Gay Networks, Greater Happiness? Associations Between Gay and Heterosexual Social Capital and Subjective Happiness Among MSM in Japan

Abstract

Background: Men who have sex with men (MSM) in Japan, and globally, face disparities in mental health and well-being, driven by stigma, exclusion, and minority stress. While social capital is known to enhance well-being, no previous research has examined the distinct roles of gay versus heterosexual social capital in relation to subjective happiness. Methods: We conducted a cross-sectional online survey of 1564 MSM in Greater Tokyo, using geosocial networking applications. Social capital was measured using an adapted Resource Generator to assess access to resources from MSM and heterosexual network members. Subjective happiness was assessed using the Japanese version of the Subjective Happiness Scale. Multivariable linear regression identified correlates of happiness. Results: Gay social capital was the strongest predictor of subjective happiness, exceeding the effects of self-rated health, education, employment, and age. High levels of heterosexual social capital were also associated with greater happiness but to a lesser extent. Unemployment, poor health, and attendance at gay cruising venues were negatively associated with happiness. Conclusions: These findings provide the first global evidence of the differential associations between gay and heterosexual social capital on subjective happiness. Community-based interventions that affirm identity and build peer connection may substantially improve well-being among MSM in Japan.

1. Introduction

Men who have sex with men (MSM) experience significant disparities in mental health and well-being compared to their heterosexual counterparts both globally and in Japan (Bostwick et al. 2010; Hill et al. 2020b; King et al. 2008). These disparities include higher rates of depression, anxiety, suicidality, and overall poorer life satisfaction, often stemming from societal stigma, minority stress , and limited access to culturally competent healthcare (Hidaka and Operario 2006; Hill et al. 2020a; Kaneko et al. 2024; Meyer 2003). Minority stress refers to the chronic stress faced by members of stigmatized groups as a result of ongoing discrimination, social exclusion, and internalized stigma (Meyer 2003). Globally, MSM report lower levels of subjective happiness due to these intersecting structural and social challenges (Septarini et al. 2021). In Japan, where HIV is a concentrated epidemic at 96 times the prevalence of HIV and 33 times the prevalence of AIDS compared to the general population in Japan (National Institute of Infectious Diseases 2020; Shiono et al. 2011), and societal expectations around conformity and non-disclosure of minority sexualities remain strong, MSM may face compounded challenges that influence happiness and well-being (Hidaka and Operario 2006; Inoue et al. 2015).

In Japan, although there has been limited research examining the happiness and well-being of MSM outside of people living with HIV (PLWH) (Inoue 2020), research has documented high levels of sexuality-based bullying, harassment, and discrimination, which are known contributors to mental distress (Hidaka and Operario 2006; Osaki 2018). Traditional norms around gender and sexuality, alongside limited legal protections for sexual minorities, exacerbate the psychological burden experienced by MSM (Human Rights Watch 2023). Despite growing visibility of sexual and gender minorities, there is still a lack of comprehensive research exploring the factors that can promote happiness and life satisfaction in this population.

Social capital, defined as the resources available to individuals through their social networks, has gained attention in public health for its positive associations with physical and psychological well-being (Hamano et al. 2010; Kawachi et al. 1999, 2008). It encompasses emotional, informational, and tangible support provided by social connections. Social capital may act as a buffer against the psychological toll of minority stress by promoting a sense of belonging, connectedness, and access to supportive resources (Kawachi et al. 2008). Importantly, social capital can stem from different networks, including those composed primarily of other MSM (hereafter referred to as ‘gay social capital’), or those primarily composed of heterosexual individuals (Hill et al. 2021; Urakawa and Page 2014; Van Der Gaag and Webber 2008).

While studies have shown the positive association between social capital and subjective happiness across various European and Asian countries among the general population, including Japan (Tanaka and Tokimatsu 2020), we know of no known studies that have explored this relationship among MSM, nor any that have distinguished between (same-gender) gay and heterosexual social capital networks based on the sexual identity of alters (the people in one’s social network). This distinction is crucial, as gay social capital may provide culturally and emotionally relevant support, while heterosexual social capital may grant access to more institutional or mainstream societal resources (Hill et al. 2020b, 2024).

This study fills this research gap by analyzing data from a large online survey of MSM in Greater Tokyo. Using the validated Japanese translation of the Subjective Happiness Scale (Shimai et al. 2004), we examine whether MSM and heterosexual social capital are associated with higher subjective happiness. In doing so, this study provides the first empirical evidence on the role of gay and heterosexual social capital networks in shaping happiness among MSM, offering a strengths-based alternative to deficit-focused research and offering actionable insights to into how to improve quality of life.

2. Materials and Methods

2.1. Survey Design and Participants

We conducted a cross-sectional survey investigating social capital among MSM in the Greater Tokyo area. Data were collected via an anonymous, self-administered online questionnaire through gay mobile applications between November 2016 and January 2017. Ethical approval for the study was obtained from the Human Research Ethics Committee at Human Research Ethics Committee at the University of Melbourne (HREC 1646197). Eligible participants were individuals aged 18 years or older who self-identified as MSM (defined as identifying as gay, homosexual, bisexual, or having sexual experience with men) and who provided informed consent. Participation was entirely voluntary. The questionnaire was translated from English into Japanese by a native translator and back-translated into English by an independent translator. MSM helplines and information services were provided to all individuals accessing the survey link. Participants could opt to receive survey results via email and enter a lottery for gift card prizes of up to USD 80.

A total of 1657 participants completed the survey. Duplicate entries were identified and removed in cleaning. Additional removals were made for incomplete responses, resulting in a final analytic sample of 1564 respondents who completed all questions relevant to the study’s primary variables.

2.2. Recruitment

Recruitment strategies followed established protocols for MSM research using location-based gay social networking apps (9Monsters, Grindr and Jack’d) and is described in detail in previous publications (Hill et al. 2018, 2019). Initially, participants were recruited via direct messaging on gay mobile applications. A systematic sampling approach was employed, in which one user from each row of app users was contacted until 50 eligible individuals were identified and invited to participate. This approach yielded 215 participants (Landovitz et al. 2013; Rice et al. 2012). Subsequently, a one-week splash screen advertising campaign was launched in January 2017. This full-screen advertisement, displayed to users upon launching the app in the Greater Tokyo area, included a link to the survey. This method, used effectively in U.S. and Australia-based MSM studies (Jayawardena et al. 2021; Rendina et al. 2014), led to the recruitment of an additional 1442 participants.

2.3. Measures

Participants provided information regarding age, gender identity, sexuality, marital status, place of birth, current residential location (Central Tokyo, Greater Tokyo, or other prefectures), education level, employment status, and health status, adapted from previous social capital surveys in Japan (Hill et al. 2018, 2019). Self-rated health was measured using a five-point Likert scale from 1 (Very unhealthy) to 5 (Very healthy) and subsequently recoded into a binary variable (“Fair/Poor” vs. “Healthy”), consistent with prior research in Japan (Ishii and Ishii 2023).

Gay community engagement was assessed by asking participants whether they had ever attended a gay bar or event, a cruising spot (hattenba), or participated in organized gay group activities, defined as community-based activities, such as political or social meetings, community events (e.g., pride parades, fairs, parties), and volunteer activities, in the past six months. Responses were coded as binary (yes/no).

Social capital was measured using an adapted version of the Resource Generator, a validated tool (Webber and Huxley 2007) designed to capture access to tangible and intangible resources via social networks (Van Der Gaag and Snijders 2005; Van Der Gaag and Webber 2008). Participants were asked whether they could access 18 types of resources (e.g., financial support, emotional support, trusted relationships) from individuals identified as either heterosexual or MSM. The Resource Generator produced two scores: MSM social capital and heterosexual social capital. For each of 18 resources, participants indicated whether they could access the resource through someone who was (1) heterosexual, (2) MSM, (3) both, or (4) neither. MSM social capital scores included responses of ‘MSM’ or ‘both’, while heterosexual social capital scores included responses of ‘heterosexual’ or ‘both’. Scores ranged from 0 to 18, with higher scores indicating greater access to resources through each respective network. This approach allowed overlapping sources of support to be reflected in both scores where appropriate (for more detail see (Hill et al. 2025)). Responses were coded as “MSM,”, “heterosexual,”, “both,”, or “neither,”, with separate total scores calculated for gay and heterosexual social capital (range: 0–18 each). Participants were classified into low, medium, or high gay and heterosexual social capital groups based on their score distributions. High social capital corresponded to one standard deviation above the mean, low to one standard deviation below, and medium to the remaining middle distribution.

Subjective happiness was assessed using the Japanese version of the Subjective Happiness Scale (SHS) (Shimai et al. 2004), a validated four-item instrument designed to capture global self-assessed happiness (Lyubomirsky and Lepper 1999). The SHS has demonstrated strong psychometric properties in Japan (Shimai et al. 2004). Each item is rated on a 7-point Likert scale, resulting in a total score range of 4 to 28, with higher scores indicating greater subjective happiness. The four items were:

• “In general, I consider myself…” (1 = not a very happy person, 7 = a very happy person)
• “Compared with most of my peers, I consider myself…” (1 = less happy, 7 = more happy)
• “Some people are generally very happy. They enjoy life regardless of what is going on, getting the most out of everything. To what extent does this characterization describe you?” (1 = not at all, 7 = a great deal)
• “Some people are generally not very happy. Although they are not depressed, they never seem as happy as they might be. To what extent does this characterization describe you?” (1 = a great deal, 7 = not at all) − reverse scored

The SHS demonstrated strong internal consistency in the present sample (Cronbach’s alpha = 0.86).

2.4. Statistical Analysis

Analyses were conducted using Stata version 16 (StataCorp, College Station, TX, USA). To examine associations between independent variables and subjective happiness (SHS total score), we performed multivariable linear regression. Variables with a p-value < 0.10 in bivariable linear regressions were retained for the final multivariable model in line with previous research using this data set (Hill et al. 2020b, 2021). Robust standard errors were used to account for heteroskedasticity. Model diagnostics indicated no issues of multicollinearity, with VIF values ranging from 1.04 to 1.27, a mean VIF of 1.12, and tolerance values between 0.78 and 0.97.

3. Results

3.1. Socio-Demographics, Health, Happiness, and Gay Community Participation

Table 1. Socio-demographic characteristics (n = 1490).

Age	n	%
Under 25	281	18.9
26 to 35	503	33.8
36 to 40	461	30.9
Over 46	245	16.4
Birth place
Japan	1437	96.4
Other	53	3.6
Residential location
Central Tokyo	801	53.8
Greater Tokyo	502	33.7
Other prefecture	187	12.6
Education level
High school	367	24.6
Two year university	254	17.0
University	723	48.5
Graduate degree	146	9.8
Occupation
Full-time work	1020	68.5
Part-time work	167	11.2
Student	148	9.9
Self-employed/freelance	92	6.2
Unemployed	63	4.2
Marital status
No	1430	96.0
Yes	60	4.0
Sex partner gender
Only men	1359	91.2
Both men and women	131	8.8
Disclosed sexuality to close friends
No	611	41.0
Yes	879	59.0
Healthy
Very Poor/Poor/Fair	542	36.4
Good/Very good	948	63.6
Ever attended Hattenba
No	770	51.7
Yes	720	48.3
Ever attended gay bar or event
No	636	42.7
Yes	854	57.3
Participated in gay group activity in past 6 months
No	1289	86.5
Yes	201	13.5
MSM Social Capital
Low MSM social capital	305	20.5
Medium MSM social capital	886	59.5
High MSM social capital	299	20.1
Heterosexual Social Capital
Low heterosexual social capital	324	21.7
Medium heterosexual social capital	921	61.8
High heterosexual social capital	245	16.4

presents the socio-demographic characteristics, health status, and levels of social capital among participants. The largest proportion of respondents were aged 26 to 35 years (33.7%) and 36 to 40 years (30.5%), followed by those under 25 years (19.4%) and those over 46 (16.4%). The overwhelming majority of participants were born in Japan (96.4). Over half of participants lived in Central Tokyo (53.9%), while 33.5% resided in Greater Tokyo and 12.7% in other prefectures. Educational attainment was varied, though about half of participants had completed university (48.5%), followed by high school (25.1%), two-year college (16.7%), and graduate-level education (9.7%). Most participants were engaged in full-time work (68.1%). Smaller proportions reported part-time work (11.5%), being students (10.0%), self-employed or freelance (6.1%), or unemployed (4.3%). The vast majority were not married (95.9%). In terms of health, 63.4% of participants rated their health as good or very good, while 36.6% rated their health as fair, poor, or very poor. The mean subjective happiness score of participants was 17.6.

Most participants (91.2%) reported having sex exclusively with men, while 8.8% reported having sex with both men and women. A majority had disclosed their sexual identity to close friends (59.0%), although a significant portion (41.0%) had not. Participation in gay community spaces and events varied: 48.1% had ever attended a gay cruising spot (hattenba), 57.0% had ever attended a gay bar or event, and 13.5% had participated in a gay group activity in the past six months.

3.2. Socio-Demographic Correlatetes of Fair or Poor Self-Rated Health—Multivariable Results

Multivariable linear regression in

Table 2. Multivariable linear regression for subjective happiness (n = 1490).

	Mean Happiness		Bivariable Regression		Multivariable Regression
Variable	n	%	β (95% CI)	p-Value	β (95% CI)	p-Value
Age
Under 25	281	17.3	REF
26 to 35	503	17.6	0.17 (−0.52–0.86)	0.638	0.09 (−0.68–0.87)	0.810
36 to 40	461	17.5	−0.00 (−0.71–0.71)	0.998	0.43 (−0.37–1.24)	0.292
Over 46	245	18.2	0.85 (0.03–1.68)	0.042	1.15 (0.24–2.06)	0.013
Birth place
Japan	1437	17.6	REF
Other	53	19.2	1.47 (0.25–2.69)	0.018	0.34 (−0.84–1.51)	0.576
Residential location
Central Tokyo	801	18	REF
Greater Tokyo	502	17.1	−0.85 (−1.39–−0.30)	0.002	−0.21 (−0.72–0.30)	0.410
Other prefecture	187	17.7	−0.30 (−1.05–0.45)	0.431	0.20 (−0.53–0.93)	0.588
Education level
High school	367	16.4	REF
Two year uni	254	17.1	0.51 (−0.25–1.26)	0.192	0.28 (−0.44–1.01)	0.438
University	723	18.1	1.50 (0.90–2.10)	0.000	0.48 (−0.12–1.08)	0.119
Graduate degree	146	19.2	2.64 (1.69–3.59)	0.000	1.47 (0.50–2.44)	0.003
Occupation
Full-time work	1020	18.1	REF
Part-time work	167	15.9	−2.05 (−2.89–−1.21)	0.000	−1.14 (−1.98–−0.31)	0.007
Student	148	17.5	−0.41 (−1.22–0.40)	0.325	−0.60 (−1.56–0.36)	0.221
Self-employed/freelance	92	18.6	0.76 (−0.38–1.89)	0.190	0.53 (−0.49–1.55)	0.309
Unemployed	63	14.2	−4.09 (−5.24–−2.95)	0.000	−2.70 (−3.77–−1.63)	0.000
Marital status
No	1430	17.6	REF
Yes	60	18.8	1.10 (−0.02–2.23)	0.055	0.93 (−0.32–2.18)	0.146
Sex partner gender
Only men	1359	17.5	REF
Both men and women	131	18.6	1.01 (0.16–1.86)	0.020	0.57 (−0.35–1.48)	0.223
Disclosed sexuality to close friends
No	611	17.3	REF
Yes	879	17.8	0.50 (0.00–0.99)	0.049	0.07 (−0.44–0.58)	0.794
Healthy
Very Poor/Poor/Fair	542	15.3	REF
Good/Very good	948	18.9	3.56 (3.08–4.05)	0.000	2.90 (2.40–3.40)	0.000
Ever attended hattenba
No	770	17.8	REF
Yes	720	17.4	−0.42 (−0.92–0.07)	0.090	−0.63 (−1.11–−0.15)	0.010
Ever attended gay bar or event
No	636	17.0	REF
Yes	854	18.1	−1.15 (−1.66–−0.64)	0.000	−0.42 (−0.94–0.10)	0.114
Participated in gay group activity in past 6 months
No	1289	17.5	REF
Yes	201	18.4	0.85 (0.10–1.60)	0.027	−0.01 (−0.72–0.70)	0.969
MSM Social Capital
Low MSM social capital	305	15.5	REF
Medium MSM social capital	886	17.6	1.83 (1.17–2.49)	0.000	1.49 (0.81–2.17)	0.000
High MSM social capital	299	20.0	4.21 (3.44–4.99)	0.000	3.04 (2.18–3.90)	0.000
Heterosexual Social Capital
Low heterosexual social capital	324	15.7	REF
Medium heterosexual social capital	921	17.7	1.82 (1.20–2.44)	0.000	1.00 (0.40–1.61)	0.001
High heterosexual social capital	245	20.0	4.18 (3.41–4.95)	0.000	2.17 (1.36–2.98)	0.000

revealed several significant predictors of subjective happiness among men who have sex with men (MSM) in the Greater Tokyo area.

Age was positively associated with happiness. Compared to participants under 25, those aged over 46 reported significantly higher happiness (β = 1.15, 95% CI: 0.24 to 2.06, p = 0.013). Educational attainment was also significant, and participants with a graduate degree had significantly higher happiness than those with only a high school education (β = 1.47, 95% CI: 0.50 to 2.44, p = 0.003).

Employment status emerged as a strong correlate of happiness. MSM working part-time (β = −1.14, 95% CI: −1.98 to −0.31, p = 0.007) and those who were unemployed (β = −2.70, 95% CI: −3.77 to −1.63, p < 0.001) reported significantly lower happiness scores compared to those in full-time work. Self-rated health was also strongly associated with happiness: participants who rated their health as good or very good were significantly happier than those who reported poor, fair, or very poor health (β = 2.90, 95% CI: 2.40 to 3.40, p < 0.001).

Lastly, some behavioural indicators were also linked to happiness. MSM who had ever attended a cruising spot (hattenba) reported lower happiness than those who had not (β = −0.63, 95% CI: −1.11 to −0.15, p = 0.010), although attendance at gay bars or participation in gay group activities in the past six months were not significantly associated in the adjusted model.

3.3. Social Capital

Social capital was one of the most robust predictors of happiness. MSM with medium levels of gay social capital reported higher happiness scores than those with low gay social capital (β = 1.49, 95% CI: 0.81 to 2.17, p < 0.001), and those with high gay social capital reported the largest increase in happiness (β = 3.04, 95% CI: 2.18 to 3.90, p < 0.001) in Table 2.

Heterosexual social capital was also positively associated with happiness, though to a lesser extent. Compared to those with low heterosexual social capital, MSM with medium levels had higher happiness scores (β = 1.00, 95% CI: 0.40 to 1.61, p = 0.001), while those with high levels showed a stronger association (β = 2.17, 95% CI: 1.36 to 2.98, p < 0.001).

4. Discussion

This study is the first globally to examine the independent associations of gay and heterosexual social capital with subjective happiness among MSM. The results showed that gay social capital was the strongest association with subjective happiness, with a larger point estimate than self-rated health or heterosexual social capital. Compared to MSM with low social capital, those with medium and high levels reported significantly greater happiness, with high gay social capital associated with an adjusted increase of over 3 points on the Subjective Happiness Scale. This further highlights the uniquely powerful role of identity-based social capital networks in enhancing the well-being of sexual minority populations (Hill et al. 2020b, 2024; Axelsson et al. 2013).

The magnitude of this association suggests that gay-specific networks offer types of support that go beyond what is available from broader, heterosexual social capital networks (Kelly et al. 2012). These may include identity affirmation, emotional safety, reduced minority stress, and access to culturally resonant resources (Axelsson et al. 2013; Kelly et al. 2012). This finding aligns with minority stress theory, which posits that supportive peer connections can buffer against stigma and discrimination by fostering community belonging and resilience (Meyer 2003; Hatzenbuehler 2009). The mean subjective happiness score of participants was 17.6, which is broadly comparable to previously reported averages among the general population in Japan. Prior studies have reported mean levels of subjective happiness ranging between 17.5 and 18.5 (Tanaka and Tokimatsu 2020; Shimai et al. 2004). Although it is not possible for this study, or for general population studies of MSM, to achieve representativeness, the results suggest that MSM in Japan fall within the normative range for happiness, although slightly lower, with a notable proportion reporting below-average levels. (A representative sample is not feasible in a context where no reliable denominator exists for hidden populations such as drug users and sexual and gender minorities (Hill et al. 2023; Kaneko et al. 2012), given the lack of LGBTQ inclusion in the Japanese national census.)

Self-rated health, a widely used indicator of well-being, was also significantly associated with happiness, though its effect size was smaller than that of gay social capital. Participants who rated their health as good or very good reported significantly higher happiness than those with poor or fair health. This finding reaffirms the interconnectedness of physical and mental health (Hill et al. 2024; Ishii and Ishii 2023), but also underscores that psychosocial resources such as social capital may, in some contexts, be even more strongly linked to subjective well-being (Kawachi et al. 1999, 2008).

Other socio-demographic and behavioural factors were independently associated with happiness. Older participants—particularly those aged over 46—reported greater happiness than those under 25, consistent with prior findings that suggest identity integration and social stability increase with age (Lorenzi et al. 2015). Conversely, younger MSM may face heightened challenges related to coming out, navigating identity, and accessing supportive networks, which may contribute to lower happiness during early adulthood (Hill et al. 2020b; Hidaka et al. 2008; Russell and Fish 2016).

Educational attainment showed a significant positive association with happiness. Participants with graduate degrees had significantly higher happiness scores than those with only high school education. This may reflect not only material advantages linked to higher education, such as income or job security, but also greater exposure to inclusive social environments that foster openness and acceptance (Hatzenbuehler 2010; Barnes et al. 2014). Employment status was a strong predictor of happiness. MSM who were unemployed or working part-time reported significantly lower levels of happiness compared to those in full-time work. These findings are consistent with previous research showing that employment provides both material and psychological benefits, including routine, social contact, and a sense of purpose (van der Noordt et al. 2014). The particularly low happiness among unemployed respondents highlights the potential compounding effects of economic marginalization and minority stress (Hill et al. 2020a; Bruce and Harper 2011).

Attendance at gay cruising venues (hattenba) was negatively associated with happiness. While these venues can provide opportunities to meet other MSM, the findings may reflect that such spaces, often involving transience and anonymity, do not necessarily foster long-term emotional connection or psychological safety (Takahashi and Magalong 2008). In the Japanese context, where public expressions of same-gender intimacy remain stigmatized, hattenba may be experienced more as hidden, transient spaces (Ichikawa and Kimura 2003) rather than sites of community or joy compared to cruising venues in some Western contexts, where such spaces may be more integrated into broader gay community life. The association observed here may also reflect underlying factors such as social isolation or limited access to affirming MSM spaces, rather than the act of attending itself.

That gay social capital had the strongest association with happiness, even greater than age, education, employment, or health, underscores the importance of investing in community-level interventions that build and sustain peer networks among MSM. Such interventions could include peer mentorship programmes, MSM-specific community centres (severely under-funded in Japan (Ministry of Health, Labor and Welfare, Japan 2025)), or digital platforms that facilitate social connection and shared identity (Liu et al. 2023). These interventions could significantly enhance well-being and help reduce disparities in mental health and happiness experienced by MSM in Japan and globally. Importantly, these findings align with a Bourdieusian understanding of social capital, where not all networks confer the same benefits, and highlights the need for health promotion strategies to recognize the differentiated impact of social capital types, rather than treating it as a uniformly positive resource (Van Der Gaag and Webber 2008).

In this context, future interventions should not only aim to build social capital but also consider the specific composition and function of those networks. Gay-oriented social networks may be best positioned to address minority stress and identity-based stigma, while heterosexual social capital may provide access to instrumental or general support, but less culturally specific affirmation (McPherson et al. 2001). Understanding how MSM, particularly those born outside of Japan or of mixed ethnicity, youth, or those at the intersection of multiple marginalizations, navigate between these forms of capital is essential for tailoring effective supports.

These findings contribute to a more refined theoretical model of social capital that is grounded in intersectionality and minority experience. By mapping the distinct strengths and limitations of gay and heterosexual social capital networks, this study offers a valuable framework for strengthening peer-led community engagement. Rather than viewing gay social capital solely as a buffer against adversity, this study positions it as a central correlate of happiness and as an important counterpoint to research that has historically focused on risk, pathology, or stigma.

Limitations

This study’s findings should be interpreted in light of several limitations. First, the cross-sectional design prevents conclusions about causality. While associations between social capital and happiness were significant, it is possible that happier individuals are also more likely to engage in supportive social networks. Future longitudinal research investigating the directionality of these relationships would be beneficial. Second, data were self-reported, which may introduce recall or social desirability bias. However, this survey was conducted anonymously online, likely limiting social desirability bias (Kelly et al. 2013). Third, recruitment via geo-social networking applications may have excluded MSM who are less digitally connected or not open about their sexuality, thereby limiting generalizability. However, this was a comparatively large sample of MSM and included a substantial proportion of non-gay community attached MSM who would be otherwise underrepresented. Although the point estimates indicated a stronger association between gay social capital and subjective happiness compared to heterosexual social capital, the overlap in confidence intervals requires caution in interpreting the magnitude of this difference. Nonetheless, the consistent direction and statistical significance of both associations reinforce the broader finding that access to diverse forms of social capital is positively linked to well-being among MSM in Japan. Finally, although the Resource Generator used to assess social capital is a validated tool, social capital is inherently context-dependent and multidimensional, which may limit comparability with studies using different social capital frameworks or definitions (Narayan and Cassidy 2001; Shortt 2004).

5. Conclusions

This study provides the first global evidence that gay social capital is strongly associated with subjective happiness among MSM. More than any other factor examined, including self-rated health, age, education, or employment status, gay social capital was the most strongly associated with subjective happiness. These findings underscore the potential of peer-driven social capital in promoting well-being, and support the development of strength-based, community-oriented public health initiatives that foster gay social networks to improve quality of life among MSM.