Introduction

The socio-legal recognition of same-sex relationships often co-evolves with discussion or even leads to the recognition of sexual minority individuals’ access to parenthood (Chauveron et al., 2017). Previous research indicates that sexual minority men largely possess lower parenting desire and intention than heterosexual men (Gato et al., 2020; Kranz et al., 2018; Riskind & Tornello, 2017; Robinson & Brewster, 2014; Shenkman, 2021; Tate et al., 2019). Myriad economic (e.g., the high cost of surrogacy), social (e.g., the anticipation of minority stress), and legal (e.g., the lack of protection for the second parent in adoption) barriers may impede sexual minority men’s path to parenthood, which are in part attributable to pervasive heteronormative expectations regarding parenthood in many societies (Berkowitz & Marsiglio, 2007; Murphy, 2013). Compared to lesbians, gay men also face additional layers of obstruction resulting from social norms regarding reproduction as an exclusively feminine domain and therefore seeing men as less warm, nurturing, and capable of properly rearing children (Amodeo et al., 2018; Carneiro et al., 2017; Perrin et al., 2019; Vinjamuri, 2015), together with the practical difficulties of finding a surrogate compared to a sperm donor.

Emerging studies have identified that legal recognition of same-sex relationships may influence sexual minority individuals’ parenting desire, intention, and subsequent parenthood experience (Park et al., 2016, 2020). While over two thirds of respondents in Perrin et al.’s (2019) US study reported having experienced stigma based on being a gay father, gay fathers living in states with few legal protections were confronted by stronger stigma against their parenthood than their counterparts living in states with more legal protections. Drawing on the legal consciousness theoretical framework, a change in the “laws on the books” can translate into how people make sense of and interact with that legal change in their daily life (Park et al., 2020). Therefore, a question remains as to whether the legalization of same-sex marriage (SSM) can readily shape gay men’s desires and perceptions related to having a child even when the methods (e.g., adoption and surrogacy) remain largely unavailable (Kazyak & Woodell, 2016). This study investigated the institutional dependency of gay men’s parenting desire by examining the effect of SSM legalization in Taiwan.

Parenting Desire Among Gay Men

Gay men share similar reasons with their heterosexual counterparts for becoming, or not becoming, a father, such as a quest for life enrichment, expected positive changes from parenthood, continuity of the family line, and consideration of financial, emotional, and physical costs (Gato et al., 2020; Goldberg et al., 2012; Santona et al., 2022; van Houten et al., 2020). Even those who choose not to pursue parenthood may perceive having a child as desirable and valuable (Kranz et al., 2018; Shenkman, 2012).

Notably, the emergence of the desire to have a child, otherwise termed “procreative consciousness,” is sensitive to social climates and legal shifts concerning their partnership and parenthood (Kazyak & Woodell, 2016). Research has sought to locate an “ah-ha” moment, a definitive event or turning point when individuals realize that they have the ability, choice, right, or an internal inclination to become a parent (Berkowitz & Marsiglio, 2007; Murphy, 2013; Pralat, 2021). These moments can arise from personal experiences or interactions, such as a friend having a child, learning about new assisted reproductive technologies (ARTs), or through discussion with an intimate partner (Murphy, 2013; Pralat, 2021).

As the procreative identity framework (Berkowitz & Marsiglio, 2007) suggests, socioeconomic, legal, and cultural barriers can intensify sexual minority individuals’ hesitance or reluctance to seek parenthood (Bauermeister, 2014; Park et al., 2016, 2020). Several studies have investigated how changes in the legal recognition of SSM may influence sexual minority individuals’ desire, intention towards, or actual parenthood. For example, Italy recognized same-sex civil unions in June 2016, yet access to parenthood through donor insemination, surrogacy, or adoption is still not allowed (Giunti & Fioravanti, 2017). Compared with the estimates of Baiocco and Laghi (2013), Scandurra et al. (2019) thus found a higher parenting desire (72.4% vs. 51.8%) and intention (64.2% vs. 26.9%) among Italian gay men after the legalization of same-sex civil unions. Scandurra et al.’s (2019) estimates were also higher than those of Riskind and Patterson (2010), who used the 2002 National Survey of Family Growth (desire: 54%; intention upon desire: 67%).

The Current Study

Although previous studies have suggested that judicial rulings and marriage equality policy can be associated with gay men’s parenting aspirations, their results are either qualitative or cross-sectional (Bauermeister, 2014; Berkowitz & Marsiglio, 2007; Park et al., 2016, 2020; Riskind et al., 2013). To date, no longitudinal research has been conducted to pinpoint the change in gay men’s parenting desire as a function of the legalization of SSM. A unique opportunity to collect prospective data emerged in Taiwan when the the Justices of the Constitutional Court made the ruling, Interpretation No. 748, on May 24, 2017, requiring the Legislative Yuan to either pass a new bill or amend the existing Civil Code to legally recognize same-sex partnerships by May 2019. On May 17, 2019, the Act for Implementation of Judicial Yuan Interpretation No. 748 was passed in Legislative Yuan and took effect on May 24, 2019. Capitalizing on this prospective process of legalizing SSM in Taiwan, this study aimed to address an empirical gap, whether the passing of SSM fostered parenting desire among gay men in Taiwan. Based on the foregoing literature review, we hypothesized that the passing of SSM legislation would increase Taiwanese gay men’s parenting desire. We also inquired about gay men’s perceptions of the importance of parenthood to understand further their attitudes about parenthood over the policy shift. Lastly, we accounted for their intention and attitudes toward getting married since previous studies suggested that parenting desire and attitudes might be more pronounced among those who consider the legal recognition of same-sex partnership important or relevant to them (Kazyak & Woodell, 2016; Tate & Patterson, 2019).

Methods

Sample Recruitment

The study adopted a prospective design to recruit a panel of self-identified gay and bisexual men living in Taiwan via Facebook advertising between May 11 and 23, 2019, although only the gay sample is included in this study. Four parameters were entered in the metrics to circulate the advertisement to Facebook users who indicated in their profiles that they were (1) men, (2) aged ≥ 18 years, (3) residents of Taiwan, and (4) interested in lesbian, gay, bisexual, and transgender community-related issues and information. Interested users were redirected to the online survey platform, SurveyMonkey, to read the detailed information about the study and eligibility criteria, including (1) being assigned male at birth, (2) being aged 18 years or older, (3) self-identifying as gay or bisexual, and (4) residing in Taiwan. Consenting respondents spent approximately 20 minutes completing the survey and were requested to provide contact details (e.g., email address, mobile phone number, or mobile communication application) to receive an e-coupon for NT$100 (~ US$3, all currency conversions as of June 2021) as an honorarium. Between September 14 and October 16, 2020, personalized invitations and information about the follow-up survey were sent to respondents who had provided contact details in the baseline survey. Consenting respondents spent approximately 13 minutes completing the survey and were prompted to leave their contact information again to receive an honorarium. We relied on the respondents’ contact information to link the data from the two waves of data. The study protocol was reviewed and approved by the Human Research Ethics Committee (#EA1904014) at the University of Hong Kong.

In the baseline survey, 1422 respondents completed the questionnaire. Forty-one were removed for being (1) minors (n = 3), (2) heterosexual (n = 24), or (3) repeat respondents according to their IP address (n = 14). Thus, the final sample size for the baseline survey was 1381. The follow-up survey collected 952 responses (attrition rate = 31.1%), of which 89 were discarded because they were from a minor (n = 1), from repeat respondents (n = 8), had missing values in the full survey (n = 25), questions on the variables of interest (n = 30); or did not indicate the respondent’s age (n = 5). Responses were also discarded if they had error messages (n = 1), were unmatched with the corresponding data in the baseline survey (n = 1), self-identified as heterosexual (n = 18) or bisexual (n = 132). The final analytic sample was 731. The demographic differences between those who completed both waves and dropped out from the follow-up survey are shown in Table 1.

Table 1 Sociodemographic differences between completers and dropouts at time 1
Full size table

Measures

Demographics

Respondents’ age, monthly income, education level, and employment status (in employment or not in employment) were collected.

Parenting Desire

Respondents’ parenting desire was sought using a binary question: “would you like to have your own child in your life?” We created a classification to represent any change in respondents’ parenting desire across two time points. Those who reported parenting desire at both waves were labeled “Consistent Yes”; those without a parenting desire at both waves were labeled “Consistent No.” The “Withheld” group refers to those indicating a parenting desire at baseline only, while the “Emergent” group refers to those who developed a parenting desire at follow-up.

Perceived Importance of Parenthood

We used three questions developed by Lo et al. (2016) to assess the perceived importance of having one’s own child for oneself, for one’s present or future partner, and for someone with whom one has a present or future intimate relationship. Respondents answered these questions on a five-point Likert scale ranging from 1 (“very unimportant”) to 5 (“very important”) (Cronbach’s alpha = 0.83). The change in perceived importance was calculated by the numerical difference (i.e., follow-up score — baseline score) to reflect the extent of changes pre- and post-legislation. We also divided the sample into three groups — “no change” (follow-up — baseline score within 0.5SD), “increase” (follow-up — baseline score above 0.5SD), and “decrease” (follow-up — baseline score below 0.5SD). This complementary procedure will show the categorical patterns of change.

Attitudes Towards Same-Sex Marriage

In the follow-up survey, we asked whether the respondents would like to get married in the future (intention of same-sex marriage); the response options comprised “yes,” “possibly,” “no,” and “already married/engaged.” We also asked the extent to which the legalization of SSM was personally important to them, with options ranging from 1 (“not at all important”) to 4 (“very important”).

Statistical analysis

Little’s Missing Completely At Random (MCAR) test (x2 = 1.098, df = 2, p = 0.577) indicated that all missing values were completely random. The change in parenting desire before and after the legalization of SSM was analyzed by McNemar’s test, and the change in the perceived importance of parenthood was analyzed by a paired-samples t-test. The relationship between the change in parenting desire and attitudes towards getting married was investigated by chi-square tests. The associations between the pattern of change in perceived importance and attitudes towards getting married were explored by analyses of variance (ANOVAs) with LSD post hoc analysis. These were followed up by multinomial regression analyses to simultaneously account for demographic and SSM-related variables. Statistical significance was indicated by p < 0.05, and all analyses were conducted with Statistical Package for the Social Sciences (SPSS) 25.

Results

Sample Characteristics

At follow-up, the mean age of respondents was 28.1 years (SD = 5.75; range = 19 to 49). A total of 27.5% had a monthly income below NT$20,000 (~ US$725); 32.6% between NT$20,001–40,000 (~ US$725–US$1450), and 39.9% above NT$40,001 (~ US$1450). A total of 66.4% were in full-time employment or self-employed. About two thirds of respondents (65.3%) were tertiary-educated, and 29.7% possessed a master’s degree.

Changes in Desire for and Perceived Importance of Parenthood

In this study, fewer respondents in the follow-up survey (55.8%) indicated parenting desire than in the baseline survey (71.7%). McNemar’s test was significant (p < 0.001), showing that the proportion of respondents with (vs. without) a parenting desire was significantly different across time. Table 2 presents the results of the changes in parenting desire and its perceived importance. About half (52.8%) of respondents reported having a consistent parenting desire before and after the legislation (“Consistent Yes”), while 25.3% reported a consistent lack of parenting desire (“Consistent No”) in both waves. Notably, 18.9% changed from positive to negative (“Withheld”), whereas only 3.0% changed from negative to positive (“Emergent”). Income was related to the changes in parenting desire (p < 0.05). A large proportion of the “Consistent Yes” group (n = 172, 44.6%) belonged to the highest income category (NTD40,000 or above), while the income distribution in other groups (i.e., “Consistent No,” “Withheld,” and “Emergent”) were relatively even. Other demographic characteristics were not significantly related to the change in parenting desire.

Table 2 Changes in childbearing desire by sample characteristics and attitudes towards same-sex marriage
Full size table

Overall, the respondents reported a mild but significant decrease in the perceived importance of parenthood after the SSM legalization, with the mean decreasing from 3.43 (SD = 0.84) to 3.21 (SD = 0.87), t (730) = 7.67, p < 0.001, Cohen’s d = 0.28. We further divided the sample into three groups according to the changes in the perceived importance of having a child, “no change” (follow-up — baseline score within 0.5SD), “increase” (follow-up — baseline score above 0.5SD), and “decrease” (follow-up — baseline score below 0.5SD). While about half of the respondents (52.9%) were classified as “no change,” 33.7% reported a decrease, and only 13.4% indicated an increase in the perceived importance of having a child. The shift in the perceived importance of parenting was not significantly associated with age, income, or employment, but was significantly associated with education, χ2 (2) = 9.247, p = 0.010. Specifically, the “no change” group included more participants (35.4%) with a postgraduate degree or more than either the “increase” (32.7%) or “decrease” groups (24.0%).

Attitudes Towards Same-Sex Marriage

More than half of the sample (58.1%) reported an intention to enter same-sex marriage in the future, and 35.2% reported they would possibly marry a same-sex partner. Only 4.4% reported a lack of intention to marry, while 2.3% were already married or engaged to a same-sex partner. In addition, 44.3% regarded the legalization of SSM as important to them, while 43.0% reported it as very important to them.

Associations Between Attitudes Towards Same-Sex Marriage and the Change in Desire for and Perceived Importance of Parenthood

Intention to marry was related to the change in parenting desire (Table 2). Specifically, 71% of the “Consistent Yes” group and 54.5% of the “Emergent” group had an intention to marry, while only 41.6% of the “Consistent No” and 44.9% of “the Withheld” reported an intention to marry. In addition, those unsure about their marriage intention accounted for the largest proportion of the “Consistent No” (48.6%) and “Withheld” groups (45.7%). Intention to marry was significantly associated with the change in the importance of having a child, F (3, 727) = 4.194, p = 0.006. After the legalization of SSM, only married/engaged respondents reported increased importance (M = 0.14, SD = 0.98) while the rest reported reduced importance (would like to marry: M =  − 0.17, SD = 0.79; would possibly marry: M =  − 0.30, SD = 0.76; would not like to marry: M =  − 0.52, SD = 0.81).

The extent to which marriage was perceived as important was also related to the change in parenting desire, F (3, 721) = 5.230, p = 0.001 (Table 2). Post hoc analyses with LSD adjustment show that those with a consistent parenting desire (i.e., the “Consistent Yes” group) (M = 3.28, SD = 0.75) attached higher importance to SSM than those with no desire for a child across both surveys (i.e., the “Consistent No”) (M = 3.16, SD = 0.72) (p = 0.004) or withheld their desire (i.e., “Withheld”) (M = 3.17, SD-0.79) (p = 0.020). The extent to which the legalization of SSM was considered important was positively associated with the importance of having a child (r = 0.170, p < 0.001) and its increased importance (r = 0.101, p = 0.007).

Multivariate Analyses

Multinomial regression analyses were conducted to investigate the independent factors associated with the change in parenting desire (with “Consistent No” as reference category) and change in the perceived importance of parenthood (with “No change” as reference category). For the change in parenting desire, first, the “Consistent Yes” group had a greater intention to get married than the “Consistent No” group (95% OR = 2.01–3.93, p < 0.001). Other factors were not significant for the contrast with the “Consistent No” group. The result of the likelihood ratio test was significant, χ2(33) = 105.047, p < 0.001. For the change in the perceived importance of parenthood, those with a reduced perceived importance of parenthood attached lower importance to SSM (95% OR = 0.62–1.00, p = 0.48) and were less likely to have a postgraduate degree (95% OR = 0.40–0.87, p = 0.008) compared to the “No change” group. The result of the likelihood ratio test was significant, χ2(22) = 37.63, p = 0.020. Other factors were not significant.

Discussion

This longitudinal study is the first to document the changes in gay men’s desire and attitudes regarding parenthood when SSM became legal in Taiwan. This study is also the first to provide longitudinal data illustrating potential impact of SSM legalization on gay men’s parenthood desire. Overall, we found a modest decline in both the desire and the perceived importance of having a child among respondents. Our hypothesis, therefore, was not supported. Why might this be the case?

A first practical issue relates to the changing nature of family formation between 2019 and 2020 in the context of the COVID-19 pandemic during which the research was undertaken. In common with other social and economic shocks in the recent past (Schneider, 2015; Sobotka et al., 2011; Testa & Basten, 2014), the disruption in both daily life and the economy brought by the pandemic may have negatively impacted fertility rates (Aassve et al., 2020). Preliminary evidence suggests that fertility preferences among heterosexual couples may have shifted in response to the pandemic, especially where the economy has been hit the hardest (Luppi et al., 2020). This might have a stronger impact on sexual minority individuals than their heterosexual counterparts because of the significant postponement or suspension of ARTs (Alviggi et al., 2020; Gromski et al., 2021). Moreover, in Taiwan, fertility rates fell to a record low in 2020, declining further in early 2021 (Everington, 2021). Previous evidence from Taiwan suggests a robust link between economic resources, non-economic values, and fertility preferences (Gietel-Basten, 2018; Hu & Chiang, 2021). Local ARTs are currently out of reach for Taiwanese gay men (Hsu et al., 2018), rendering their suspension less of a factor impacting decreased desire of parenthood. However, amidst the grim economic outlook, our findings may be partially confounded by the pandemic-caused revision of fertility preferences among heterosexual couples and sexual minorities. This requires further investigation.

Secondly, Kolk and Andersson (2020) investigated the effects of two decades of legalization of marriage and parenthood equality policies on Swedish family dynamics, and their results may shed light on the reasons behind our study’s unexpected finding. Using population-based registry data for 1995–2021, they found that neither the legislation of registered partnerships nor adopting fully gender-neutral marriage legislation markedly increased actual childbearing among same-sex couples. Rather, the subsequent increase in childbearing resulted from legal changes in the joint adoption of a child and access to donor insemination for women in same-sex cohabitating relationships. Notably, the increase in childbearing following these legal changes was observed among women only, but not men in SSM.

Across the two decades of legal changes in Sweden, men in SSM had consistently low rates of having a child because the options for family building (e.g., surrogacy and second-parent adoption) remained costly and inaccessible. Similarly, the legal means to have a child remain limited for Taiwanese gay men. According to Taiwan’s Artificial Reproduction Act, only heterosexual and legally married couples are currently allowed to undertake ARTs to conceive a child (Hsu et al., 2018). While married same-sex couples have yet to be permitted to adopt, gay men are routinely discredited in the determination of qualification as single adoptive parents. Against this legal and structural backdrop, the legalization of SSM alone cannot elevate gay men’s desire for parenthood if the means and avenues remain absent. Further studies are warranted to observe the trend of desire for parenthood among same-sex couples when the legal barriers to reproduction are lifted.

This study is also the first to quantitatively relate changes in gay men’s desire and attitudes toward parenthood to their attitudes regarding SSM. Although gay men’s parenthood desire and the perceived importance of parenthood were found to be fairly stable over the study period, the legal access to SSM appears to have implications for the issue of parenthood only for those who considered or entered marriage. Married or engaged respondents expressed a high level of desire for parenthood across time, whereas those who felt unsure or reluctant to marry reported little or declined desire. Those who considered SSM important to them reported a consistent desire for parenthood, while married or engaged gay men even expressed increased importance of having a child. These findings suggest that the legal status of same-sex partnerships is heavily involved in the desire to become parents and form families (Kazyak & Woodell, 2016). They might also reflect that gay men in Taiwan tend to see marriage and parenthood as intertwined issues.

This study also illustrates the underlying link between access to SSM and parenthood desire. This connection has two layers of meaning. First, this link could be seen as a product of heteronormativity — a set of ideologies and practices, including romantic love, monogamy, and reproductive sexuality — that are built upon heterosexuality and gender hierarchy (Hopkins et al., 2013). Previous studies indicate that gay men’s desire to marry and have a child might involve an intent to assimilate into a heteronormative society (Dempsey, 2013). A recent study with childless lesbian and gay individuals in China has also identified a positive association between internalized homonegativity and parenting desire, meaning that those with negative perceptions of their own sexual orientation reported a greater desire to become parents (Wang & Zheng, 2021). Viewed from a critical queer lens, SSM, on the one hand, is an affirmative response to sexual minority people’s demand to legitimize their intimate relationships. On the other hand, it is also argued that SSM feeds into post-gay/assimilation politics that reify the resemblance of same-sex couples and families to their heterosexual counterparts instead of recognizing their distinction (Ghaziani, 2011). Second, the link between SSM and parenthood represents the robust bundle of legal marital status and rights and benefits in Taiwan. Similar to heterosexual couples, gay men may also develop a desire to rear a child in a legally recognized relationship for the sake of relationship stability and resources and entitlements attached to legal marriage (Shieh, 2006). Knowing that the current social and welfare systems in Taiwan continue to disadvantage non-marital child childbearing, gay men’s intention or decision to marry becomes a critical factor for developing a desire to have a child. These speculations about the findings of this study can be examined by future research.

This study enriches the procreative identity framework (Berkowitz & Marsiglio, 2007), originally proposed to describe how cisgender heterosexual men make reproductive decisions and react to pregnancy loss, miscarriage, and abortion. Fantus and Newman (2022) have revisited this framework by accounting for sexual minority men’s distinctive experiences, suggesting that gay men’s procreative consciousness is widely tied to individual life events, romantic interpersonal relationships, and sociohistorical contexts. Building on this framework, the current study with Taiwanese gay men in same-sex relationships delineates the role of sociocultural factors in fostering men’s desire for parenthood, identifying that gay men’s procreative consciousness partly depends on their attitudes towards marriage. Particularly in Chinese societies, where having a genetically connected child to extend family lineage has been considered a man’s filial responsibility, sexual minority men’s failure to comply with this cultural script could expose them to additional pressure and prejudice (Kong, 2021; Lau et al., 2022; Wang et al., 2009). In this sociocultural context, Taiwanese gay men’s attitudes towards marriage were involved in their desire for parenthood as SSM becomes an avenue not only to fulfill their personal desire to be a father but also to achieve the cultural expectations for adult men to form a family that includes children.

Policy Implications

This research provides important data that SSM legalization has fostered parenting desire among those intending to marry or who have already registered their partnership. This finding has two important implications. First, despite the progression towards marriage equality in Taiwan and many other countries, legal and viable methods of having a child remain beyond the reach of married same-sex couples, whether through surrogacy or adoption. Those with the necessary time and money might be forced to realize their aspirations elsewhere. Given that those considering marriage are more likely to desire a child, policymakers should engage in reciprocal communications with them to understand their needs while conveying the legal concerns from the policymakers’ perspectives. Such processes should be instrumental in responding to the unique needs of those seeking to become parents. Second, the study findings provide information for policymakers to gauge the possible number of gay men who might want to have a child and consider resource allocation and deliberation on policy changes related to reproduction.

Limitations

Several limitations and methodological concerns relating to the study are noteworthy. First, despite the relatively large sample size and the panel design allowing for controlling within-subject changes, the study sample was not representative, so the findings’ external validity (e.g., generalizability) is constrained. Second, the generalizability of the study findings is limited because of the non-probability sampling procedure and the specific characteristics of this convenience community sample of gay men (e.g., relatively younger and well-educated). It should also be noted that only follow-up data were analyzed, and sample attrition could have introduced self-selection bias. That is, the follow-up sample may share some unobserved characteristics distinguishing them from respondents who dropped out. Third, as already observed, the follow-up survey took place during the COVID-19 pandemic, which could strongly undermine respondents’ mental health and childbearing aspirations (Micelli et al., 2020). This significant historic event might be partly responsible for the observed decline in parenting desire and highlights the need for further investigation into the interplay between parenting desire and large-scale socio-economic changes. Fourth, we acknowledge that the definition of parenting desire in this study is more specific to men and may not adequately grasp the full spectrum of ideation regarding family formation (i.e., from ideals through to intentions). Furthermore, while this study was formulated on the hypothesis that legal recognition of same-sex relationships influences the parenting desires of gay men within a short period (i.e., 1½ years between the two waves of data collection), parenthood equality policy may also, in turn, influence the uptake of SSM in the longer term (Kolk & Andersson, 2020). Therefore, continuous follow-up studies are recommended to observe longer-term changes in gay men’s family dynamics, including SSM and parenthood, as well as attitudes and decisions regarding them when adoption and assisted reproduction become available for them.

Conclusion

The very existence of gay fathers in Taiwan is a clear indication that some gay men do not feel a need to forgo their desire for parenthood. This study offers a snapshot of parenting desires among gay men in Taiwan within a short timeframe, identifies the reproductive implications of legalizing SSM, and informs ongoing discussions about sexual minority men’s emerging rights to parenthood.