Reporting within the Corridor of Conformance: Managerial Perspectives on Work Environment Disclosures in Corporate Social Responsibility Reporting

: This paper examines managerial perspectives on work environment (WE) disclosures in corporate social responsibility (CSR) reports. WE encompasses all aspects of the design and management of the work system that a ﬀ ect the employees’ interactions with the workplace. The data are drawn from interviews with 20 CSR managers in large companies that are recognized as high performers in CSR. Managers reported that WE disclosures were an important part of CSR reporting and had several beneﬁts—for example, it helped to maintain companies’ reputations as good places to work, which were of interest to both investors and potential employees. However, WE reporting was at a low level, focusing predominantly on occupational health and safety performance indicators, such as the number of employee injuries per year. We suggest that organizations derive legitimacy by reporting WE disclosures within a corridor of conformance that permits a low level of reporting and a great deal of latitude regarding what topics and how topics are disclosed, provided organizations meet institutional expectations. The corridor is maintained by an institutional environment in which, from participants’ perspectives, few external stakeholders were interested in WE disclosures representing employee well-being (e.g., psychological health) and the prevailing CSR reporting standards and guidelines provide scant information about employee health. come up. What usually does come up is environmental issues . . . donations and what we’re doing in the community, but not so much on the work environment piece, and that’s also true when it comes to external stakeholder engagement that we do with our customers. Because we also ask them to rank the issues that are important to them, like with what we should be doing, and how we treat our employees and health and safety is always part of that, and it’s never up on the top. It’s usually close to the bottom” (#8).


Introduction
Work-related injuries and fatalities continue to have a staggering effect on employees. Globally, in 2015, there were an estimated 2.78 million work-attributable deaths [1]. The costs of workplace injury and illness to national economies are enormous [2] and represent substantial threats to performance and profits in the business world [3,4].
There have been calls to improve the health and well-being of employees through Corporate Social Responsibility (CSR) [5][6][7] but relative to other areas of CSR, such as the natural environment, there is a lack of knowledge about how employee concerns are addressed in CSR reporting. Companies are conforming to the institution of CSR insofar as they have adopted reporting protocols and are disclosing information. Within their CSR reports, companies are attending to work environment (WE) topics; however, the literature suggests much of this reporting is low quality. Organizations' WE reporting is incomplete in different but important senses. First, there is a significant incongruity regarding what is reported about employee well-being and the actual practices and occurrences within concerning CSR reporting, an area that has received little attention [25]. The paper goes beyond analyses of publicly available disclosures in CSR reports, which are extremely valuable, but rarely shed light on the process underlying the decisions about what and how to disclose. Examining the views of CSR managers regarding WE disclosures permits a study of the WE elements that are deemed relevant by managers, why these are deemed relevant, and why they are subsequently disclosed through CSR reporting. The rationale for these CSR reporting decisions remains relatively understudied [26]. Finally, the paper contributes to the discussion about WE in CSR reporting -a noted gap in the CSR literature [9,12,27,28].
The paper begins with a review of the literature that pertains to CSR reporting and includes an overview of institutional theory, which guides the study, and a discussion of the literature from which the research questions emerge. The paper then describes the sample and data collection methods that were used in the study. Following this, the findings are presented. This is followed by the discussion in which the findings of the study are reviewed and, implications, limitations of the project, and future research are considered. Finally, the key findings are summarized in the conclusion.

Literature Review and Research Questions
Institutional theory has been effectively used to examine the adoption and extent of adoption of organizational forms such as CSR [29,30] and been used both to investigate how organizations are shaped by their institutional environment and, in turn, shape institutions. Initially used to explain sameness across organizations in the same sector, institutional theory has, in recent years, been used to explain differences across organizations and the presence of non-conformity among groups of organizations. In this paper, we use institutional theory to examine the integration (or the lack thereof) of WE disclosures in CSR reporting.
Institutional theory argues that organizations structure themselves based on what is symbolically acceptable, i.e., an appropriate form of organization in their field. This is not necessarily what is most effective and efficient financially. DiMaggio & Powell [31] refer to an organizational field as "those sets of organizations that, in the aggregate, constitute a recognized area of institutional life; key suppliers, resource and product consumers, regulatory agencies, and other organizations that produce similar services or products" (p. 148). Organizations structure themselves based on their interaction with other organizations and their institutional environment, which provide the conventions concerning how best to structure an organization [32]. Therefore, organizational forms may be partly the product of their functionality, effectiveness or efficiency, but also heavily influenced by the symbolic conventions set and reinforced by key actors and institutions in an organization's environment.
Isomorphism is the process by which organizations become similar to other organizations, as they conform to prevailing conventions in their environment [31]. Aligning with existing expectations or an organizational template enables organizations to tap into required resources (e.g., legitimacy, customer support). Conversely, neglecting to construct organizational forms that reflect institutional pressures may result in a loss of legitimacy, increased scrutiny, and a potential suspension of resource flows [32,33].
Organizations may encounter pressure from three sources [31]. Organizations may encounter coercive pressures from other organizations on which they depend for resources such as government, customers, social movements and organizations. Organizations may also mimic established organizational forms in times of uncertainty. Conformity is also made more likely to the extent that an organization's members are drawn from a common hiring pool and have undergone similar formal training and processes of socialization. This is because managers may feel compelled to develop their organization in a similar fashion to align with other organizations or prevailing guidelines.

Theoretical Background
Institutional researchers point out that organizations are not "passive" players, involuntarily and automatically adopting organizational forms and practices [34][35][36]. Rather, they have a degree of agency and are creators of organizational forms. Organizational actors may, therefore, seek to strategically legitimize their organization according to one or more logics within an environment [37,38]. This means that organizational personnel engage in a range of responses to institutional pressure ranging from passive acceptance [36] to partial responses that act as compromises [39] to resistance and working outside of institutions [40].
Organizational fields are not uniform; influences within a field may be contradictory, with multiple logics at play [40][41][42][43][44][45]. An institutional logic can be defined as "a set of material practices and symbolic constructions which constitutes its organizing principles and which is available to organizations and individuals to elaborate" [41], (p. 248). Pressure for companies to conform varies by organizational field, and the segments within them. Within some fields, organizations are highly regulated and thus face considerable coercive pressures to adopt particular organizational forms. Aligning an organization to an institution, however, comes with challenges. Scott [46] notes that many organizational fields are "fragmented or conflicted, containing competing requirements or prescriptions" (p. 430). One challenge that results from the pressure of multiple conflicting logics is that orienting an organization to only one may mean losing legitimacy in the eyes of another [36,41,47]. The non-uniform character of the logics and pressures to conform can create variations among organizations; so how one group of organizations responds to their environment differs from another group of organizations.
Goodrick and Salancik [48] explained that how organizations respond to institutions is also affected by the degree of completeness of the institutions in a particular field. As Goodrick and Salancik [48] explain, "When institutional standards are uncertain . . . they will be unable to constrain actions completely. Existing alternatives may then be selected without loss of legitimacy. With the discretion resulting from uncertain prevailing institutional standards, organizational or individual interests may frame strategic choices" (p. 3). Organizations have the opportunity to exercise discretion when there is uncertainty about institutions. Uncertainty can emerge under particular circumstances, such as when there are known institutional goals but the means to achieve them are unspecified, underspecified or unclear (p. 4). Institutional uncertainty may also emerge when there is confusion about institutions themselves "such that beliefs about legitimate purposes are in conflict or even contradictory" (p. 5). The fragmented, incomplete nature of institutions provides organizational actors with discretion, enabling them to selectively innovate in response to external pressures [40,49].

WE, an Institution of CSR, and a Corridor of Conformance
Building on Goodrick and Salancik [48], a key "institutional goal" for CSR reporting is to disclose information on the triple bottom line of economic, environmental, and social performance, including employees. How reporting on employees should be carried out, however, is largely unspecified. The institutional expectation to report on WE coupled with the uncertainty regarding the institution creates what we refer to here as a corridor of conformance: a "passageway" in which an organization's reporting must fall within to attain (retain) legitimacy among multiple relevant stakeholders such as rating agencies, investors, customers, regulatory bodies, and the general public. The corridor's "walls" consist of institutional CSR conventions but, so long as an organization's reporting stays within the confines of the corridor, the corridor permits them a high level of freedom to select how to respond to institutional pressures. The corridor's composition influences what is acceptable to report, such as the WE topics to disclose. The effect of the corridor is fourfold: creating an expectation that companies will report within the corridor's confines and work to remain within it to achieve legitimacy; influencing companies to disclose WE information at least at a minimal level; suggesting disclosing information about a range of WE topics with a good deal of discretion regarding what topics are reported and the format reporting takes (thus, the extremely high level of variation in WE disclosures); and, because of the corridor's low barrier to entry combined with its wide width, hindering more holistic reporting, such as reporting on employees' psychological health.
The institutional conditions that support the discretionary selection of WE disclosures are what we focus on with regard to WE CSR reporting. The selectivity by which companies can conform to the CSR institutional environment is facilitated by several critical factors, including the voluntary nature of CSR and the ambiguity around the predominant concepts that have emerged in CSR discourse: environment, social, and community [13,21,50]. Indeed, there are few mandatory CSR reporting guidelines [19]. This can be seen, for example, in the great range of indicators companies choose to report on WE in CSR reports [13]. The lack of universally applicable guidelines that minimize idiosyncratic interpretation and that are enforceable has created a context in which companies conform in principle to CSR reporting, but yet can be extremely selective in how they choose to do so. Agreement on what "good" or satisfactory reporting is on areas such as WE is problematic because this is an area that has received relatively little attention in the CSR literature compared to the natural environment [12,51]. These conditions promote an institutional corridor of conformance in which organizational managers need to provide some minimal level of WE disclosures but can be selective, picking and choosing WE topics and related indicators to report on and the depth to which they report.

Research Questions
Organizations are under a great deal of institutional pressure to adopt CSR practices and reporting mechanisms [30]. Companies that do not undertake CSR reporting are seen as deviating from prevailing societal norms and risk censure, or at the very least, appearing less legitimate than the organizations in their field that adopt CSR reporting [52]. Given these institutional pressures, the majority of large companies engage in CSR reporting [53].
There are multiple motivations for companies to engage in CSR practices and reporting. The institutional conditions in which companies are enmeshed play a critical role in the extent to which companies engage in CSR. As Campbell [29] explains, "Corporations are more likely to act in socially responsible ways the more they encounter strong state regulation, collective industrial self-regulation, NGOs and other independent organizations that monitor them, and a normative institutional environment that encourages socially responsible behavior" (p. 962). Other motivations for CSR reporting have been widely explored [54][55][56].
The form that WE reporting takes can be partly understood by recognizing the prevailing form that CSR reporting, more broadly, has taken. One of the most prominent forms of CSR reporting is the instrumental or strategic form [57]. As Bondy et al. point out, this strategic form is "a business innovation designed to support profit generation" (p. 282) and departs in important ways from the initial aim of CSR to improve society. The focus on CSR reporting for normative reasons has largely been subordinated to instrumental concerns. This orientation has important implications for the coverage and reporting of WE. Studies that focus on employee-related disclosures demonstrate that companies' CSR practices and reporting are heavily influenced by an instrumental orientation to CSR rather than a normative one [8,17,27].
There is an extensive literature on what companies are reporting in CSR and a growing body of literature that focuses exclusively on WE disclosures in CSR reporting [10][11][12][13]58]. However, these studies rarely examine managers' perspectives regarding WE and how it should be reported. As such, the literature on the process for deciding what to report is limited, particularly when the focus is on WE.
Managers may be motivated to report on WE in their CSR reporting for several reasons. As employees are a core part of the "triple bottom line" [59], reporting on them is expected among stakeholders and, therefore, reporting on employee-related areas garners legitimacy for companies [14,16,60]. In terms of competitive advantage, reporting on employee-related issues may improve recruitment and retention [61,62]. However, motivations for reporting on WE issues have largely been unexplored. This is an important knowledge gap given the significant impact WE can have on employee health and well-being, product quality, and productivity. Based on the above, we propose our first research question: RQ1: What are managers' perspectives on their current WE disclosures in CSR reporting?
In an institutional environment that can be described as "patchy" [40], with a weak regulatory environment and few standards for WE reporting in CSR that are monitored, companies have a good deal of latitude regarding what to report [48]. For example, companies have a wide range of WE indicators on which they can report. However, organizations have tended to report on a fairly narrow band of indicators (with the majority of these focusing on physical health and safety) [12,58] and in ways that make their interpretation extremely difficult [10]. Leading indicators are indicators of an organization's ongoing performance such as prevention, training programs that may reduce workers' exposure to injury or illness hazards. Lagging indicators are indicators such as how many injuries or fatalities have actually occurred in a company. Evidence in the CSR literature [13] suggest that companies tend to do a better job of reporting lagging, rather than leading, indicators.
In their deliberations about how to compare CSR reports, managers are faced with a formidable choice in selecting the indicators on which they will report [26,54,56]. However, a good deal of the open-endedness of selecting the topics and indicators about which to report is reduced by the influence of prevailing reporting norms, such as reporting guidelines [52]. A number of reporting frameworks and assessment tools, such as the Dow Jones Sustainability Index (DJSI), the FTSE4 Good, the United Nations Global Compact (UNGC), and Global Reporting Initiative (GRI) have gained prominence and influence what companies choose to report [63]. The lack of consistency between the various standards and guidelines, reminds us of Goodrick and Salancik's [48] notion that there may be agreement about how some practices fulfill institutional goals but not others. This fosters a context in which companies have considerable discretion concerning the content of their CSR reporting and what format the information should be reported in.
Consider the example offered by the most well-known CSR reporting framework, the Global Reporting Initiative (GRI) [64]. The GRI standards include a set of indicators that companies can use as a framework to report on their CSR efforts. Over the course of its development since the late 1990s, the GRI has become institutionalized [52,65]. Demonstrating alignment with this institutionalized practice can confer a number of potential advantages on companies that choose to employ the GRI. Most prominently, adopting internationally recognized practices can demonstrate an organization's accountability with respect to CSR performance, contributing to its desire for enhanced institutional legitimacy [66]. Moreover, as more companies align their reporting with the GRI requirements, the common reference point it provides can improve comparability of CSR reporting. The great variation in reporting practices has previously been highlighted as an impediment to improved work environment reporting [13]. The set of standardized indicators included in the GRI can provide a needed framework for companies undertaking CSR reporting. Critical appraisals of the GRI, however, point out that there are problems with it. In particular, the GRI framework has been criticized on the grounds that it better represents the concerns of business rather than non-governmental actors and labor [52], its use as a guide to disclosure obscures rather than reveals information about sustainability [67]. It also obscures the negative activities of organizations despite the GRI's stated aims of transparency and credibility [9,68]. The GRI does contain employee-related indicators, some of which focus on worker health. However, the coverage of employee-related areas is limited [12].
The extent to which WE indicators have been reported is narrow, with these efforts often aligning with the guidance of CSR reporting standards, such as the GRI. In particular, reporting on the concept of WE has been conflated with reporting on occupational health and safety. As a result, what is reported is focused on a narrow conception of WE. Issues such as disorders and illnesses resulting from cumulative trauma and work-related psychosocial hazards are rarely addressed-though both are leading causes of worker-related sickness and absences [6,10,12,69]. It is, therefore, important to better understand the role institutional standards play in guiding companies in their WE reporting and the extent to which this helps or hinders reporting on a broad array of WE issues. Based on the above, we propose our second research question: RQ 2: What role do institutional pressures, such as reporting guidelines, play in shaping the form that WE reporting will take?

Research Methods
To address the research questions, twenty CSR managers from large companies in Canada were interviewed. Firm size has been shown to increase the likelihood that a company is engaged in CSR reporting [70]. Each interviewee represented a different company. Reporting on the working environment is a leading-edge practice and so we focused on leaders as we expected them to have well-developed CSR reporting practices regarding the work environment. The companies were identified as leaders based on MacLean's [71] top 50 companies in CSR. MacLean's news magazine, in collaboration with Sustainalytics, selects the top 50 by extracting the top 5 companies from 10 industry groups. According to MacLean's, the selected companies "have demonstrated strong performance in areas such as environmental initiatives, impact on local communities, treatment of employees and supply chain management." Most CSR reporting in Canada is done voluntarily; there are few mandatory requirements. Canadian banks, insurance companies, trust and loan companies, however, are required to publicly disclose their performance on a limited set of CSR-related indicators, such as the total value of their charitable donations, under a federal government regulation (SOR/2002-133) requiring the publication of a public accountability statement. A number of papers have studied the publicly-available CSR disclosures of Canadian companies [13,63].
We contacted all fifty companies via email. Each company was emailed at least three times and, in each case, we attempted to identify and contact a person who was directly involved in the company's CSR reporting. Of the 50 companies we emailed, 30 responded to our email and, of these, 20 agreed to participate in the project. The companies in the sample were from a range of industries and details about the companies, including the industry and title of participants, are provided in Table 1. The companies in the sample were large, with the range of employees between nearly 800 and 50,000. Most of the companies had more than 10,000 employees. The participants all self-identified as CSR managers. The interviewees had been with the company from 1 to 11 years, with a mean of 3.5 years. Interviews ranged in length from 32 to 66 min (average of 51 min) and were audio recorded. The majority of interviews (17) were carried out over the phone; the others (3) were face-to-face. The interview format was semi-structured and allowed for flexibility to follow the lead of participants.
The interviewer (the paper's lead author) used probes and follow-up questions to discuss issues that emerged during the interview [72,73]. The interviews consisted of questions about the company's CSR reporting, how the company decided what to report on, and how the company conceived of and reported on WE, in particular. For example, the questions included "How does WE relate to your company's strategic objectives?", "How does your company assess its WE performance?", "How satisfied are you with the metrics your company is currently using?", "How does your company determine what WE elements to discuss in its CSR report?", and "What message does your company hope to convey to readers of its CSR report about WE?" The analysis proceeded through three general steps: data extraction, data organization and then analysis. The interviews were transcribed verbatim and the transcripts were imported into NVivo. A coding scheme was created based on predetermined concepts (such as pressures, benefits and satisfaction) in combination with material from the interviews based on multiple readings of the interview transcriptions. The interviews were coded and the data was examined for relevant themes. The analysis involved moving back and forth from the raw interview data to the interpretations [74]. The constant comparative method [75] was employed, comparing within and across the interviews to identify consistencies and differences. To enhance the "credibility" of the analysis [76], all three of the authors read the transcripts, wrote comments on the transcripts, and participated in discussions about the findings. These practices functioned as "analyst triangulation" and "peer debriefing" to assess the interpretations of the data (p. 30), [77].

Findings
In the first part of this section, we examine managers' perspectives on inclusion of WE in CSR reporting. We then focus on the key influences on reporting, with particular emphasis on the institutional pressures. For the companies in our study, WE was viewed as a fundamental part of the "triple bottom line" [59]. A key reason for these companies to publish WE information in their CSR reports was to partially fulfill their social responsibility goals. As one interviewee pointed out, "one of our core key values is putting people first, and so how our employees see the company and our job is critical to our success, it's an important thing for us" (#19). Similarly, another CSR manager noted the close connection between the corporation and CSR reporting: "So work environment fits into our corporate code of conduct, specifically safety elements of it" (#12). Interviewees also stated that stakeholders, such as investors and customers, wanted to see WE information and that corporate sustainability performance indices, such as the Dow Jones Sustainability Index, expect it. CSR managers discussed that WE reporting had strategic value to the company-in particular, the participants noted that WE reporting had benefited their companies' reputations. Building and maintaining reputation were defined as important among companies and reporting on WE was part of that. A common sentiment was that CSR reporting enhanced a company's value. Interviewees variously noted that WE disclosures contributed to "building value" (#4), creating "value through reputation" and "build brand awareness" (#10) and "we're firm believers in reputation driving our value" (#13).
A key benefit of WE reporting was that it could attract potential employees. Interviewees talked about conveying to readers that their company performed well across a suite of employee-related areas such as training, diversity, engagement, compensation and benefit packages. In the words of one interviewee, the CSR report allowed the company to demonstrate that it was "a great company to work for" (#6). Another participant explained "people see [company name] as a fair and equitable employer and would like to work with us, which means we can attract the best workforce" (#2). Similarly, an interviewee noted, "we are very growth focused right now. It helps also to attract new employees" (#12).
Another benefit was that it allowed the companies to demonstrate transparency. This helped to establish a company's status as a credible CSR reporter. As one interviewee explained, "We always say, the whole point of reporting is really to show the good, the bad, and the ugly, and to drive our reputation through what we do . . . Why report it? Because we have nothing to hide" (#13). For two of the companies, transparency was particularly important as the managers acknowledged that people tended to hold negative pre-conceptions of the companies within their industries. In the words of one of these managers, people had "pre-established judgments" (#10) of firms. For these two firms, reporting on WE was an important means by which to counter negative industry stereotypes and to differentiate their companies from others. As one interviewee stated, in this context it was important "just being able to say 'well look, we are transparent, here's the information and . . . we have made a commitment to be transparent about our activities'" (#19).
Interviewees (10 in total) also noted that WE disclosures were also of interest to investors. Reporting information about WE signaled to investors that risks regarding labor relations, employment standards, and occupational health and safety are being monitored and managed. Reporting on these risks may differentiate a company from others that are not and increase a company's attractiveness as an investment opportunity. For instance, one interviewee said, "It's about I'd say risk mitigation, and looking then at maximizing the opportunities, and I'd say that that's really what they're looking at, . . . from an HR perspective they're certainly interested in understanding our labour practices and in particular our labour relations. Our relationship with the union, of the [number] people that work for our company must be something like [three-quarters] that are unionized. So, it's quite a big portion. If we're going to be a successful, healthy, profitable company, then we need to make sure that we have good labour relations and healthy and happy people who are well trained to be able to run [company name] and be safe" (#3).
Another participant, highlighting the importance of disclosing information about employee retention, stated: "I think that they're [investors] interested in seeing that we are able to attract and retain qualified talent, . . . we need to make sure we have the people there that are able to work and . . . get the work done, I think that a low turnover rate also indicates that overall employee satisfaction is quite high" (#4).
WE reporting was generally thought to positively enhance company reputation; however, company representatives confided that reputational concerns also meant that some material that could possibly present the company in a negative light would be excluded. There were WE-related issues that the companies did not want reported because they were concerned how the publication of this material would affect the company's reputation. For instance, this interviewee noted that: "The number of cuts, or bruises and things like that, we don't like saying that in the report because of just the way it sounds, that our employees are getting cut on the job, we avoid that kind of stuff. The other thing is employee turnover and those types of numbers, we don't like to report on those, because it happens, I mean in the [industry] it happens quite frequently, and I think that by doing that wouldn't, I mean if another industry saw that number, it's not comparable" (#8).
These reputational concerns are not uncommon. O'Neill et al. [10], for example, have referred to practices in which companies exclude or deemphasize information regarding health and safety that would threaten, or at least blemish, a company's reputation as "safewashing."

Satisfaction with WE Reporting
When discussing their current WE reporting, two key themes emerged from the participants. First, all of the CSR representatives stated that they were generally satisfied with their company's WE reporting. The second theme, however, was that most representatives (18) stated that their WE reporting could be improved. In the words of one CSR representative, WE was the "weak link in our report" (#17). The following excerpts typified the sentiment that, although managers were mainly satisfied, they saw that WE reporting needed improvement.
"I would say I'm 85 to 90% satisfied with them. I feel like we've really, for what we can measure, and how we can break it down, so that we are the most honest in our disclosure, we've done a good job of slicing and dicing" (#9).
"Yeah, I think what we're reporting is fit for purpose right now, and for where we are as an organization, but it would be great to see us reporting more, and to expand into I guess I'll call them more non-traditional metrics" (#15).
Importantly, participants' satisfaction with their company's WE reporting, needs to be placed in the institutional context in which reporting occurs. At work here is "reference bias", in which assessments of WE reporting are largely based on what (a) other comparable companies are doing, (b) is being asked for by, or is of perceived importance to, external parties (e.g., rating agencies, investors), and (c) is outlined in the GRI and, to a lesser extent, other CSR reporting guidelines, which we discuss in more detail later in the paper. Importantly, none of these external groups in the firms' organizational field demanded much in terms of WE reporting. Their reference point for assessing their WE reporting and subsequent satisfaction were linked closely to the "best practices"-the dominant form of WE reporting, which was low-level reporting. This participant's perspective nicely captures how firms evaluated their WE reporting, "I think we're satisfied with what we've been reporting, from a best practices point of view. The methods that we employ for reporting I think is the way that companies are going, I mean having a priority matrix so that you're taking what your stakeholders are wanting from you and then, with a framework such as GRI, as a way to determine what your report is really I think best practices, and so I think that we're quite satisfied with that" (#4).
Another important element is that participants' interpretations of how well they were reporting on WE were based on an instrumental orientation to reporting, not on a normative orientation (what was best for the health and well-being of workers); on what reporting tools external stakeholders expected to see [57,78]. Notable by its absence was discussion of moral or even strategic organizational performance in their WE reporting.

Nature of WE Indicators
The participants reported using a combination of leading and lagging indicators. There were key differences in how companies said they performed in reporting on lagging indicators and leading indicators. Interviewees recognized the importance of leading indicators, often noting that there was important work to do to enhance how they reported on leading indicators. As one interviewee pointed out, "I'd say generally no, we don't have many of them" (#13). Another participant elaborated: "I think we need a little bit more work in those areas as well, . . . maybe adding the average hours of training per employee, and gender . . . we have the percentage of employees receiving performance and internal development reviews, I'm trying to see [looking at CSR report] anything else that's related to education . . . . I know there's measurement of these things, so we haven't, we'd like to see maybe some more in the training and education, especially putting some emphasis on it" (#9).
While acknowledging the need for improved leading indicators, participants also recognized that conceptualizing and reporting leading indicators was more difficult than measuring lagging indicators. As one participant noted, "Leading indicators are always tougher to narrow down" (#13).
One participant discussed the comprehensiveness of his/her company's reporting on employee-relevant topics: "So it's kind of traditional stuff like percentage of employees receiving performance reviews, demographics around-percentage of our new employees that are male versus female, age demographics, employee turnover, return to work, . . . retention rates after parental leave, . . . training and development, education and assistance, scholarships, and we also track employee wages, employee benefits, percentage of basic salary of women to men, . . . equal opportunities and diversity metrics, again breaking down our demographics into workforce diversity and percentage that's Aboriginals, visible minorities, persons with disabilities and then age and then we look at diversity in our management levels. . . . we also do health and safety, but it's more on incidents and injuries, it's not necessarily looking at internal work environments . . . so I think the kinds of metrics that you're talking about we actually don't report on" (#15).
The companies in our sample followed the lead of other companies in that they published a fairly narrow range of WE indicators. This finding converges with other research that has found similar results, pointing to the influence of popular CSR guidelines [13]. We discuss these points in more depth in the next section.

Identifying Relevant WE Topics to Disclose in CSR Reporting
The participants explained that determining which WE topics would appear in a CSR report was the product of a multi-stage process that involved drawing on information from several sources including stakeholders. Companies sought input from their internal stakeholders (e.g., employees) and from external stakeholders, such as customers, analysts, investors, and rating agencies that rank organizations on their CSR performance, such as the Dow Jones Sustainability Index (DJSI). Companies also gathered insights about what to report by examining other company's reports.
All of the companies based their decisions about what to report on, in their words, "materiality" or what was relevant to the firm and this was largely influenced by entities outside of the organizations. Determining materiality allowed companies to overcome the uncertainty of what to report and prioritize areas upon which to focus-a critical exercise given the broad range of possible CSR issues and the large number of stakeholders that may occupy a firm's organizational field. To determine materiality, companies often undertook a "materiality assessment," or "materiality review," a formal means of identifying the topics that are most relevant to a company's CSR reporting. One interviewee explained: "We hold a materiality assessment with our key stakeholders. . . . so we do an internal prioritization with the leadership, so again anyone from HR, Aboriginal Relations, and again anything that sort of relates to Corporate Responsibility Reporting, and then we have a smaller group of external stakeholders that come and help us validate that information and prioritize it . . . . and that's how we decide if we are missing anything that our stakeholders would like us to talk about in our report. And then obviously there's some work environment issues that are on there that sort of get moved around whether they're higher or lower in priority, so you know voluntary turnover or benefits, learning and development, things like that" (#9).
Although using a less formal approach than other companies, another interviewee provided additional detail on how they use materiality as a means of determining what to report, "It's a combination of things: it's GRI, it's kind of what we've done in the past, which is largely based on what the sort of norm is in our industry, what you talk about and how you talk about it, and so this past year we did a sort of materiality assessment like sort of a top ten, "these are the things that people are most interested in hearing about", "these are the things that we get the most questions about," and made an effort of saying these are the things that we're actually going to give text to, or more These processes were an exercise to determine what was most valuable to report. A theme that ran through all these companies' efforts to prioritize the WE issues that would be disclosed in CSR reports is the high degree of value that they placed on reporting the information that external stakeholders wanted to know about. The companies looked at several pieces of information to provide a basis for decision making about what was relevant for the company. The companies assessed the relevance of topics based on stakeholders' and managers' interests.

Key Influences on Reporting
We found there were several key influences on the selection of WE topics. Company representatives talked about the importance of incorporating current reporting trends into their reporting and stressed the importance of considering stakeholders' reporting expectations. We found that the companies' selection of the WE topics that would be reported was heavily shaped by prevailing CSR reports, reporting standards, and guidelines. In particular, according to the CSR managers interviewed in this sample, the GRI provided integral guidance regarding the topics that should be reported and how information should be reported. Sixteen companies discussed the GRI's influences on their reporting. This excerpt exemplified the comments about the GRI's influence: "The framework for our sustainability report is really the Global Reporting Initiative, GRI. So we're very much guided by that framework in terms of the content of our report" (#3). Other interviewees elaborated on this point. For example, one interviewee explained: "So our CSR report is based on the GRI framework, the guideline, and we follow the Level B, so we basically picked which indicator we want to report, and so some of the work environment elements or aspects . . . based on what was outlined in the GRI, in terms of health and safety ratio, head counts of employees, and compensation. So those are driven by those indicators in GRI, and we try to expand on it by giving more stories behind just the indicators" (#7).
For the CSR managers interviewed, the GRI reporting framework held a vaunted place in CSR reporting. Some pointed to its prominence as a reporting framework, noting "it's leading best practice" (#20) or "the GRI has become the internationally accepted framework for reporting" (#5). Others noted that there was "credibility" attached to using the GRI. "GRI, that framework is really I would say an established framework particularly for any S&P 500 for example. In big business, people who are serious are looking at the GRI" (#3). The implication here is that if a company wants to symbolize that it is taking CSR reporting seriously, then it should adopt GRI reporting practices.
Interviewees also pointed out that using the GRI framework allowed them to provide "standardized" reporting. This provided important advantages for the companies composing the reports: it reduced the volume of reporting that had to be done. As one person put it, it's "overwhelming" not to have a standardized way of reporting. An interviewee explained "a lot of other reporting initiatives are now harmonizing with the GRI . . . your CSR report, if it's a GRI compliant report, now can satisfy other reporting initiatives" (#15). It also enabled the companies to use a common language that, according to the interviewees, was globally recognized. Thus, it enabled companies to compare themselves to competitors and allowing external audiences (e.g., analysts, customers) to compare one company to another on a consistent set of measures. Using the GRI and its indicators "create that ability . . . for consistency for comparison across different organizations" (#12). Another explained, "What's the value in it, for us? For us it's definitely having a standardized way of reporting, we can compare against our peers, it allows others to compare us against our peers" (#4). Importantly, standardization was considered important as it aided internal and external stakeholders to meaningfully assess a company's efforts at CSR reporting.
The perspectives of the CSR managers align with the CSR literature, which suggests that reporting standards such as the GRI, AccountAbility, the Dow Jones Sustainability Index, and FTSE4Good have become increasingly important in CSR reporting [52]. These standards give companies guidelines regarding what to report and how to report it. Their use also provides companies with increased legitimacy, and because there is shared understanding of the criteria used in these standards, they are widely recognized.

Weak Influence to Disclose WE information
While managers discussed that disclosing WE information was a core part of CSR reporting, they also indicated that there was only weak pressure from stakeholders to include WE elements beyond basic occupational health and safety (OHS) data, such as the total number and rates of workplace injuries. This interviewee's statement was typical: "They're not really asking for anything. I think they're all expecting that we put together a report and a GRI and as long as we're putting out our report to the common standard, I think that we're not getting requests for different information" (#19). As one interviewee reported, "in terms of what we report, the content in our report, that's all determined by our materiality review, and workforce elements never really rank very high in terms of impact to [company name] and impact to our stakeholders. It's not something that we're asked about externally, so there's no real external driver" (#15). Another interviewee noted that in his/her conversations with investment stakeholders, WE elements rarely came up: "I'm the one who takes the phone calls from the Canadian Pension Plan investment board or [bank name] when they're doing their ethical screening, CSR screening, and I don't ever remember being asked about wellness, employee wellness. They'll ask about labour relations, that's it. . . . I'm pretty sure if we did our own internal materiality assessment or if we did it with a group of external stakeholders, with the exception of our employees, I don't know how frequently that would come up. . . . It would be up there somewhere but it would be fairly low" (#16).
Others noted the pattern of communicating with stakeholders and seldom hearing that WE should be prioritized. This interviewee's excerpt is representative, "We do stakeholder engagement every year. We include various different stakeholders and that [WE] doesn't really come up very often. . . . one of the questions is, "Are we covering the areas that you think we should be, as a company?" And usually when it comes to workplace environment that traditionally hasn't come up. What usually does come up is environmental issues . . . donations and what we're doing in the community, but not so much on the work environment piece, and that's also true when it comes to external stakeholder engagement that we do with our customers. Because we also ask them to rank the issues that are important to them, like with what we should be doing, and how we treat our employees and health and safety is always part of that, and it's never up on the top. It's usually close to the bottom" (#8).
Notable in the above excerpts are that companies placed a great deal of importance on how external groups define WE and that WE issues were ranked low in importance across a range of stakeholders, including customers.
Interviewees, the majority of whom indicated that their company used the GRI, also discussed the lack of attention to WE elements in the GRI guidelines. As one interviewee noted "we report on turnover not on mental health. And . . . that was a comment I gave to GRI: . . . they don't really have a lot on the health side within the GRI" (#13). Similarly, with regard to worker health, an interviewee said, "the GRI doesn't give a ton of guidance on that" (#16). Another noted that WE was a broader concept but this breadth was not captured in the GRI "there's not a whole lot of demand for that information from the usual forms like GRI, and they've got indicators around health and safety and grievance procedures and training, but there's a whole lot more that goes into a work environment" (#14). Despite managers' acknowledgement of the lack of attention that the GRI places on WE, they did not see this as reason to alter their current reporting practices.
External parties, such as investors and rating agencies, played an integral role in determining the content of the CSR reports. The GRI played a significant role in influencing the content that should be included and how this should be formatted. It was clear that participants believed that few external parties were interested in WE and, therefore, companies were not interested in reporting on it. This included a lack of focus on WE in the GRI. Interestingly, when managers discuss the influences on CSR reporting practices, they did not discuss the normative importance of reporting on WE factors.
Their rationale for reporting and reasons for reporting were focused on signaling and the symbolic importance of reporting [78].

Discussion
This study revealed several important findings about CSR managers' perspectives regarding WE disclosures in CSR reports. First, participants reported that there was value in WE reporting and it was a fundamental part of their companies' CSR reporting. The CSR managers stated their satisfaction with current WE reporting. However, in-depth reporting of WE was not prioritized among the companies. Second, reporting of WE was heavily influenced by prevailing reporting practices, external stakeholder expectations, and reporting standards (e.g., GRI). These institutional pressures greatly influenced managers' selection of the topics to report and the indicators to disclose. Third, the institutional environment encouraged companies to report on a narrow band of WE dimensions. Indeed, external stakeholders had little to no expectation for reporting on the broader area of WE, such as psychological health.
Research Question (RQ) 1 Managers' perspectives on their current WE disclosures in CSR reporting. The interviewees discussed the benefits of reporting on WE, noting the place that WE played in maintaining a companies' reputation as a good reporter. While generally satisfied with their WE reporting, interviewees acknowledged that some areas of WE reporting needed improvement, such as the reporting of leading indicators. Traditional occupational health and safety indicators (e.g., number of injuries) were often used by the companies but indicators that could represent a broader conceptualization of physical, mental and social well-being were under reported by most companies. This aligns with the findings in a growing number of studies regarding worker elements and health and safety that show that some types of indicators of worker well-being are rarely used and, importantly, those that are used may give audiences an incomplete picture of worker health within a company [6,10,12,13,69,79]. While some research has indicated that consumers would preferentially select, and maybe pay a premium for, goods made under healthy working conditions [80,81], this motivation for reporting WE was not discussed by the study participants. Furthermore, there appears to be some conflicting response from participants, between the drive to ensure employee-related information is a key feature of their CSR reporting and the reality that external actors are not asking about WE in their reports. Further research in this area is warranted. RQ 2: The role of institutional pressures in shaping the form that WE reporting will take. The literature on institutional theory suggests that institutional pressures motivate companies to adopt CSR reporting practices and include aspects of the triple bottom line, including WE aspects, in their reporting. However, in some institutional contexts, the specific means by which institutional goals must be met is unclear and organizations have discretion in how to respond to institutions [48,82]. This is the case with CSR fostered by its voluntary nature-and is particularly relevant in the context of WE reporting given that there is little prescriptive direction from stakeholders, reporting guidelines, or other reference points about how, and the extent to which, WE should be reported. This aligns with Goodrick and Salancik's [48], (p. 5) suggestion that when there is institutional uncertainty around how organizations should respond, this fosters conditions in which organizations can use discretion in how they respond and often do so in ways that meet their strategic interests. There are institutional expectations that companies report on WE-it is a fundamental part of the triple bottom line-but these expectations are limited to a narrow definition of WE that often only includes a small number of lagging health and safety indicators. The disclosure of a heterogeneous mix of OHS indicators is a reporting practice shared by companies [10,13]. This context has created a wide corridor of conformance in which companies can disclose a vast array of WE information, even at a shallow level, and this enables an organization to align itself with prevailing CSR norms and provides the legitimacy of doing so. From the perspective of those interviewed, even this limited form of reporting provides firms with legitimacy among a diverse set of stakeholders, rating agencies, investors, and prospective employees. For interviewees, reporting on WE issues contributes to a firm's credibility, especially, when using GRI reporting guidelines, which most companies did, enhancing their firm's reputation among key stakeholders. Importantly, this suggests that WE reporting is not completely undetermined or completely voluntary and demonstrates from where some prevailing reporting practices are emanating. Indeed, what makes this a corridor is that reporting within it is delimited and to enter into this "passageway", or corridor of conformance, a firm needs to be addressing WE to at least a minimum expected standard, in this case OHS metrics such as the annual number of workplace injuries. This finding aligns with other research in the area [26,56,83].
The current state of WE reporting for these companies is fostered by several factors. First, the voluntary nature of CSR disclosures largely leaves open both the WE topics that companies should be reporting and the level at which these topics should be reported. While companies are expected to engage in CSR reporting, there is a lack of institutional pressure emanating from the environment around WE reporting, specifically, and there are few "rules" about what needs to be reported regarding WE. Second, prevailing reporting guidelines, such as the GRI, provide legitimacy, but set the expectation for WE reporting at a very low level [79]. Third, the manager's expectations about what WE disclosures there should be were based on stakeholders in the organizational fields of the companies we interviewed, such as comparator organizations, customers and rating agencies, all of which did not request WE reporting. Fourth, the CSR form that has gained prominence, instrumental reporting, does not focus on WE [17,27,57]. The lack of pressure for a normative orientation to reporting is reflected in managers' orientations toward reporting. In a context in which instrumental reporting is defined as best practice, managers will principally focus on how reporting aligns with external stakeholders and whether that confers legitimacy rather than reporting on topics that genuinely benefit the community, workers or others who may be affected by the actions (or inactions) of the corporation. Currently, the primary criterion for achieving legitimacy in WE reporting is not whether there is appropriate coverage of workers' health and well-being (more normative), but rather whether the reporting conforms to agreed-upon reporting practice (more instrumental given the prevailing institutional practices), which does not cover WE in a meaningful way. Furthermore, because of the corridor of conformance that is supported, there is a range of ways to achieve legitimacy.
The results show that the companies' WE reporting was heavily influenced by CSR standards and guidelines. In particular, the GRI shaped the content of WE disclosures. Interviewees interpreted the use of the GRI as almost compulsory for reporting. According to interviewees, using the GRI provided credibility, signaling to relevant actors in the organizational field that the company was undertaking high-quality reporting and provided a high level of standardization permitting stakeholders to do intra-firm comparisons across years as well as inter-firm comparisons. Use of the GRI enables companies to become isomorphic, thereby creating legitimacy. However, modeling a firm's WE disclosures on the GRI's guidelines does not foster the presentation of a comprehensive picture of WE [6,12,15]. With regard to WE, anything but comparisons that focus on the most basic WE indicators, such as the number of injuries, are difficult to do as companies use a vast array of indicators to report on WE [13]. The large number of indicators used in reports, however, does not reflect the quality of reporting [13]. This finding converges with other examinations of CSR reporting [52,66]. Achieving a transparent look at WE is challenging because existing standards and guidelines, including the GRI, often include few WE aspects [6,12,13,15,69,79]. The GRI in particular has been criticized for its lack of specificity regarding how topics should be reported [50], its few indicators that exclusively focus on employees, and its reliance on a narrow conception of health and well-being.

Implications
This qualitative study helps extend institutional theory [27,31,34] by shedding light on the forces influencing organizational practices around the reporting of WE in the context of CSR. While qualitative research may be subject to sampling and social-desirability biases [73,84], we report here the espoused views of knowledgeable informants from organizations in a range of industries that make higher than usual efforts to engage in CSR reporting. The conditions within the organizational field of the organizations examined in this study promote the reproduction of low-level WE reporting, which facilitates incomplete reporting, a practice that has been well-documented in CSR reporting generally [54,68,82,83]; and reporting with regard to employee health and well-being specifically [10,14,17,18,58], to the extent that some companies have been criticized for engaging in "safewashing" [10] or "worker washing" [18]. The prevailing WE reporting practices foster reporting that does not consider the range of factors that affect workers or that mental health and well-being should be re-considered [6,12,69]. The current conditions do little to enable cross-company comparisons of WE, which is a criticism of CSR reporting in general [13,50]. There is a need for improved guidelines for the reporting of WE aspects within CSR reporting contexts [79].
Through their use of prominent CSR standards and instruments, which are inadequate in terms of assessing and reporting WE comprehensively [6,15,79], companies ultimately risk creating conditions that have negative outcomes for workers, company performance, and society [85,86]. Companies can innovate and find new ways of WE reporting so that a more holistic conception of WE is adopted and standards and guidelines could be modified and new ones created so as to better capture WE. While further research can extend the sample, or explore the depth to which individual issues and other factors are operating in different regions, this work provides useful knowledge to develop strategies for improvement in this area that can help companies foster actions to create better WEs in their operations through the process of regular CSR disclosures.

Study Limitations
This study has provided some important insights into managers' perspectives regarding work environment reporting. However, the study does have limitations. The sample size, not unusual for a qualitative study, is small and that brings with it some constraints. Qualitative interviews provided a great deal of insight into the perspectives of managers and how they are interpreting the CSR environment and responding to it. However, qualitative interviews do not allow for generalizations to a larger population. Examining the extent to which our findings align with the trends in a larger population would be beneficial to the literature on WE reporting in CSR. Our sample consisted of managers from large companies that were known to be CSR reporters. Large companies are more likely to have the slack resources needed to devote to CSR reporting, which may mean this sample of companies was more concerned about WE reporting than companies with access to fewer resources. Earlier research has shown that large firms tend to publicly disclose more CSR-related information than small firms [70]. Further, the fact that the companies were high-level CSR reporters may mean that we might find that other companies pay less attention to WE. These elements are worth examining in future studies. While there was some variation with regard to the industry sector in the sample, the study sample was not large enough to permit a detailed cross-sector comparison. Given there is evidence of sector-specific responses to CSR reporting [87], it would be beneficial to do some cross-sector comparisons (e.g., resource extraction versus service sector) to explore whether there is a "sector effect" on how managers operate in the corridor of conformance.

Conclusions
This article contributes to the conversation about WE disclosures in CSR reporting, which is an important but largely understudied area. The findings shed light on managers' perspectives on the current state of WE reporting and give us insights into why companies tend to report on a limited number of WE dimensions. Importantly, the findings point to the fact that companies see the importance of WE but are content for the most part to report at a low level. Indeed, companies perceived the reputational importance of reporting on WE and saw WE as a fundamental part of CSR reporting. The extent to which WE is reported on is affected by institutional expectations that do not involve reporting on WE in depth. WE reporting was heavily influenced by the institutional environment, including reporting guidelines such as the GRI and managers' perspectives on the requirements of stakeholders, which typically did not request information that went beyond disclosing basic OHS information. Companies were clear that when developing their CSR reports stakeholders did not request reporting on WE. The GRI guidelines, which were widely recognized as a credible reporting framework with significant symbolic value, adopted by the majority of companies and played a key role in influencing CSR, have little influence on WE and, as such, the companies reported on few WE areas. This context has created a corridor of conformance in which organizations can, with relative ease, disclose minimal WE information and be selective about the WE disclosures they include in their reporting. In so doing they are conforming to the prevailing CSR institutional expectations and receive the attendant legitimacy. An implication of this is that there is little external motivation for companies to broaden their WE reporting to include health and well-being dimensions, such as psychological health, musculoskeletal health, or occupational disease and go beyond OHS (e.g., lost-time injury) numbers. Results suggest that improved WE reporting guidelines could help broaden the current bandwidth of WE issues covered in CSR reports and help standardize WE disclosure practices across companies, which in turn, can help companies foster more humanly sustainable workplaces.