Introduction

Hong Kong’s laws have been reformed significantly to protect the rights of sexual orientation minorities. In 1991, Hong Kong’s Legislative Council decriminalized sodomy. More recently, sexual orientation rights have advanced through case law. In 2006, the Hong Kong Court of Appeal held that Hong Kong’s disparate age-of-consent laws for anal and vaginal sex violated the Basic Law and Bill of Rights (Leung v. Secretary for Justice). In that case, the court set important legal precedent, treating sexual orientation discrimination as a matter equally as grave as discrimination on the bases of race and sex (Lau 2008). In 2007, the Hong Kong Court of Final Appeal applied similar reasoning to hold that, by punishing public indecency by same-sex couples and opposite-sex couples unequally, the criminal code violated the Basic Law and Bill of Rights (Secretary for Justice v. Yau Yuk Lung Zigo and Another). Finally, in 2008, the High Court of Hong Kong held that the government broadcasting authority impermissibly discriminated when it objected to a television documentary on same-sex couples (Cho Man Kit v. Broadcasting Authority).

Despite these protections against governmental discrimination, there remains no legal protection against sexual orientation discrimination in the private sphere, including private-sector employment. This article furthers the understanding of sexual orientation-based employment discrimination in Hong Kong and its public policy implications. Specifically, utilizing survey data from 792 sexual orientation minorities in Hong Kong,Footnote 1 we address three main questions: First, how prevalent is sexual orientation discrimination in Hong Kong? Second, who is most vulnerable to sexual orientation discrimination? And, third, what is the relationship between experiences of sexual orientation discrimination and self-reports of psychological outcomes?

Although there currently is no legal prohibition of sexual orientation-based employment discrimination in the private sector, such legislation has been proposed in Hong Kong. The first major attempt to pass such legislation occurred in 1994, when legislator Anna Wu proposed a comprehensive antidiscrimination law that would have included sexual orientation. To date, Hong Kong has passed antidiscrimination laws covering discrimination on the grounds of sex (1995), disability (1995), family status (1997), and race (2008)Footnote 2 (e.g., Chan 2005; Kapai 2009; Petersen 1997; To 2004), but not sexual orientation. Hong Kong’s Equal Opportunities Commission (EOC) applies the Disability Discrimination Ordinance to certain complaints based on individuals’ transgender status (Shamdasani 2004; Yau 2008); however, the EOC has also been criticized for handling these cases insensitively (Yau 2008).

In 2001, Hong Kong’s Legislative Council established a Subcommittee to study sexual orientation discrimination, but the Subcommittee never tabled a bill for consideration. The Hong Kong government has, however, launched public education campaigns to confront sexual orientation discrimination, issued non-binding declarations against sexual orientation discrimination in the workplace, and established its Gender Identity and Sexual Orientation Unit (GISOU). GISOU takes complaints of sexual orientation and gender identity discrimination and seeks to mediate claims of discrimination. However, GISOU does not offer enforceable remedies (Lau 2008).

There has been growing pressure on the Hong Kong government to pass legislation banning employment discrimination based on sexual orientation. Advocacy for sexual orientation rights in Hong Kong has become more visible in recent years. For example, since 2005, there have been annual gay rights marches with slogans including calls for antidiscrimination legislation (Shamdasani 2005a; Wong 2007). Pressure has also come from outside Hong Kong. The United Nations Human Rights Committee (1999) and the United Nations Committee on Economic, Social, and Cultural Rights (2001, 2005) have chided Hong Kong for failing to enact a sexual orientation antidiscrimination law and their criticisms received considerable press coverage in Hong Kong (e.g., Hong Kong Economic Times 2005; Lee 2004; Shamdasani 2005b). Similar criticism has been launched within academic circles (e.g., Chan 2005; Petersen 1997). This mounting pressure for reform begs a better understanding of how sexual orientation discrimination manifests in Hong Kong. This article helps to fill the gaps in knowledge on sexual orientation-based employment discrimination in Hong Kong.

In the remainder of this article, we begin by providing background on existing research. We then discuss the hypotheses, methodology, and findings of our online survey on discrimination experienced by sexual orientation minorities in Hong Kong. Finally, we discuss our findings’ implications for public policy and provide directions for future research.

Existing Research

In light of the push for antidiscrimination legislation, research on sexual orientation discrimination is useful not only to scholars of discrimination, but also to policymakers. Indeed, the government of Hong Kong itself has signaled an interest in better understanding sexual orientation discrimination within its jurisdiction. In this section, we provide background on the few existing studies of sexual orientation discrimination in Hong Kong, including a recent survey commissioned by the Hong Kong government. We also briefly explore research conducted in other parts of the world.

Hong Kong-based Studies

Empirical research on sexual orientation discrimination in Hong Kong is nascent. In this section, we discuss five notable studies, the first of which was a telephone survey commissioned by the Hong Kong government in 2005. The government stated that two purposes of its research were to “assess the extent of the problem faced by homosexuals” and “to examine what can be done to address the problem of discrimination faced by homosexuals” (Hong Kong Home Affairs Bureau 2006). A random sample of Hongkongers was surveyed, but the study did not ask about each respondent’s sexual orientation. Because the study did not disaggregate responses from sexual orientation minorities from other responses, the government’s findings potentially obscure the actual situation in Hong Kong.

Of the 2,040 respondents who participated in the survey 29.7% of respondents thought, overall, discrimination based on sexual orientation was a “very serious/serious” problem in Hong Kong and 41.7% thought the problem was “average.” Less than one-third of the respondents thought that discrimination was a serious problem with regard to employment, flat rental, or membership in private clubs. The perception of respondents is, however, likely to be a poor barometer for the seriousness of sexual orientation discrimination because 76.5% of the respondents reported having had no contact with homosexuals. Notably, the 4.8% of respondents who reported frequent contact with homosexuals were more likely to perceive sexual orientation discrimination as a serious problem (41.1% of those with frequent contact compared to 29.7% of all respondents). Even those respondents who reported frequent contact with homosexuals might be a poor barometer for the severity of sexual orientation discrimination because they are at least one step removed from suffering incidences of discrimination, unless they themselves identify as homosexual.

The government study also asked respondents about their attitudes toward homosexuality. Only 47% of all respondents thought that homosexuality is “psychologically normal,” and 38.9% thought that homosexuality “contradicts the morals of the community.” One should note that individuals do not always form their views on civil rights and treatment of others based on moral attitudes (Brewer 2003).Footnote 3 Although 38.9% of respondents in Hong Kong thought that homosexuality contradicts morals, the large majority of respondents stated that they were accepting of homosexual colleagues (79.9%), superiors at work (77.5%), and neighbors (78.0%); only 40%, however, were accepting of homosexual family members.

In addition to the government’s study, two studies by non-profit groups and two academic studies have provided preliminary insights into sexual orientation discrimination in Hong Kong. In 2005, Women’s Coalition surveyed 693 respondents, 98% of whom were self-identified lesbian or bisexual women in Hong Kong, and 2% of whom were men (Chan n.d.). From this sample, 10.8% of respondents reported experiencing harassment at work, 5.0% reported knowing or suspecting that they were dismissed from work because of their sexual orientation, 4.9% reported being discriminated against during job applications, and 15.4% reported a suspicion of being discriminated against during job applications. In 2006, the press reported on an online survey of 150 gay and lesbian respondents conducted by Civil Rights for Sexual Diversities, F’Union, Hong Kong Christian Institute, and Blessed Minority Christian Fellowship (Leung 2006). In that study, 50% of respondents reported experience with workplace discrimination. In published academic research, Lau et al. (2008) studied 411 men who have sex with men (“MSM”) in Hong Kong and found that men who experienced social discrimination based on MSM status (56% of the sample) were more likely than other MSM to report sexual dysfunctions. Taking a different approach, Wong and Tang (2004) surveyed 187 ethnic Chinese men in Hong Kong and found that respondents’ general perceptions of discrimination against the gay community correlated with respondents’ lack of coming-out behaviors. Respondents’ reports of personal experience with discrimination, however, were found not to be a predictor of coming-out behavior.

Taken together, the government study, the NGO research, and the published academic articles offer little more than a glimpse into the nature and scope of employment discrimination against sexual orientation minorities in Hong Kong. The two unpublished studies by NGOs are helpful because, unlike the government study, they specifically surveyed sexual orientation minorities. These two studies’ samples may, however, be skewed. It is possible that certain segments of the community are disproportionately likely to participate in surveys conducted by advocacy groups. While the two academic studies shed some light on sexual orientation discrimination, the scope of their findings is extremely limited. Neither study focused on discrimination as the primary research variable; they both measured discrimination only as a possible explanatory variable for other variables of interest. Moreover, these studies only addressed subpopulations among male sexual orientation minorities.

These existing studies demonstrate that sexual orientation discrimination exists in Hong Kong. However, the nature and extent of the discrimination is still under-researched. The correlates of discrimination are also not well understood. Foreign research may offer further insights into the possible manifestations and consequences of sexual orientation discrimination in Hong Kong.

Foreign Studies

While studies on sexual orientation discrimination in Hong Kong are limited, many studies have been conducted in other parts of the world, especially the United States. A recent literature review of 15 survey-based studies of sexual orientation discrimination in the United States (Badgett et al. 2009) found that between 10-43% of sexual orientation minorities reported experiences of sexual orientation discrimination. The reported rates varied depending on the scope of individuals surveyed and the studies’ definitions of “discrimination.” Rubenstein’s research (2001), based on complaints filed at equal opportunity agencies in the United States, suggests that the rate of sexual orientation discrimination complaints per gay worker is at least comparable to the rate of sex discrimination complaints filed by women and of race discrimination complaints filed by persons of color. Sexual orientation discrimination in the United States and other parts of the world has also been studied through other—perhaps better—methodologies, such as wage analyses and controlled experiments (Arabsheibani et al. 2004; Badgett and Frank 2007; Badgett et al. 2009).

Foreign researchers have also studied variables potentially associated with sexual orientation discrimination. The results of this research have varied, depending on the populations and variables studied. For example, Arabsheibani et al. (2006) researched sexual orientation-based wage gaps as a proxy for workplace discrimination; they found that age and education interact with the wage differentials differently in the United States and in the United Kingdom. For instance, the wage gap between gay men and straight men was greater for older men in the United States, but the opposite was true for men in the United Kingdom. Researchers have also examined whether sex correlates with experiences of discrimination. In survey-based studies from the United States, lesbian and bisexual women have reported experiences with sexual orientation discrimination in the workplace; however, wage analyses comparing lesbians and straight women have produced inconsistent results, often showing that lesbians earn more than similarly situated straight women (for a review of these findings and possible explanations for them, see Badgett 2006; Badgett et al. 2009). In contrast to the findings on women, wage analyses have rather consistently demonstrated that gay men earn less than similarly situated straight men (Badgett et al. 2009).

A final variable worth noting is employees’ degree of “outness.” Data from interviews and surveys suggest that sexual orientation minorities often avoid “coming out” at work to defend themselves against discrimination (e.g., Badgett 2001; Levine and Leonard 1984; Woods 1994). Even individuals who do not explicitly disclose their sexual orientation can, however, be subject to sexual orientation discrimination. Employers might discriminate against an individual perceived to be gay, even if that individual never explicitly disclosed her sexual orientation. In addition, an individual who never discloses her sexual orientation can be outed by others.

Looking beyond discrimination’s effects on the labor market, a significant body of literature outside of Hong Kong suggests that reported experiences of discrimination are positively associated with psychological distress (e.g., Kessler et al. 1999). Researchers studying populations outside of Hong Kong—for example, the United States and the United Kingdom—have shown that experience with sexual orientation discrimination correlates with reduced mental well-being (Cochran 2001; Mays and Cochran 2001; Warner et al. 2004). Coercing sexual orientation minorities to conceal their sexual orientation is a specific form of discrimination associated with psychological distress. Foreign research suggests that disclosing one’s sexual orientation is associated with improved self-esteem and reduced distress (Badgett 2001; Corrigan and Matthews 2003). Note that Wong and Tang reported similar dynamics in Hong Kong; they found that, among gay men in Hong Kong, disclosure was positively related to overall life satisfaction (Wong and Tang 2003) and negatively correlated with mental distress (Wong and Tang 2004).

While these foreign studies inform some of our hypotheses discussed below, the applicability of their findings to Hong Kong is arguably limited because of cultural differences between Hong Kong and other parts of the world. For example, scholars have argued that the nature of homophobia in Chinese societies differs from that in Western societies. Homophobia is perhaps more pervasive in many Chinese contexts, but is likely less severe. Numerous scholars of cultural studies believe that homophobia is less virulent in Chinese societies because it is not as deeply rooted in religious doctrines, as is the case in the West (e.g., Chou 2001; Martin 2000; Rubin 2003; Wong 2007). Instead of deriving homophobia from religion, Hongkongers may be more likely to perceive same-sex relationships to be incongruent with good reputation because same-sex relationships are viewed as a challenge to traditional Chinese notions of family integrity (Chou 2001; Tang et al. 1997). As Liu and Ding have suggested, “Homosexuals and queers in present-day Taiwan [and other Chinese settings] do not have less, but rather differently, difficult lives because homophobic forces do not operate as overtly and violently but rather to protect everyone else’s face” (Liu and Ding 2005, p. 36).

Cultural differences might explain disparities between how the relationship between homosexuality and morality is perceived. Recall that, in the Hong Kong government study, 38.9% of respondents thought homosexuality contradicts community morals. For comparison purposes, consider the United States. The Pew Foundation (2003) found that 55% of Americans believe engaging in homosexual activity is a sin; it also found (Pew Foundation 2006) that 50% of Americans believe homosexuality is immoral.

In a global city such as Hong Kong, homophobia is likely informed by both Western and Eastern cultural origins (Ho 2008; Wong 2007). Thus, although empirical studies from Western societies can serve as the basis of research hypotheses, it is critical to be mindful of the unique blend of cultures in Hong Kong when predicting and interpreting sexual orientation-based employment discrimination. This study helps to clarify how sexual orientation in Hong Kong, specifically, relates to employment discrimination, and how experience of employment discrimination is related to sexual orientation minorities’ overall well-being.

Survey Aims and Method

Aims

To inform the current debate on legislative reform in Hong Kong, we seek to answer three questions: First, how prevalent is sexual orientation discrimination in Hong Kong? Second, who is most vulnerable to sexual orientation discrimination? And, third, what is the relationship between experiences of sexual orientation discrimination and self-reports of psychological outcomes?

Adopting an inductive approach to theory generation, this exploratory study was intended to provide fundamental data to support the future development of sophisticated theories on sexual orientation discrimination in Hong Kong. Although we did not take theory as our primary starting point, we did form simple hypotheses based on the notion that our data would comport with existing preliminary data. That is to say, we formed hypotheses based on the limited existing data on sexual orientation discrimination in Hong Kong and on data from Western countries, with an understanding that cultural differences between Hong Kong and Western societies might limit the predictive value of data from the West.

First, we hypothesized that, as research by Hong Kong advocacy groups suggested, sexual orientation minorities do indeed experience an amount of employment discrimination that warrants concern. While commentators have argued that homophobia in Chinese societies is expressed less violently and less overtly than in the West, we hypothesized that a considerable amount of employment discrimination exists because employment discrimination is generally a non-violent form of discrimination. For example, refusing or terminating employment based on sexual orientation can occur subtly without overt expressions of disgust or hate. Even workplace harassment can occur through subtle but insidious remarks. Moreover, we were mindful that Hong Kong society has likely absorbed more virulent strains of homophobia from allegedly Western origins. We expected, however, that rates of discrimination may be lower than rates previously reported by advocacy groups due to potential sampling biases in their studies.

Second, we hypothesized that respondents’ age, sex, educational attainment, and disclosure might be associated with rates of discrimination. The hypothesis was relatively tentative with regard to age and sex because foreign studies have been inconsistent in showing associations with age (e.g., Arabsheibani et al. 2006) and sex (e.g., Badgett 2006; Badgett et al. 2009). It is worth emphasizing here the exploratory nature of this Hong Kong study. Rather than seeking to resolve the inconsistent findings from abroad with a deep theory on how age and sex might interact with sexual orientation discrimination in Hong Kong, we sought to generate data to inform future theorizing.

We hypothesized more strongly that educational attainment would be negatively associated with discrimination experience. We did so because the Hong Kong government’s research found that individuals with higher levels of attainment are more accepting of homosexuality and Hongkongers may be likely to work most closely with colleagues of similar levels of educational attainment. Based on foreign studies (e.g., Badgett 2001), we also hypothesized that respondents who concealed their sexual orientation would be the least likely to experience discrimination, but among respondents who did not conceal, the degree of disclosure would not necessarily be associated with higher rates of discrimination.

Third and lastly, we hypothesized that experiences of sexual orientation discrimination in employment will be positively associated with reports of reduced psychological well-being and life satisfaction. This result would be similar to findings from other countries (e.g., Cochran 2001; Mays and Cochran 2001; Warner et al. 2004).

To test our hypotheses, a brief and anonymous Internet survey was conducted using a snowball sampling technique. The survey asked demographic questions, questions about attitudes and overall life satisfaction, and questions regarding experiences with discrimination and violence in Hong Kong. Snowball sampling is a survey methodology that has been used by other researchers to access populations of sexual orientation minorities (e.g., Badgett 1997; Bucher and Raess 2007; Croteau 1996; Weiss 2007). It is a widely accepted technique for accessing hard-to-reach populations, especially those that wish to remain anonymous. Although the technique compromises true randomization because it over-relies on social networks, thus leaving the possibility of excluding “isolates” who are not intricately connected to the networks, researchers have suggested that increasing sample size and varying methods of survey distribution help to assuage such concerns (Atkinson and Flint 2001; Southern et al. 2008). Accordingly, this study sought a large sample through a variety of recruitment channels.

Recruitment

We recruited sexual orientation minorities instead of Hongkongers generally because, as Rubenstein has argued, the urgency of passing antidiscrimination legislation on sexual orientation is best gauged by studying rates of discrimination per capita among members of the self-identified minority community, as opposed to the general public (Rubenstein 2001). We recruited by circulating a solicitation message among people likely to self-identify as sexual orientation minorities. The solicitation message directed individuals to a website hosted by UCLA that featured an information sheet serving also as a consent form. Before proceeding to the rest of the survey, individuals needed to click on a button indicating that they self-identify as a sexual orientation minority in Hong Kong, had read the information sheet, and were over 18 years old. The survey was hosted on Zoomerang, a commercial online surveying tool chosen because of its capacity to post questions and record answers both in English and Chinese.Footnote 4 Survey participants were free to select their language preference for the survey.

Along with asking individuals to take the survey, the solicitation requested that they pass the solicitation to other sexual orientation minorities. The solicitation message was circulated primarily online because of the high level of Internet access in Hong Kong. A report from the City University of Hong Kong found that 83% of Hong Kong’s homes were connected to the Internet in 2008 (HK Internet Project 2009). We also chose to use an Internet-based survey because it ensured anonymity; no IP addresses or other identifying information were collected.

There were four entry points in the snowball methodology, the first of which was email. We sent a wave of solicitation emails in January of 2008 and another wave in early June 2008. The survey was open from January to the end of June. During the waves, we sent emails to individuals who were known to have email addresses for other self-identifying sexual orientation minorities. For example, we sent solicitations to individuals who managed listservs for sexual orientation community groups—such as social networking forums, religious organizations, public interest groups, student organizations, and newsletter communities. Second, we forwarded the solicitation to self-identified sexual orientation minorities on two social networking websites; one website catered to individuals seeking sexual activity while the other catered to individuals seeking to maintain professional and plutonic social networks. Third, the solicitation was posted on Internet bulletin boards and weblogs (“blogs”) that target sexual orientation minority communities. These bulletin boards and weblogs ranged from those frequented by closeted individuals seeking sexual encounters to those that appeal to a more openly gay population. Fourth, we distributed the solicitation message in the form of postcards at gay bars and at Hong Kong’s 2007 International Day Against Homophobia march. The text of the solicitation message was identical for all four entry points. As is the case with all snowball samples, no response rates can be calculated despite the extensive sampling strategy.

Measures

The survey contained multiple parts. First, the survey asked demographic questions concerning variables such as age, sex, citizenship, race/ethnicity, Hong Kong permanent residency status, and relationship status. Next, the survey asked basic questions about life circumstances, such as disclosing sexual orientation to family members, friends, and peers; experiences of discrimination in the workplace; disclosing sexual orientation to colleagues and supervisors in the workplace; discrimination in the receipt of goods and services; and experiences of violence. Last, there were also questions about psychological concerns, such as internalized homophobia and life satisfaction. Although the survey addressed all those areas, this article focuses on experiences of employment discrimination and the factors associated with such discrimination.

The survey included multiple choice questions in the demographic section, scaled measures, and yes/no questions. The scaled measures were used to assess personality factors. Specifically, we used the five-item Revised Internalized Homophobia Scale (Herek et al. 2009) based on the original nine-item scale developed by Martin and Dean (1987). While Herek et al. (2009) created separate scales for lesbians and gay men, we asked individuals to respond to gender-neutral statements such as “I have tried to stop being attracted to members of the same sex” instead of “I have tried to stop being attracted to men.” Because all the items on this scale were extremely negative, five positive items (e.g., “I am proud to be a sexual orientation minority”) were added to balance the valence of the questions. However, all analyses were conducted on the original five items (α = .71). We also included the Satisfaction with Life Scale (Diener et al. 1985), a five-item scale using a 1-7 (agree to disagree) range to determine a person’s overall satisfaction with her life by measuring agreement with statements such as, “I am satisfied with my life” (α = .87). In addition, we employed yes/no questions to ask participants about their level of disclosure at work. They were asked yes/no questions regarding whether they have mentioned their sexual orientation to colleagues and supervisors, and if not, whether they believed their colleagues and supervisors suspected their sexual orientation.

Discrimination was measured with a series of yes/no questions, asking if, during the last five years in Hong Kong, respondents had experienced sexual orientation discrimination. Rather than use a single item simply asking respondents about their overall belief of whether they had experienced discrimination, we used several specific items (e.g., asking respondents whether they had been rejected for a job, denied promotion, or fired). The use of specific items helped to avoid suspect validity that might result from differences in how people interpret a broader single question. The specific items all addressed scenarios that would be considered employment discrimination as it has been defined by major international human rights institutions.Footnote 5

Participants

Participants were all living in Hong Kong. Approximating Hong Kong’s diversity, the respondents came from a broad cross-section of society. A total of 792 people took the survey, though not all participants answered every question. The first page of the survey asked respondents to continue only if they were over the age of 18, but some people under 18 continued anyway. Thus, the age range was wide, with respondents as young as 12 and as old as 72 (M = 26.9, SD = 9.1). There was nearly an even split between males (41.6%) and females (57.1%), with another 0.9% identifying as other types of sex (e.g., transgender). Educational attainment ranged broadly: 13.9% held graduate degrees, 36.4% held bachelor’s degrees, and 8.4% held associate’s degrees. Another 30.0% reported that their highest level of education attained was Form 7 or less (Form 7 being the final year of secondary education in the Hong Kong school system). In addition, 11.2% stated that their highest level of educational attainment was some form of secondary education under a foreign or international system.Footnote 6 Among the participants who identified their race/ethnicity, 88.2% identified themselves as Chinese, 8.9% identified as white, 1.5% reported being of another Asian ethnicity, and 1.3% reported belonging to other categories.

While the sample reflects Hong Kong’s demographics, it is an imperfect reflection. The average age of respondents was slightly lower than that of the general population. Overall, the survey’s participants also held higher levels of educational attainment. It is unclear to what extent such higher levels of education reflect a sampling bias as opposed to actual differences between sexual orientation minorities and the general population in Hong Kong; for comparison purposes, note that population samples from the United States have shown that, on average, gays and lesbians achieve higher levels of education than other individuals (Black et al. 2000). Assuming sexual diversity is consistent across ethnicities, the current survey sample is slightly skewed by an over-representation of white respondents, compared to the number of whites in Hong Kong’s general population.

The survey sample also reflected an interesting dispersion of relationship statuses. Of those who answered the relationship questions, 64.5% of respondents were in a same-sex relationship, among whom 33.7% lived with their same-sex partner. Thus, overall, 22.2% of respondents were living with same-sex partners. In addition to living with same-sex partners, 4.2% of respondents were in legal relationships (i.e., marriages, civil unions, etc.) registered abroad and there were an additional 3.2% of participants who had been in a legally recognized relationship from abroad that ended (i.e., through divorce, etc.). There was further diversity with regard to participants’ experiences with opposite-sex relationships. For example, 4.7% stated that they were in opposite-sex relationships. Only 21.6% of those respondents were living with their opposite-sex partner and 24.3% were married to their opposite-sex partner. Of the total participants, 3.5% had been in an opposite-sex marriage that ended.

Results

Prevalence of Discrimination

When considering the entire sample who completed the survey questions on employment discrimination, 24.6% (n = 161) reported experiencing at least one type of discrimination in the last five years. However, this number is misleading because the overall sample included many people, such as full-time students, self-employed individuals, and retired individuals who responded that the discrimination questions were not applicable to their situation. Including these individuals in our analysis of employment discrimination would fail to consider employment itself as a factor, thus understating the rate of employment discrimination. Therefore, for the remainder of the paper, we ran analyses focusing on individuals who are vulnerable to employment discrimination, namely employees (i.e., those currently employed by others).

Among the sample of employees, 29.3% (n = 161) reported experiencing sexual orientation-based employment discrimination within the last five years. This discrimination took several forms: 7.7% (n = 42) reported having been rejected for a job and another 2.6% (n = 14) reported being fired or asked to leave work. There was also evidence of more subtle types of discrimination: 4.9% (n = 27) reported having been pressured to leave a job, 4.4% (n = 24) reported being denied a promotion, and 6.8% (n = 37) reported being mistreated in the assignment of tasks. Regarding harassment and work environment, 13.3% (n = 88) reported being harassed or persecuted by their colleagues and/or supervisors, and 19.1% (n = 104) reported being treated with less respect because of their sexual orientation.

Vulnerability to Discrimination

To test our second hypothesis, we examined whether the rates of reported discrimination varied among employees based on sex, educational attainment, age, and disclosure of sexual orientation in the workplace. Regarding sex, 28.2% (n = 72) of males and 29.6% (n = 85) of females reported experiencing discrimination, which was not a statistically significant difference (χ2 = .125, p = n.s.). Thus, it seems that men and women are both experiencing similar rates of employment discrimination based on sexual orientation.

When comparing those respondents whose highest level of educational attainment was a bachelor’s or post-graduate degree with respondents with less educational attainment, a statistically significant difference was detected (χ2 = 8.05, p < .01). Among 297 respondents with bachelor’s and/or advanced degrees, 24% (n = 72) responded that they had been victims of employment discrimination. Meanwhile, among the 249 respondents whose highest level of educational attainment was an associate degree or some level of secondary school, 35% (n = 88) had experienced discrimination. These results suggest that a higher level of education was a protective factor for sexual orientation minorities (OR = .585 [CI: .40-.85], p < .01).

To determine the relationship between age and experiences of discrimination, we divided the respondents into four categories to maintain adequate sample sizes for each age group: younger than 20, 20-29, 30-39, over 39. The youngest respondents were the most vulnerable to discrimination, with 47.8% (n = 33) of employees younger than 20 reporting experiences of discrimination in the last five years. The other three age groups reported similar rates of discrimination; 26.4% (n = 72) of those in their 20s, 28.5% (n = 36) of those in their 30s, and 26.1% (n = 17) of those 40 or older reported discrimination in the last five years. This difference between employees under 20 and all other respondents was statistically significant (χ2 = 12.78, p < .005), suggesting that the youngest cohort was the most vulnerable to sexual orientation-based employment discrimination. It is important to keep in mind, however, that age and education are interrelated. Employees in the age group reporting the highest rate of discrimination (employees under 20) were also the least likely to have had a bachelor’s degree (only 3 out of the 68 employees who were under 20 years old had a bachelor’s degree). Notably, when excluding the under-20 cohort from analysis, the effect of having a bachelor’s degree still holds (χ2 = 4.09, p < .05), where respondents who had attained at least a bachelor’s degree reported less discrimination (OR = .65 [CI: .43-.98], p < .05).

Rates of reported discrimination also varied based on respondents’ level of sexual orientation disclosure. The survey asked respondents a variety of questions regarding disclosure at work. For simplicity’s sake, we only asked respondents about their current disclosure, even though we asked respondents about their experiences with discrimination over the past five years. Specifically, we asked whether respondents disclosed their sexual orientation in some way (e.g., proclaiming gay pride or referring to a same-sex partner) and, in the absence of disclosure, whether they believe that others at work suspected their sexual orientation. We asked respondents to answer these questions with regard to their supervisors and colleagues.

Our questions produced nine categories of outness by crossing three levels of disclosure (disclosed, suspected, and concealed) with three categories of information recipients (supervisors, colleagues, and both). Analyzing all nine groups independently would be cumbersome. In addition, some groups were very small. Therefore, to provide an overall sketch of how disclosure may impact work experiences, we chose to analyze the following three groups, which represented the extremities and midpoint of outness:

  1. 1.

    Concealed – Respondents who reported that they have not disclosed their sexual orientation to colleagues/supervisors and do not think colleagues/supervisors have any suspicion of their sexual orientation (n = 100).

  2. 2.

    Suspected – Respondents who have not disclosed their sexual orientation to colleagues/supervisors, but think colleagues and supervisors are suspicious of their sexual orientation (n = 57).

  3. 3.

    Disclosed – Respondents who have disclosed their sexual orientation to their colleagues and supervisors. Recall that disclosure could take various forms, e.g., by self-identifying as gay or by mentioning a same-sex partner (n = 138).

Within these three groups, only 8% of the concealed group reported experiencing employment discrimination based on sexual orientation within the past five years. Meanwhile, 46% of the suspected group reported discrimination and 34% of the disclosed group reported experiencing discrimination.

These findings comport with the view that sexual orientation minorities often avoid “coming out” at work to defend themselves against discrimination, unless they perceive that their workplace is accepting of sexual orientation minorities (for an elaboration of this view, see Badgett 2001). Respondents whose sexual orientation was concealed at the time of the survey reported the least discrimination in the last five years. This low rate of discrimination likely results from the fact that these respondents’ colleagues/supervisors do not have information upon which to act, even if the colleagues/supervisors harbor bias against sexual orientation minorities. Although there is no way from these survey questions to determine causation, the eight respondents who reported discrimination in the last five years and yet were concealed at the time of the survey likely suffered discrimination from previous or potential employers who had knowledge or suspicion of the respondents’ sexual orientation (recall that the survey asked respondents about their current disclosure and experiences of discrimination over the past five years).

Commentators have opined that individuals are more likely to disclose their sexual orientation at workplaces perceived as socially tolerant. Employees’ decisions whether to disclose at work are not usually based solely on their personal preferences. Instead, individuals’ assessments of discrimination risk likely factor into their decisions on whether to disclose (Badgett 2001). Accordingly, disclosed respondents possibly worked in environments that are, on average, more tolerant than the environments where suspected respondents worked. This dynamic may explain why the suspected respondents were more likely than the disclosed respondents to have experienced discrimination. That is to say, if the suspected respondents’ workplaces were not socially tolerant, yet people in their workplaces suspected their sexual orientation, such circumstances would contribute to a uniquely high rate of discrimination. It is important, however, to be mindful that the suspected group was half as big as the other two groups and that the differences in sample size may have skewed the comparison. Perhaps most importantly, it is worth emphasizing that concealed respondents were the least vulnerable to discrimination. This finding supports the idea that sexual orientation minorities have incentives to try their best to conceal their sexual orientation, in order to avoid discrimination.

Altogether, these results foster thinking about sexual orientation discrimination in Hong Kong with greater nuance. On one hand, the results suggest that there were no sex-based differences in discrimination. On the other hand, being younger than 20 years old, having less education, and being “out” at work were associated with increased likelihood of having experienced workplace discrimination.

Associations Between Discrimination and Psychological Outcomes

Our study examined the relationships among sexual orientation-based employment discrimination, overall life satisfaction, and internalized homophobia in Hong Kong. Due to survey time constraints, we did not ask respondents questions regarding specific forms of mental distress such as depression, anxiety, and suicidal ideation. Previous research, however, has found internalized homophobia to be related to such forms of distress (e.g., Herek et al. 2009; Herek and Garnets 2007; Rosser et al. 2008; Szymanksi and Gupta 2009). Moreover, for reasons discussed below, internalized homophobia is itself a dignitary harm that should trouble policymakers.

Examining our entire sample of employees’ overall life satisfaction and internalized homophobia is heartening. The average reported life satisfaction reflected an overall moderate to high level of satisfaction (M = 4.39, SD = 1.24) and the overall reported scores for internalized homophobia were quite low (M = 2.84, SD = 1.17). As predicted in the literature, these two variables were negatively correlated (r = -2.38, p < .0001).

We found, however, that discrimination is significantly associated with both reduced life satisfaction and increased internalized homophobia. Among employees who reported having no experiences of discrimination, the average score of internalized homophobia was 2.77 (SD = 1.18). Meanwhile, employees who reported having at least one experience of discrimination reported a higher level of internalized homophobia (M = 3.00, SD = 1.19) that was statistically significant (t (544) = -2.00, p < .046).

There was also a statistically significant association between discrimination and reduced life satisfaction. Among those who reported no experiences of discrimination, the average score on the life satisfaction scale was 4.52 (SD = 1.19), while those who had experienced discrimination reported a mean score of 4.09 (SD = 1.31). This difference was statistically significant (t (544) = 3.65, p < .001), demonstrating that those who reported recent experiences with discrimination were more likely to have reported decreased life satisfaction.

Although no causal direction can be determined from these findings, the associations we found are consistent with the view that experiencing discrimination has negative psychological effects. (For a discussion of this view among public health scholars, see Mays and Cochran 2001.) Coupled with the foreign studies discussed above, our study provides data for inductively developing inferences that the effects of sexual orientation-based employment discrimination extend beyond financial loss and market concerns, affecting a segment of the population’s dignity and mental well-being.

Limitations and Directions for Future Research

While this study was intended to provide a snapshot of Hong Kong’s sexual orientation minorities, the snapshot has its limitations. The survey utilized a snowball sample based on convenience, as opposed to a random probability sample; as a result, our findings may not be generalizable to reflect Hong Kong’s sexual orientation minorities at large. As noted above, the sample’s proportions of white respondents, highly educated respondents, and young respondents were slightly greater than corresponding proportions in Hong Kong’s general population. Because the survey was conducted over the Internet, the sample may have been biased toward the technology proficient. Assuming that the elderly are less proficient with computers, the technology bias might help to explain the survey’s age skew.

Our study might be characterized by an additional limitation: it is possible that individuals who have experienced discrimination felt more compelled to complete the survey after they realized that the survey’s questions focused largely on discrimination. Individuals who have not experienced discrimination may have avoided the survey or quit the survey early, feeling that the questions did not pertain to them.

It is important to note that while our study examined employees’ perceived experiences of discrimination,Footnote 7 perceptions are sometimes incorrect. It is unclear, however, how this inaccuracy pans out (Badgett et al. 2009). On one hand, employees who believe that they have suffered discrimination might misperceive their employers’ motives, perceiving discrimination when none existed. On the other hand, employers may conceal their motives so well that employees end up reporting less discrimination than actually exists. These two countervailing possibilities might cancel each other out. Future studies of sexual orientation should include studies of actual inequalities. Researchers should consider employing wage analyses and controlled experiments to assess such inequalities (for background on these methods, see Badgett et al. 2009).

In creating future surveys on sexual orientation-based employment discrimination in Hong Kong, researchers should include new questions. We limited the length of this article’s survey, believing that respondents are more likely to complete surveys requiring no more than 20 minutes to finish. As a result, this study provides a snapshot of circumstances in Hong Kong, allowing future studies to zoom-in and view specific parts of the picture in greater detail.

Discussion

Our survey’s findings support the push for legislation in Hong Kong to prohibit employment discrimination based on sexual orientation. The data support such legislation in at least four regards. First, the data show that a troubling number of self-identified sexual orientation minorities report experience with employment discrimination based on sexual orientation. Second, the study suggests that, even if they do not report experience with more overt forms of discrimination, sexual orientation minorities in Hong Kong often suffer indirect discrimination in the form of pressure to remain closeted. Third, the data suggest that, among sexual orientation minorities in Hong Kong, individuals who experience discrimination often belong to subpopulations that are less resilient to the harms of discrimination. Fourth, the data is consistent with the idea that sexual orientation-based workplace discrimination produces dignitary and psychological harms that warrant attention. This section will discuss each of these four factors, which coalesce, weighing in favor of antidiscrimination legislation.

This survey found that nearly one in three respondents reported experiencing sexual orientation discrimination in the workplace. This prevalence of discrimination reported by self-identified sexual orientation minorities reinforces arguments for passing antidiscrimination legislation. Rights advocates might contend that empirical evidence of sexual orientation discrimination should not be a precondition for enacting antidiscrimination legislation. The argument is that, even if only one person suffers from sexual orientation discrimination, one instance of discrimination still warrants a remedy. This article, however, is agnostic on that point. This article suggests that a considerable portion of Hong Kong’s sexual orientation minorities are already suffering from discrimination.

Moreover, viewing our study holistically suggests that the problem of sexual orientation discrimination is more severe than the one-in-three prevalence figure might signify. As discussed above, our study suggests that many sexual orientation minorities have incentives to conceal their sexual orientation in order to avoid discrimination. This pressure to remain closeted is a form of indirect discrimination because, unlike sexual orientation minorities, heterosexual employees generally are not pressured to conceal their sexual orientation.

One might note that Chou Wah-shan has argued that coming out is less important of a personal goal to Hong Kong’s sexual orientation minorities, compared to sexual orientation minorities in the West (Chou 2001). This position has, however, been questioned by numerous other scholars (e.g., Martin 2000; Rubin 2003; Wong 2007). These scholars would likely maintain that pressure to remain closeted is experienced by many—if not most—Hong Kongers as oppressive. Wong and Tang’s (2004) empirical research also suggests that coming out contributes to the psychological well-being of gay Chinese men in Hong Kong. Foreign studies have similarly suggested that coerced closeted-ness compromises mental health (for a review, see Badgett 2001). These studies have also supported the idea that coerced closeted-ness compromises career advancement because the psychological distress caused by closeted-ness leads to increases in workplace stress, absenteeism and turnover, while decreasing productivity, self-confidence, and comfort in professional networking (Badgett 2001). Thus, although our survey found that nearly one-third of respondents reported experiencing sexual orientation discrimination in traditionally defined terms, one should recognize that the rate of discrimination is actually higher once one acknowledges that pressures to remain closeted are a form of indirect discrimination.

The fact that educational attainment is associated with discrimination experience further illuminates the harms of sexual orientation discrimination. Often, the tangible impacts of discrimination are magnified when its victims are of some form of lower social status, such as educational status. For example, individuals of higher educational status may have greater ease in finding new employment after losing a job. Because job opportunities are more limited for individuals with less educational attainment, job replacement is more burdensome for these individuals. In a similar vein, individuals of higher socioeconomic status are likely to have greater savings to serve as a financial cushion. Our survey’s finding that educational attainment and sexual orientation discrimination are related suggests the urgency of passing a sexual orientation antidiscrimination law. Arguably, such legislation would be less important if victims of discrimination are highly resilient to the harms of discrimination. Our survey, however, illuminates that sexual orientation minorities who report workplace discrimination in Hong Kong tend to come from vulnerable subpopulations with relatively less financial resilience.

Finally, our study suggests that sexual orientation discrimination in Hong Kong is associated with more than just financial distress; the relationship detected between reported discrimination and psychological outcomes is consistent with the view that discrimination has the potential to cause dignitary harms and psychological distress. Recall that our findings show an association between discrimination experience and internalized homophobia. Although the association is not itself proof of a causal relationship, the finding bolsters the inferred perspective that experiencing sexual orientation discrimination contributes to internalized homophobia. Internalized homophobia as a dignitary harm should concern policymakers. As the Supreme Court of Canada has succinctly put it: “Human dignity means that an individual or group feels self-respect and self-worth” (Law v. Canada (Minister of Employment and Immigration)). In that vein, the Hong Kong Court of Appeal has stated that sexual orientation minorities have important interests in self-worth (Leung v. Secretary for Justice). Among sexual orientation minorities, internalized homophobia is a form of diminished self-worth that should concern policymakers.

Recall also that the survey found an association between discrimination experience and reduced life satisfaction. Although it is unclear how reduced life satisfaction among respondents manifested in terms of specific symptoms, the general findings on reduced life satisfaction should at least raise a flag, giving policymakers reason to query whether sexual orientation-based employment discrimination generates severe psychological harms such as depression, anxiety, and suicidal ideation.

This study’s findings buttress calls for passing legislation in Hong Kong to combat employment discrimination based on sexual orientation. Passing sexual orientation antidiscrimination legislation in Hong Kong would help to further the sexual orientation antidiscrimination principle that is, according to Hong Kong’s courts, enshrined in Hong Kong’s constitutional law. Legislation is necessary to remedy instances of discrimination, to deter discrimination, and also to empower more sexual orientation minorities not to fear disclosing their identities. To be sure, legislation would probably not directly remedy some of the harms explored in this section. For example, antidiscrimination laws generally do not give sexual orientation minorities grounds to sue employers for social pressure within a workplace on employees to remain closeted. With that said, antidiscrimination laws can ameliorate such pressure by emboldening sexual orientation minorities to disclose their identity since discrimination resulting from disclosure would be remediable.

Conclusion

This article has advanced the understanding of sexual orientation discrimination in Hong Kong by analyzing data from a survey of self-identified sexual orientation minorities. Nearly one-third of the survey’s respondents reported having experienced workplace discrimination over the past five years. That figure, however, does not fully convey the harms of discriminations. Those who reported discrimination in the last five years had, on average, lower levels of educational attainment and were younger than those who did not report discrimination. To the extent that lower educational attainment is associated with fewer job opportunities, discrimination suffered by sexual orientation minorities with lower educational attainment is particularly harmful. Some respondents who did not report direct acts of discrimination likely suffer indirect discrimination because the threat of discrimination creates pressure for sexual orientation minorities to remain closeted. Finally, the results comport with the perspective that experiences of discrimination create psychological consequences in the forms of increased internalized homophobia (an important predictor of mental health and psychological distress) and decreased life satisfaction. A sexual orientation antidiscrimination law is necessary to address these harms in Hong Kong’s private employment sector.