, that dictate the conditions under which deliberation in social media takes place (e.g., Whelan et al. 2013). Indeed, recent incidents have shown how the configuration of power by these corporations (e.g., Facebook, Google, Microsoft) can enable censorship and silence certain voices. Overall, this study suggests that the promising affordances of social media for the inclusion of various voices in the public negotiation of the expectations and responsibilities of organizations tend to be compromised—not just through the activities of the largest information technology organizations like Google or China Mobile, but through social media guidelines of global corporations across economic sectors. In many ways, these findings support widely held assumptions about organizations distorting and/or closing particular discourses (e.g., Deetz 2007; Kuhn and Deetz 2008). We suggest that social media policies are understudied “mechanisms of closure” (Christensen et al. 2015, p. 142), through which managers and corporations limit their members’ communicative freedoms and intrude upon their private and public lives in ways that can have significant impact on how CSR evolves in the workplace. When intruding upon the private communicative sphere of employees, corporations hinder individuals’ capacities to enact their interests as citizens in the online “public arenas of citizenship” (Whelan et al. 2013, p. 780).

Our findings suggest that social media use is most often conceived within a business or risk frame and that more than two-thirds of companies’ policies violate internationally agreed upon CSR communicative norms to some degree. However, the picture is not completely bleak. Our findings also suggest that almost a third of social media guidelines embrace—at least to some extent—the affordances of social media from a third-generation CSR perspective. In these cases, employees are explicitly encouraged to express their own opinions and to engage and share information with various internal and external stakeholders. Thus, employees’ communicative rights are respected and their voices are valued in the ongoing public negotiation of corporate responsibilities. At the local organizational level, these guidelines may further enhance and facilitate the creation of an online “corporate arena” (Whelan et al. 2013, p. 781), wherein employees who use social media are entitled to participate, and even encouraged to discuss, debate, negotiate, and organize CSR issues.

Moreover, although corporations have always exercised their hegemony over workers with great control (Cochran 1972) and the rise of new information and communication technologies has increased corporations’ “panoptic power” (Brown 1996, p. 1238), there are some guidelines that explicitly recognize the private sphere and make a distinction between individuals’ communication in their capacities as employees versus their roles as private citizens. Such guidelines align with Article 8 of the Human Rights Act of 1998, which provides a person with “the right to have one’s private life respected.”

In other words, although the majority of guidelines minimize and/or restrict employee voice, some guidelines do frame social media from a co-creation perspective, where social media are viewed as a public forum for transparency, employee dialogue, deliberation, and community rights. In a global context where norms and regulations regarding employees’ social media use are ambiguous and evolving, those guidelines that clearly and explicitly empower employees may pave the way for other organizations to view social media as deliberative tools. Potentially, corporations that instantiate the communicative tenets of CSR are fostering the possibility of enacting a new generation of CSR (Seele and Locke 2014).

Overall, the empirical evidence regarding the institutionalization of CSR communication tenets within social media policies and the larger context in which these policies are being developed is mixed. As suggested above, normative expectations regarding the scope of organizational control are expanding throughout the world, but in some venues, the protection of workers’ rights on social media is being recognized. The landmark cases in 2012 seem to have affected the prevalent framing of social media guidelines such that avoiding risk rather than improving business has become the predominant policy rationale. However, a recent report also notes that although the legal position on social media guidelines is still in flux across the globe, recent NLRB rulings give broader latitude to employees, viewing online exchanges more like a water cooler conversation among employees than public broadcasts to actual or potential customers (Proskauer 2014). Specifically, the NLRB ruled in 2014 that “the rights of employees to act together to address conditions at work, with or without a union, are protected,” and that “this protection extends to certain work-related conversations conducted on social media, such as Facebook and Twitter.” Citizens’ reactions to these policies are also having an effect. Public outrage regarding the Times of India social media policy disaster, in which they demanded control of all employee social media accounts (i.e., both personal and professional) even after employees had left the company, resulted in a modification of the policy—one that assured a minimal level of communication rights but maintained other extreme regulations (Sruthijith 2014). Whether these kinds of public outrage, continuing discussions, and demands for public accountability regarding CSR will influence and be influenced by these policies remains to be seen.

We suggest that as organizations continue to develop and refine their social media policies in response to increased vigilance and activism by stakeholders, prevalent use of CSR consultants, and new rulings and regulations by (inter)national governance and legal institutions (DiMaggio and Powell 1983), it will be important to track if the acceptance of third-generation communication rights will become more engrained within social media policies. Moreover, the tensions between competing views of employee and employer communication rights and responsibilities need further consideration. Thus, future research should look into the inherent contradictions, paradoxes, and communicative ambiguities in corporate policies, as well as how employees navigate them. As social media guidelines become a ubiquitous part of the communication landscape, they will also become an inherent part of our conceptions of what constitutes a socially responsible corporation.

Appendix

See Table 7.