The disclosure of concealable stigmas: Analysis anchored in trust

Abstract Prior research on the disclosure of concealable stigmas at work has mostly overlooked the moderating and mediating role of employees’ trust in their supervisor and organization at large. The absence of trust from this field of study limits organizational efforts to foster inclusion at work. Thus, this paper presents a framework for examining the multiple linkages between employees’ trust in their direct supervisors and their organization, and the disclosure decision. Trust is proposed to be embedded in intra and extra organizational context, both in terms of the legal framework and the HR policies and practices. On the basis of synthesis of the literature, the article extends previous research and reviews of diversity by providing systematic review and recommendations that can help promote diversity management efforts and ultimately contribute to employees’ well-being as well as positive organizational outcomes.


Introduction
The primary purpose of this paper is to explore the role of trust in the disclosure of concealable stigmas at work, and to propose a new framework for understanding this role. Employees living with PUBLIC INTEREST STATEMENT This paper deals with a sensitive issue related to the disclosure decision of employees living with concealable stigmas. It addresses a difficult situation affecting a large part of the population on a daily basis, whether they are sexual minorities, people suffering from chronic illnesses, psychological disabilities, employees of minority religions, and so on. This paper attempts to explain how people make their disclosure decision what are the conditions that could facilitate a sense of safety for individuals facing this situation. The analysis is anchored in trust, a critical condition for reducing one's sense of vulnerability. This paper suggests that given the impact of discrimination and of the emotional toll associated with hiding one's stigma, it is of utmost important to understand how to create a more supportive workplace environment.
concealable stigmas constitute a large portion of the workforce; among other groups of employees, they include Lesbian, Gay, Bisexual, and Transgender (LGBT) individuals, employees with invisible medical conditions, members of minority religions, people who experienced stigmatizing life experiences (such as rape), those suffering from a mental illness, and so on. Although precise figures on the total number of employees with concealable stigmatized identities are hard to obtain, conservative estimates using US census data suggest that LGBT individuals alone make up approximately 4.1% of the workforce (Sears, Hunter, & Mallroy, 2009), which translates to roughly 6.5 million employees in the US and 9.5 million in the EU (CIA World Factbook, 2014). Whereas people suffering from chronic medical conditions with varying degrees of severity form close to a third of the population, translating to 125 million individuals in the US, among whom nearly half have more than one condition (Anderson & Horvath, 2002).
Research indicates that learning how to create a supportive environment where employees do not have to worry about concealing their stigma can benefit both the employees and their organization (American Psychological Association, 2002;Clair, Beatty, & MacLean, 2005;Jones & King, 2014;King & Cortina, 2010). The underlying assumption is that when employees do not need to hide who they are, they feel more free, comfortable, and empowered, and as a consequence, their positive state of well-being also affects the organization as they become more engaged and productive (Clair et al., 2005;Colgan, Creegan, McKearney, & Wright, 2006;Griffith & Hebl, 2002;Jones & King, 2014). Furthermore, it has been suggested that a climate of openness fosters the psychological safety needed for self-disclosure, which generates positive individual and interpersonal psychological mechanisms that ultimately result in higher group performance (Druskat & Wolff, 2001;Murphy, Steele, & Gross, 2007;Roberge & van Dick, 2010).
Over the years, multiple studies have attempted to identify the conditions that make employees with concealable stigmas feel comfortable enough to disclose this aspect of their identity at work. Efforts aimed at providing insights into their disclosure decision have focused mainly on antecedents, such as a person's identity centrality or their level of outness in their private life; their company's HRM diversity and inclusion practices; and the social climate or legal environment (Button, 2001;Day & Schoenrade, 2000;Griffith & Hebl, 2002;Law, Martinez, Ruggs, Hebl, & Akers, 2011;Pennington, 2010;Ragins, 2008;Ragins & Cornwell, 2001). This line of study has largely overlooked the relevance of employees' trust in their supervisors and their organization, and how they are embedded in and interact with both intra and extra organizational factors. This is surprising since trust in these two referents was found to facilitate the disclosure of other types of sensitive information, including feelings, opinions, concerns, mistakes, and wrongdoing (e.g. Dietz & Den Hartog, 2006;Gillespie, 2003;Holtzhausen, 2009;Mayer & Gavin, 2005;Milliken, Morrison, & Hewlin, 2003). For instance, in her study, Gillespie (2003) found that employees who trusted their managers were more likely to discuss their original ideas with them and how they felt about their work.
Shifting the perspective from studying the impact of systems such as HRM practice and/or individual differences such as identity centrality to how these variables simultaneously interact with trust may well be significant for both theory and practice. Using this perspective can help to deal with questions, such as: How do HRM policies influence trust in the organization and trust in the supervisors? How do external antidiscrimination laws impact on this relationship? How are individual difference variables related to these relationships? For instance, several studies have found that inclusive HRM policies and practice do not necessarily produce the intended outcomes such as promoting disclosure or generating a better workplace environment for the employees they intend to protect (Botsford & King, 2008;Griffith & Hebl, 2002;Tejeda, 2006;Waldo, 1999). Yet these studies failed to incorporate the important mediating 1 mechanism that trust can have on the impact of HRM practice on employees' workplace attitudes (Aryee, Budhwar, & Chen, 2002;Deery, Iverson, & Walsh, 2006;Gould-Williams, 2003;Tan & Lim, 2009;Tzafrir, 2005). For management, learning how trust impacts on disclosure and a positive climate can provide them with additional tools that should result in a more satisfied and productive workforce (e.g. Bijlsma & van de Bunt, 2003;Hurley, 2012 Accordingly, this article intends to provide a better understanding of the disclosure of concealable stigmas by incorporating trust into this field of study. The model proposes that the interaction of trust with different personal and environmental variables will eventually affect the extent to which employees with concealable stigmas will feel comfortable when taking the risk associated with coming out. The model developed here draws on findings and theories from the fields of identity-verification (Ragins, 2008;Swann, 1983Swann, , 1996; trust (Dirks & Ferrin, 2001;Mishra, 1992Mishra, , 1996; psychological contracts (Robinson, 1996); and the disclosure of various types of sensitive information. Based on our discussion, we also make recommendations that can guide organizations in fostering this type of trust and in creating a better social environment for their employees.
To facilitate our discussion, this paper is divided into four main sections. Section 1 provides an introduction to the disclosure dilemma, Section 2 discusses the relevant literature on trust and information sharing. Section 3 is where we develop our theoretical propositions, and the Sections 4 and 5 comprises our conclusions and limitations.

The disclosure dilemma: antecedents and the consequences of coming out
I haven't come out because I fear some homophobia. I have encountered some homophobia among colleagues from other countries, and I suspect that one of the senior managers at my company may be homophobic. I don't feel great about my decision not to come out. I may decide to come out on a very limited basis in the future. (Lesbian Employee, Silva & Warren, 2009, p. 12) Unlike employees whose diversity characteristics are visible such as racial minorities, employees with concealable stigmas must decide whether to disclose or conceal their differentness ("come out" or "pass"/"stay closeted")-and then to manage their identity appropriately (Bergart, 2004;Chung, 2001;Goffman, 1963;Hill, 2009;Munir, Leka, & Griffiths, 2005;Ward & Winstanley, 2005). The decision whether or not to come out is a very difficult one in the lives of these employees due to the potential consequences of their disclosure and the invisible nature of the stigmatized identity (Clair et al., 2005;Jones & King, 2014;Quinn & Chaudoir, 2009;Ragins, 2008). This difficult decision is typically referred to as the "disclosure dilemma": On the one hand, concealing one's identity has been found to produce high levels of stress and anxiety, mainly resulting from the fear of being outed involuntary, and the constant need to conceal their stigma from co-workers (Corrigan & Matthews, 2003;Flett, 2012;Goffman, 1963;Hill, 2009;Smart & Wegner, 2000;Ward & Winstanley, 2005). On the other hand, coming out involves the risk of discrimination, harassment, social hostility, and even physical harm (Chung, Williams, & Dispenza, 2009;FRA (European Union Agency for Fundamental Rights), 2009;Jones, 2011;Sears & Mallroy, 2011;Ragins, 2008;Quinn & Chaudoir, 2009;Rocco, 2004).
The complexity of this decision results not only from its potential consequences, but also from its very nature. With the objective of avoiding potential discrimination at their workplace, individuals can be selective in disclosing their stigma, meaning they can be out to everyone, to some people, or to no one (Clair et al., 2005;Munir et al., 2005;Ragins, 2004;Rodkjaer, Sodemann, Ostergaard, & Lomborg, 2011), and to manage how much real information they share about their private lives (Bergart, 2004;Chung, 2001;Chung et al., 2009;Goffman, 1963;Griffin, 1992;Munir et al., 2005). When individuals experience ambiguity concerning the anticipated acceptance of their stigma, they initiate a "signalling" process intended to determine the risk involved in disclosure (Jones, 2013;Jones & King, 2014;Ragins, 2008).
The level of comfort that individuals feel about how much they can share about themselves has consequences not only for their personal well-being, but also for their organization. Decades of studies have indicated that the outcome of the disclosure decision has significant effects on employees' work attitudes and contributions. Employees who are out in a supportive environment establish better relationships with their colleagues, and are more committed, productive, and participative than employees who are passing or are out in a negative environment (American Psychological Association, 2002;Clair et al., 2005;Colgan et al., 2006;Day & Schoenrade, 2000;Fesko, 2001;Gignac & Cao, 2009;Jones & King, 2014;Ragins, Singh, & Cornwell, 2007;Waldo, 1999).
The importance of promoting inclusion has prompted research and practical efforts aimed at creating a more welcoming environment for employees with diverse backgrounds. For instance, increasing numbers of organizations have implemented HRM LGBT inclusion policies and practices with the intention of promoting a more supportive working environment for sexual minorities, and to comply with anti-discrimination legislation (Day & Greene, 2008). Nevertheless, research on the impact of these policies has produced mixed results. While some studies indicate that HRM policies and practice do promote disclosure and inclusiveness (Button, 2001;Law et al., 2011;Ragins & Cornwell, 2001), others have found that they may have either no or very limited effects on employees' openness and workplace attitudes and well-being (Day & Schoenrade, 2000;Griffith & Hebl, 2002). Additionally, some scholars have found that non-discrimination policies are at times actually associated with increased hostility toward the employees they are trying to protect. Numerous studies on gender, race, and sexual orientation have already demonstrated that the introduction of diversity initiatives can promote a counter reaction and backlash from the dominant majority (Bond & Pyle, 1998;Hill, 2009;Kalev, Dobbin, & Kelly, 2006;Linnehan & Konrad, 1999;Tejeda, 2006;Thomas & Plaut, 2007). Clair et al. (2005) suggested that it is not the existence of these policies, but rather the employees' confidence in their superiors' support for them that ultimately determines how comfortable they feel in coming out. Studies on the disclosure of invisible disabilities indeed underscore the importance of trust in management. As Pennington (2010) explained, even in places where employers are required by law to accommodate for employees with disabilities, many employees will be hesitant to formally "coming out" about their disability and will base their decision on the perceived reaction for their disclosure. Therefore, their level of trust in the way their organization and supervisors will use the information about their medical condition is critical for disclosure (Cunningham & James, 2001).
Studies on the disclosure of sexual orientation point in the same direction. In one of the earliest studies on the disclosure of concealable stigmas, Waldo (1999) found that HRM practices had no impact on reducing hostility towards gays and lesbians unless the organization took these issues seriously. Similarly, later studies showed that while HRM policies and practice had little or no impact on an LGBT employee's disclosure, or on workplace attitudes, the perceived degree of management supportiveness of inclusion was an important determinant (Day & Schoenrade, 2000;Griffith & Hebl, 2002).

Defining trust and the disclosure of sensitive information
The proper definition of trust is still being debated in the academic literature, and various scholars have produced different classifications for this conceptual construct (e.g. Lewis & Weigert, 1985;Mayer, Davis, & Schoorman, 1995;. Nevertheless, Rousseau, Sitkin, Burt, and Camerer (1998) concluded that, despite some areas of disagreement, there appeared to be a convergence around various key elements that comprise the construct of trust: (1) confident expectations of others, and (2) the willingness to become vulnerable or to rely on another person. Consequently, the authors proposed a definition that suggests that "trust is a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behavior of another" (p. 395). Trust, as suggested by Das and Teng (2004), reduces the perceived risk associated with the vulnerability present in the situation; the more a person trusts another to take action favorable to them, the less they perceive putting their faith in the other as being risky, and the more likely they are to take this course of action. A conscious disclosure of potentially damaging information by one party is therefore seen as risk-taking behavior, indicating trust for the other party (Dietz & Den Hartog, 2006;Gillespie, 2003). Indeed, ongoing research indicates an established relationship between employees' trust in their organizational members and their openness about information of a sensitive nature, including views, opinions, mistakes, problems, feelings, knowledge, medical conditions, and wrongdoing (Arthur & Kim, 2005;Gillespie, 2003;Lee, Gillespie, Mann, & Wearing, 2010;Mäkelä & Brewster, 2009;Milliken et al., 2003;Muthusamy & White, 2005;Zand, 1972). For instance, studies on disability show that the readiness of employees to disclose to their management their medical condition depends on the employees' track record in the company, the perception that the managers have positive attitudes towards disability, and the legal context (Ellison, Russinova, MacDonald-Wilson, & Lyass, 2003;Stanley, Ridley, Manthorpe, Harris, & Hurst, 2007). A simple illustration of how trust affects disclosure is an individual who takes a risk and shares with his supervisor that he feels underqualified for that task. The employee knows that sharing this information with a superior could make him vulnerable and hurt his career. Nevertheless, because he perceives his supervisor to be trustworthy and expects her to guide and support him, he decides to take the risk and to discuss his concerns.

Trust in the supervisor and trust in the organization
Employees develop different forms of trust in relation to proximity, the nature of the interaction, and the power relationship between themselves and the target of trust. Research on trust typically distinguishes between three levels of foci an employee trusts: (1) proximate supervisors; (2) the organization; and (3) colleagues or team (Fulmer & Gelfand, 2012). This distinction is significant as there are important differences in the form of trust at each of these levels.
The form of trust employees develop in both their direct supervisor and their co-workers is considered interpersonal (Fulmer & Gelfand, 2012) and is formed based on dyadic relationships (McKnight, Cummings, & Chervany, 1998). Nevertheless, trust in the supervisor and trust in colleagues is not the same due to the important power differential between supervisor and employee, which does not exist in the more horizontal relationships between colleagues (Schoorman, Mayer, & Davis, 2007;Tan & Tan, 2000). By contrast, trust in the organization is institutional and impersonal (Costigan, Insinga, Kranas, Kureshov, & Ilter, 2004;Mayer & Davis, 1999;Mayer & Gavin, 2005), and it addresses more general targets such as the employer (Deery et al., 2006;Robinson & Rousseau, 1994) or the top management team (Costigan et al., 2004;Mayer & Davis, 1999;Tzafrir, 2009). This form of trust is based on the perception of the institution's norms and procedures, rather than on direct one-onone experiences with others (Gillespie, Hurley, Dietz, & Bachmann, 2012), and can be referred to as "system trust" (Luhmann, 1979) or "institutional-based trust" (Zucker, 1986). Organizational arrangements (e.g. HRM) therefore function not only as coordination mechanisms, but also as a source of organizational reputation by influencing their employees' expectation, intentions, and attributions of the organization's trustworthiness. The institution's norms and procedures serve to channel social behaviors into predictable patterns; they can consequently influence the formation of interpersonal trust, yet cannot guarantee that individuals will always follow the rules and norms (Bachmann, 2011;Berger & Luckmann, 1966;Giddens, 1984).
While recognizing the importance of trust in co-workers, our discussion will focus on employees' trust in their supervisors and the organization (institution) simultaneously. There are two main reasons for doing so. The first is that our intention is to focus on disclosure at the more formal level, and not the interpersonal one. As discussed previously, the relationship between trust and the disclosure of stigma in the context of intimate personal relationships has already been established in the literature (Boon & Miller, 1999;Cain, 1991;Derlega et al., 1998;Obermeyer et al., 2011). Second, both the organization and its direct supervisors hold formal roles of responsibility and power over their employees; they are the ones who signal the desired role behaviors in the organization, and are considered critical for the implementation of diversity programs and the creation of psychosocial safety (Cox, 1994;Dollard & Bakker, 2010;Roberge, Lewicki, Hietapelto, & Abdyldaeva, 2011;Zohar & Luria, 2005). As explained by , the signals employees receive from senior leaders, HRM policies, and management are critical for the successful implementation of a climate for inclusion. Also, as found by Dollard and Bakker (2010), the importance senior leaders give to their employees' psychological well-being helps create a psychological safety climate, which in turn affects employees' psychological working conditions, health, and engagement. Furthermore, evidence implies that managers have an important role in providing emotional and practical support for those disclosing and making themselves vulnerable at work (Gignac & Cao, 2009;Huffman, Watrous-Rodriguez, & King, 2008;King & Botsford, 2009;Munir, Pryce, Haslam, Leka, & Griffiths, 2006;Munir, Randall, Yarker, & Nielsen, 2009). For instance, Gignac and Cao (2009) found that managerial support helps to reduce the stress employees with arthritis experience following the disclosure of their medical conditions. Accordingly, Munir et al., 2005 found that employees with chronic illnesses are likely to disclose their full condition to their line managers if they consider receiving support from their supervisor in relation to their chronic illness as important. Along similar lines, studies have found that the support supervisors can provide women who decide to disclose intimate partner abuse, helps buffer the effects these negative experiences may have on the victim's employment (Perrin, Yragui, Hanson, & Glass, 2011;Swanberg, Logan, & Macke, 2005).

Disclosure: trust and its interaction with intra and extra organizational factors
Coming out at work is a voluntary act of sharing information that increases one's vulnerability. Accordingly, the purpose of this section is to discuss how trust is embedded in intra and extra organizational contexts, and to determine the expected outcome of the interactions between trust and the other critical antecedents of disclosure. Figure 1 summarizes the framework that we have developed. We begin by developing research propositions and discussing the interaction between trust and HRM policies and practices, then the interaction of trust with individual-level variables, and finally its interaction with the legal context.

HRM policies, trust and disclosure
A valuable starting point for understanding disclosure in an organizational context is to consider the ways in which protective HRM policies and practices impact employees' willingness to take the risk associated with coming out. As disclosure of stigma involves risk, employees are likely to look for ways to assess whether or not they can make themselves vulnerable. The implementation of inclusive HRM policies can provide employees with some indication as to the degree they can trust their organizations and managers. HRM policies and practices can be classified into the following three main roles: Motivation, Support, and Symbols. The motivation role focuses on stimulating, directing, and maintaining employees toward attaining a specific goal (Greenberg & Baron, 2008), such as an open and fair environment. The support role pertains to the efforts made by organizations to help employees to do their jobs by reducing obstacles and supplying resources such as employee assistance programs. Finally, symbols in the workplace make up the third role and look at the existence of formal and informal attitudes, behaviors, and procedures for resolving difficulties, predicaments, and dilemmas in the workplace, such as discrimination. Not surprisingly, Delaney and Lundy (1996) suggested that one of the hallmarks of equitable HRM systems is their enactment of objective standards that remove bias and subjectivity in the implementation of HRM practices. With these objectives in mind, different organizations have, over the years, implemented a variety of HR policies and practices that are intended to create a more inclusive, fair, and safe working environment for their employees. These practices include non-discrimination policies, Employee Resource Groups, guidelines for inclusive communication, and diversity awareness training (Chrobot-Mason, 2003;Day & Greene, 2008;Johnston & Malina, 2008;King & Botsford, 2009). Obviously, HRM policies and practices, as well as anti-discrimination laws, do not cover all sources of stigma and bases for discrimination. These mechanisms tend to address specific groups or categories of individuals (sexual minorities, employees with disabilities, and so on), and so leave unprotected others who may be stigmatized due to attributes, such as victims of rape or parents of drug addicts (Kulik, Bainbridge, & Cregan, 2008;Paetzold, Dipboye, & Elsbach, 2008).
Although these HRM policies and practices can build, develop, and maintain trust in an organization and its supervisors, they do not operate in a vacuum. Employees tend to interpret these policies and practices together with their attributions and the perceptions of their managers' behavior (Searle, 2013;Skinner, Saunders, & Duckett, 2004;Tzafrir, 2005;Weibel et al., 2009;Whitener, 2001). As noted by Clair et al. (2005) it is not the existence of policies or practices that will promote disclosure; rather, it is the reassurance that their management provides them with. For instance, research on the impact of LGBT inclusion and HRM systems appears to imply that the existence of managerial support for LGBT employees is a necessary condition for their success in reducing heterosexism and facilitating disclosure (Day & Schoenrade, 2000;Waldo, 1999). Similar conclusions were reached when considering the way family-friendly policies can make pregnant workers feel more comfortable when taking the risks associated with the disclosure of early-stage pregnancy. King and Botsford (2009) pointed out that as managers can override organizational family-friendly policies in both supportive and destructive ways, managerial support for these policies is important to ease the fears associated with disclosure. Similarly, Lewis (2011) argued that while organizations can encourage employees to report wrongdoing by introducing policies and procedures, much more important is the employees' experience of what happens when concerns are raised. Finally, research has shown that when organizations were auditing disability, some employees were reluctant to disclose their disability, apparently due to not trusting how this information would be used (Cunningham & James, 2001).
A large body of data propose that employees' trust in their organization and/or their supervisors acts as a mediator between HRM practices and employees' workplace behaviors and attitudes (Aryee et al., 2002;Chen, Aryee, & Lee, 2005;Lee et al., 2010;More & Tzafrir, 2009;Searle & Dietz, 2012). For instance, a study by Chen et al. (2005) found that perceived organizational support influenced employees' level of trust in their organization, which in turn impacted their role performance and commitment. In an earlier study by Gould-Williams (2003), HRM practices related to selection, training, and job design were found to predict organizational trust and interpersonal trust. This high level of trust consequently contributed to the employees' level of satisfaction, commitment, and overall organizational performance. Finally, a recent study by Seifert, Stammerjohan, and Martin (2014), suggests that trust in the supervisor and organization mediates, or intervenes, between various forms of organizational justice and employees' readiness to disclose wrongdoing.
Research shows that HRM policies and practices have the potential to generate trust in both the organization and the supervisors Tan & Tan, 2000;Whitener, Brodt, Korsgaard, & Werner, 1998). The ways these organizational systems generate trust in these two areas of trust are related, yet distinct. HRM policies and practices can foster organizational trust by signaling a message of support and commitment to all employees, by creating a sense of certainty and security, and through a sense of fairness and inclusion of professionalism (Mayer & Davis, 1999;Tzafrir, Baruch, Dolan, & Dolan, 2004;Vanhala & Ahteela, 2011). In that sense, the time and effort HR personnel and senior leaders invest in developing well-crafted policies and practices can signal to those employees who are at risk of discrimination that they can trust their organization.
The way in which HRM systems influence trust in the supervisors may follow a different route. An HRM philosophy (Welbourne & Andrews, 1996) and the existence of HRM policies and practices can promote an environment and conditions that engender trust between employees and their supervisors Whitener et al., 1998). For instance, managers trained in inclusion, and who follow anti-discrimination policies and fair procedures are likely to act in a way that will increase their employees' sense of confidence in them. It is then that these organizational practices and policies have the potential to create a positive employee social environment (Tzafrir, Gur, & Blumen, 2014) where employees feel comfortable about "taking the leap of faith" (Möllering, 2006) involved in trust . More specifically, in the case of employees with stigmatized identities, the integration between HRM practices and policies with trust enables one to take the risks associated with disclosure.
One way to explain the way these HRM policies and practices influence employees' trust and their consequent disclosure is by viewing them through the lens of psychological contracts (Rousseau, 1989). The usefulness of the theory for our context is that they can explain the inconsistent way HRM policies and practices influence disclosure. All in all, HRM policies and practices build and develop employees' expectations, as well as creating more reciprocation in terms of their interaction with their organization and managers (Tzafrir, 2005).
A psychological contract is an individual's beliefs about the terms and conditions of a reciprocal exchange agreement between that person and another party (Rousseau, 1989). Studies show that employees' trust and commitment depend on their perception of how their employer has met the expectations they have set with their policies (Deery et al., 2006;Robinson & Rousseau, 1994;Skinner et al., 2004). The explanation for this process lies in the essence of trust, which is one's expectations concerning the treatment he or she receives from the other party (Robinson, 1996). The enactment of HRM policies, in terms of content and implementation, is likely to affect employees' perceptions concerning the extent that their organization adheres to their implicit and explicit contract and that it can be trusted . A study by Chrobot-Mason (2003) on racially diverse employees helps illustrate this point. The research showed that when organizations set up diversity initiatives, they create expectations ("diversity promises") concerning equal treatment. When these expectations are not met, they generate a psychological contract breech that produces cynicism and lowers employees' commitment and job satisfaction.
The way HRM policies and practices are implemented can offer the employees tangible evidence of the degree to which the organization and its supervisors' intentions are genuine and can be trusted . It is therefore likely that when an organization introduces diversity policies and practices, it creates expectations concerning organizational and managerial behavior. The perceived success of HRM programs to promote a more inclusive environment will therefore impact the level of trust in the organization and management, and consequently employees' readiness to come out. Employees who believe that their employer adheres to its stated HRM inclusion policies will feel the organization can be trusted Skinner et al., 2004), which will increase the likelihood for disclosure. On the other hand, if an employee perceives their employer as not having fulfilled the set expectations, trust decreases, and so does the employee's readiness to take the risk associated with making their stigma known.
Taking the Lewicki and Bunker (1996), one may say that the perceived success of the stated HRM policies serve as a knowledge base for employees' trust-based decisions. The theory of managerial decision-making underscores that the complexity of the decision process increases with uncertainty (e.g. Eisenhardt & Zbaracki, 1992). To arrive at an optimal decision, individuals need to examine thoroughly all the alternatives. Increased uncertainty enhances the complexity of each of the alternatives to a point that individuals may not be able to efficiently explore the trade-offs between the costs and benefits (Bingham, Eisenhardt, & Furr, 2007;Busenitz & Barney, 1997). Thus, trust may serve as a cognitive mechanism reducing the uncertainty and facilitating disclosure decision. Taking all of the above into consideration, trust reflects employees' decisions to make themselves vulnerable at work based on their assessment of their organization's and managers' commitment to inclusion (Clark & Payne, 1997;Dietz & Den Hartog, 2006). Accordingly, it is assumed that when HRM systems serve to promote the inclusion of employees they are meant to protect, they foster a trust that will eventually promote disclosure. However, when they fail to do so, for example, when employees notice that the management does not curb anti-gay talk, or when employees who disclose their disability end up experiencing unfair treatment, trust will decrease and so will their willingness to take the risks associated with coming out.
Proposition 1a: Trust in the organization will mediate the relationship between HRM inclusion policies and practices and the stigma disclosure of the protected employees.
Proposition 1b: Trust in the supervisors will mediate the relationship between HRM inclusion policies and practices and the stigma disclosure of the protected employees.

Trust and individual differences
In addition to the mediating role that trust can play in the disclosure decision, it can also play an important moderating 2 role on individual differences when predicting disclosure. Studies on different types of disclosure (whistleblowing, work-related information, minority religions, etc.) have suggested that trust can be expected to moderate the relationship between individual variables that predict disclosure, such as the centrality of one's identity, motivation to cooperate, or level of selfesteem, to the disclosure itself (Dirks & Ferrin, 2001;Fleig-Palmer & Schoorman, 2011;Hecht, Faulkner, & Faulkner, 2000;LePine & Van Dyne, 1998). Findings from over two decades of scholarly work have demonstrated two important individual-level antecedents of disclosure: (1) an individual's self-view, which refers to the level of identification and comfort with their stigma; and (2) outness, which is the degree to which people are open about their stigma in their private lives (Clair et al., 2005;Corrigan & Matthews, 2003;Friskopp & Silverstein, 1996;Griffith & Hebl, 2002;Huffman et al., 2008;King, Reilly, & Hebl, 2008;Law et al., 2011;Ragins, 2008). These two individual variables are closely linked, and show people's motivation to affirm their identity and achieve congruence or harmony across different areas of life (Friskopp & Silverstein, 1996;Kwang & Swann, 2010;Ragins, 2008;Swann, 1983Swann, , 1996. Simply put, what identity theory implies is that people who feel positive and strong about who they are, and who are open about it to their family and friends, will have a stronger need to be out at work than those who do not. On the other hand, people who experience their stigmatized identity as less important or positive, and who are in the closet in their private lives, have less inclination to come out at work. The importance of individual variables such as identity has been repeatedly validated in studies on various concealable stigmas. For instance, an important precursor phase to a person's readiness to disclose that they are HIV-positive is their ability to construct a new personal identity as "a person living with HIV" (Rodkjaer et al., 2011). Similar findings have even been found in research exploring people's readiness to disclose medical conditions of a lesser severity, such as asthma (Adams, Pill, & Jones, 1997). The literature on whistleblowing also provides ample support for the relevance of psychological antecedents in deciding whether to take the risks associated with disclosing wrongdoing. As the essence of whistleblowing is the reporting of illegal, immoral, or illegitimate practices, scholars point out that one's level of moral identity predicts the decision to blow the whistle (Liyanarachchi & Newdick, 2009;Miceli, Near, & Schwenk, 1991;Vadera, Aguilera, & Caza, 2009). These individual propensity variables were even found to have higher association with the act of whistleblowing than did organizational propensity variables (Cassematis & Wortley, 2013). Hence, unsurprisingly, people who see standing up for morality as something important in their lives will be more inclined to take the risks associated with retaliation for reporting wrongdoing. The question therefore revolves around whether and how these individual variables or motivators interact with an employee's trust in the decision to come out. Dirks (1999) and Dirks and Ferrin (2001) conclude that trust impacts individual outcomes and behaviors at work by moderating the effects of motivational constructs. This line of thought implies in our context that trust is supposed to alter the relationship between the independent motivational variables, that is identity, outness, and disclosure (Hayes, 2012). One explanation is provided by Fleig-Palmer and Schoorman (2011) in their study on trust and knowledge transfer in mentoring relations. The authors propose that, although for mentors who are highly motivated to participate in the transfer of knowledge, trust is hardly needed to prompt them to action, for mentors with little motivation to transfer their knowledge, having a trusting relationship with the mentee is vital in determining their level of collaboration in the knowledge transfer process. These claims are supported by studies on other forms of disclosure such as religious identity. A study by Hecht et al. (2000) on the disclosure of Jewish identity among Jewish-Americans leads to similar conclusions. They mention that for some of the people they interviewed, their Jewish identity was so strong that they simply could not conceal it, and even made it clearly visible to others by wearing Jewish symbols such as the Star of David. For others, for whom their religious identity played a minor role, disclosing their Jewish faith depended to a large degree on their level of comfort with the target of disclosure.
What these studies suggested is that the way individual variables impact disclosure is not necessary linear; instead, trust may moderate these patterns of relationship. Accordingly, it is expected that the two foci of trust will moderate the way central individual variables, outness in private life and self-view, predict employees' readiness to come out at work. For employees whose stigma does not form an important part of their identity, or who are not completely open about this aspect of who they are in their private lives, trust in their line managers and organization may make an important difference as to whether or not they are out at work, and may even prompt them to come out. On the other hand, the more a person views their identity as important and positive, or the more he or she is out in their private lives, trust in their organization and trust in their managers will be less significant. Therefore, we propose the following: Proposition 2a: Trust moderates the relationship between self-view and stigma disclosure, such that a stronger trust in their organization reduces the impact of employees' self-views on disclosure.
Proposition 2b: Trust moderates the relationship between outness and stigma disclosure, such that a stronger trust in their organization reduces the impact of employees' outness in their private lives on disclosure.
Proposition 2c: Trust moderates the relationship between self-view and stigma disclosure, such that a stronger trust in their supervisor reduces the impact of employees' self-view on disclosure.
Proposition 2d: Trust moderates the relationship between outness and stigma disclosure, such that a stronger trust in their supervisor reduces the impact of employees' outness in their private lives on disclosure.

Differing levels of trust between trust in the supervisor and trust in the organization
Although the employees' level of trust in their organization and supervisors can be similar, it can also be distinct (Konovsky & Pugh, 1994;McEvily, Radzevick, & Weber, 2012). An employee, for instance, can trust her supervisor based on the good relationship they have, while at the same time not trust the organization based on other observations (Tan & Tan, 2000). As people can be selective concerning whom they come out to and how open they want to be about their stigma, it is likely that trust in the supervisor and trust in the organization will impact the disclosure patterns (Chung et al., 2009;Clair et al., 2005). Research so far provides little indication as to how differing levels of trust might impact disclosure. While studies shows that both targets of trust have a role in promoting diversity and psychological safety at work, such studies failed to contemplate the impact of contradictory levels of trust (Cox, 1994;Day & Schoenrade, 2000;Dollard & Bakker, 2010;Huffman et al., 2008;Singh, Winkel, & Selvarajan, 2013;Stoner & Russell-Chapin, 1997;Waldo, 1999).
While research on related fields such as hostility and harassment at work may provide some relevant insights, studies addressing this topic have failed to address this issue. There is wide evidence indicating that employees have grounded reasons to fear coming out to their supervisor when they cannot trust their reaction, even when working in an overall positive organizational climate. First, due to elements of power in the relationship, mistreatment or hostility by the supervisor tends to take a heavy toll on the employees' well-being, especially when an employee is singled out (Huo, Lam, & Chen, 2012;Langan-Fox, Cooper, & Klimoski, 2007;Tepper et al., 2009). Then, despite the existence of policies and grievance procedures, many employees will still be reluctant to call for support due to fear of reprisal by the supervisor or colleagues (Denissen, 2010;Pershing, 2003;Vijayasiri, 2008). Such a finding suggests that employees can trust their organization and find it supportive, but at the same time not trust their supervisor. In this case, their tendency will be to disclose to formal organizational members (e.g. HR), yet avoid sharing the same information with their supervisor and team. Therefore, we propose that: Proposition 3a: When employees trust their organization but not their supervisor, their tendency will be disclosure to formal organizational members and, yet avoid disclosure to their supervisor and team.
On the other hand, the opposite situation can also occur. For instance, a lesbian employee may not trust that her organization will treat her fairly if she discloses her sexual orientation, yet she knows her supervisor well and feels that he is trustworthy. Under this condition, we suggest that: Proposition 3b: When employees trust their supervisor but not their organization, their tendency will be disclosure to proximate organization members, yet avoid disclosure to more formal organizational members.

Legal protection and trust
Legislative bodies are one of the most important and visible fronts in the efforts to achieve greater inclusion and equality at work. Overall, the trend in many Western developed countries suggests increasing legal protection for groups vulnerable to discrimination, such as members of minority religions or races, women, employees with a disability, LGBT individuals, and older employees (Barron & Hebl, 2010). Unsurprisingly, the available research has shown that legal protection has a positive impact on employees, who are thus more likely to come out (Ragins & Cornwell, 2001). The enactment of anti-discrimination laws adds an additional dimension as these laws are "imposed" on the organization by the external legal system (King & Botsford, 2009).
Nevertheless, although there is some overlap between legal requirements and HRM policies and practices, the two are not the same (Linnehan & Konrad, 1999). For instance, in places where there is protective legislation, organizations can take a passive compliance approach that centers on statutory requirements and policy mandates, which implies acting in accordance with the letter (but not the spirit) of the laws or standards (Miller, 1994;Rocco, Landorf, & Delgado, 2009). The organization in this case merely upholds the law in a passive way that does not signal a true commitment to inclusion. In fact, in organizations that only express passive compliance with the law, employees are still at risk of being discriminated against, and are recommended to carefully consider whether they should make a disclosure (Rocco et al., 2009).
While anti-discrimination laws have been found to contribute to an individual's willingness to come out at work, legislation alone is not likely to satisfy all of the conditions necessary to make employees feel comfortable about coming out, as employees can still remain vulnerable (Beatty & Kirby, 2006;Ragins & Cornwell, 2001;Searle, 2013). Studies on the disclosure of disabilities show that while protective legislation appears to give employees more confidence in disclosing their condition, many will still hesitate from doing so, even at the expense of receiving accommodation (Cunningham & James, 2001;Pennington, 2010;Vickers, 1997). Indeed, research has shown that discrimination can be still present even in places where protective legislation does exist (FRA (European Union Agency for Fundamental Rights), 2013; Drydakis, 2009;Sears & Mallroy, 2011), resulting in a negative employee social environment (Tzafrir et al., 2014). Nevertheless, both scholars and experts predict that over time, the enforcement of anti-discrimination laws will create a more equal and safe environment (Barron & Hebl, 2010;Beatty & Kirby, 2006;Budgett, Ramos, & Sears, 2008).
What makes the discussion on anti-discrimination laws interesting in our context is that they are meant to offer protection, which is not at the discretion of an organization's management. This means that both the foci of trust, even if not intrinsically supportive of inclusion, may be required by law to offer equal treatment to their protected employees. Employees, although not fully protected from discrimination, benefit from both an expected lower probability of experiencing discrimination and from protective legal mechanisms if needed. Recent court decisions ordering compensation on the basis of discrimination are likely to reassure employees about the protection available to them and to deter management from tolerating discrimination against others (Diamond, 2008;EEOO, US Equal Employment Opportunity Commission, 2013;Stacy, 2013). Furthermore, anti-discrimination legislation is suggested to serve as both an indicator of already more acceptance of broader social norms toward a given group, and as a symbolic and instrumental mechanism for further reducing prejudice and discrimination (Barron & Hebl, 2010).
From the employee's point of view, anti-discrimination legislation offers protection and reassurance, thus reducing the risks associated with disclosure. Because trust and risk are interdependent (Gambetta, 1988;Mayer et al., 1995;Rousseau et al., 1998), the reduced risk is expected to make employees' trust in their superiors and organization less relevant in their disclosure decision. Therefore, we propose that: Proposition 4a: Legal Protection and trust in the organization have an independent moderating effect on the individual's antecedents for disclosure: the presence of laws protecting against discrimination reduces the impact of trust in the organization on disclosure.
Proposition 4b: Legal Protection and trust in the supervisor have an independent moderating effect on the individual's antecedents for disclosure: the presence of laws protecting against discrimination reduces the impact of trust in the supervisor on disclosure.

Implications for research and practice
The main contribution of this article is a theoretical framework that helps us to understand the role that employees' levels of trust in their organizations and supervisors plays in their disclosure decisions. Although individual differences and HRM policies and practices are important variables in predicting employees' disclosure, the way they impact disclosure is suggested to depend mainly on the level of trust that the organization and its managers develop with their employees. Our framework also suggests that the level of trust interacts with the macro-level legal context that exists outside of the organization.
Future research could help us to test and expand our understanding of the effects of trust in various ways. First, we can learn more about the construct of trust by examining how trust in one's superiors relates to information sharing, not just in the domain of work-related information, but also through interactions that are relevant to employees' willingness to share sensitive, potentially stigmatizing personal information. One way this can be done is by adapting generic scales and modifying them to the specific situation (e.g. generic trust in the supervisor's ability versus trust in his ability to support diverse employees) and by testing how this trust interacts with personal and environmental variables.
Current research models that seek to explain how employees decide whether to come out at work typically measure the impact of individual and situational antecedents on disclosure in a relatively linear and direct way (e.g. Day & Schoenrade, 2000;Law et al., 2011). Our model suggests that these factors do not have such a direct effect; instead, they influence disclosure following their interaction with trust. Future research could examine the moderating and mediating role trust is expected to play between the individual and system-level variables and the decision to come out in this complex decision-making process. This research could help create the necessary knowledge to drive changes in organizations. An interesting venue for research could be the underlying mechanisms between the individual variables to disclosure at work, and how they interact with trust in an organizational context. This line of research could examine, for instance, the way "or disposition to trust, an individual trait reflecting a person's likelihood of trusting others, impact disclosure in both their private and work life" (Colquitt, Scott, & LePine, 2007;Mayer et al., 1995;. As this tendency to expect others "to be trustworthy also influences trust in institutions and managers it is possible it can explain why people who disclose their stigma in private life also tend to do so at work" (Clair et al., 2005;Kee & Knox, 1970;Lucassen & Schraagen, 2012;Nambudiri, 2012;Ragins, 2008).
Third, similar recommendations could be made for the growing research on the disclosure at work of stigmas in general, and sexual minority identity specifically. Future research could look into the way trust in leadership impacts the success of HR diversity programs targeting different groups of employees. The researchers Cox (1994), Guillaume, Dawson, Priola, et al. (2013), Miller (1994, Rocco et al. (2009) all agree that the commitment of senior management to diversity is a key factor for making real progress in this area. A key reason is the fact that the positive benefits of diversity programs depends on fostering cultural change which is conducive to an inclusive work environment (Avery & McKay, 2010;Guillaume, Dawson, Priola, et al., 2013;. Failure to foster this cultural change can result in immense human and economic damage (Gonzalez, 2010). These conclusions should be absolutely critical in the case of many employees in terms of living with concealable stigmas as a key premise to the success of diversity programs, which is their ability to create a climate in which they feel comfortable enough to come out and to be safe once they do. In the future, studies could then examine the element of trust in the context of a diversity climate, and how it impacts the success of HRM inclusion programs that target employees with concealable stigmas.
The relationships presented in this paper could be examined by analyzing the data obtained from different groups of employees (LGBTs, employees with invisible disabilities, and so on). Due to the sensitive nature and the concealable nature of this research area, reaching a diverse pool of individuals who are working in different organizational settings could be done, for instance, by sending out an invitation to the study via different community networks. Scales measuring the different variables, such as outness in private, identity, HRM programs and policies, and disclosure at work, could be obtained from previous studies (Day & Greene, 2008;Griffith & Hebl, 2002;Huffman et al., 2008;Waldo, 1999). With regard to trust, it will be important to adapt a generic trust scale (see, e.g. Mishra, 1996) to the specific context of the study. The researchers must of course guarantee anonymity. A list of measures can be found at Appendix 1 of this paper.
Finally, the research could also examine the way trust develops, or fails to develop, over time in this specific context. Trust is dynamic and evolving. In this paper, we discussed how it develops by using the lenses of psychological contract theory and knowledge-based trust (Lewicki & Bunker, 1996;Robinson, 1996). Whether we use this specific lens or other ones, such as those that divide trust into different forms or phases (Jones & George, 1998;Lewicki & Bunker, 1995), the conclusions are the same: trusting relationships seem to develop over time based on personal experience as the target of trust. In the case of concealable stigmas, there is one critical phase, which is the disclosure of identity. Once disclosed, it is impossible to undisclose. Further studies could therefore examine how this process of the creation of trust happens in this specific context. Any such studies should take into consideration the ethical aspects and sensitivities related to this research topic. As the subjects might not have disclosed their stigma to some of their social groups or at work, the research should guarantee that no identifiable data will be shared at any stage. Furthermore, all personal details must be kept secured and be available only to the researchers themselves. These conditions should be specified clearly in the invitation letter. Researchers should understand that exposing such data could expose their subjects to discrimination and even legal risks, for instance, due to laws criminalizing homosexual relationships.
From a practical perspective, it is argued that a greater understanding of how trust affects coming out at work can help foster an inclusive environment in which employees with concealable stigmatized identities are more healthy, comfortable, and productive.
The discussion in this paper highlighted the motives for management to increase the level of trust that employees have in their organization and supervisors. Insights from the extensive body of literature on trust have served to guide these efforts. As trust is formed based on various dimensions (Dietz & Den Hartog, 2006), organizations are advised to focus their efforts on strengthening these relevant areas. One framework that appears very pertinent to our discussion is Mishra's (1992Mishra's ( , 1996 four-dimensional model, which suggests that trust is formed based on the dimensions of care, competence, openness, and reliability. Through using this framework, organizations are urged to improve their level of trust by making progress in each of these dimensions. For example, the dimension of competence can be enhanced by managers demonstrating knowledge and professionalism in the way they address issues, such as disability and sexual or religious diversity. The dimension of care can be advanced by stating that support is not conditional on one's diversity attributes, for instance, by showing interest in the well-being of their employees while clearly signaling their acceptance of who they are, and by taking concrete action to protect them against any form of hostility and discrimination. Progress can be made in the dimension of openness by being transparent about the criteria for selection and promotion, and even by discussing one's own vulnerabilities. Finally, reliability can be strengthened by maintaining consistency in applying all of the above. These indications of trustworthiness should be made clearly visible in order to reduce the level of ambiguity concerning the possible consequences of disclosure. At the same time, managers should be sensitive to possible signaling cues by employees, who might be "testing the water" in a subtle way before deciding to come out. When this signaling behavior is detected, managers should respond in a way that leaves no ambiguity concerning their level of support.

Conclusion and future research
A promising avenue for future research would be to further investigate the differentiation between general trust and the dimension of trust (or antecedents to trust), thus conceptualizing trust in a more comprehensive manner.
In our article, we have looked at the effects of general trust by the organization and supervisors on the decision to disclose by the employee. However, studies have revealed several conditions/antecedents/dimensions for trust (e.g. see Mayer et al., 1995), and these dimensions may have various impacts on employees' disclosure behaviors. For instance, Mishra's four-dimensional model suggests four factors that affect judgments on the trustworthiness of an organization. Thus, a call to incorporate a contextual (Johns, 2006) as well as a dimensional (Mayer et al., 1995) manner into this line of research is highly important. From this perspective, would trust in an organization based on the perceived openness of the management have a different effect on disclosure than trust based on the perceptions of competence? In fact, we need to better understand how the various dimensions of trust shape the employees' behavior in greater detail, especially if they may suggest different relationships.
This article contributes to the research on trust by broadening our understanding of how trust influences the willingness of employees to become vulnerable by sharing sensitive personal information, and how this decision and its outcomes affect both the employees and the organization. We extended the notion by Tzafrir et al. (2014) about the employment social arena as a "microfoundation" (Barney & Felin, 2013) representing individuals' characteristics, various forms of social interactions, and the process dynamics involved (Ferris et al., 1998) in different levels of organizational analysis." Integrating the growing evidence supporting the views that environmental (the legal system) and organizational (the HRM system) variables have on disclosure decisions, we demonstrated the important role of the two foci of trust when making these decisions. These new insights can promote future research and provide organizations with useful practical information on how to improve their management and HR practices. Research and theory suggest that the outcome of these changes can contribute to both the well-being of the employees and to the performance and competitiveness of the organizations in which they work.  . Trust Questionnaire.
The following instructions prefaced the scales. Indicate the degree to which you agree with each statement by using the following scale.
(1) Disagree strongly Employees/managers will keep the promises they make (7) Employees/managers really look out for what is important to the managers/employees (8) Employees/managers have a lot of knowledge about the work that needs to be done (9) Employees/managers are known to be successful in the things they attempt to accomplish (10) If I make a mistake, my employees/managers are willing to "forgive and forget" (11) Employees'/managers' actions and behaviors are not consistent. (R)