The herbivore’s dilemma in Japan: In 2015 around 1 in 4 women and 1 in 3 men in their thirties were unmarried and not in a heterosexual relationship; half of the singles reported no interest in romantic relationships

The herbivore’s dilemma: Trends in and factors associated with heterosexual relationship status and interest in romantic relationships among young adults in Japan—Analysis of national surveys, 1987–2015. Cyrus Ghaznavi, Haruka Sakamoto, Shuhei Nomura, Anna Kubota, Daisuke Yoneoka, Kenji Shibuya, Peter Ueda. PLOS One, November 9, 2020. https://doi.org/10.1371/journal.pone.0241571

Abstract

Background: It has been suggested that an increasing proportion of young adults in Japan have lost interest in romantic relationships, a phenomenon termed “herbivorization”. We assessed trends in heterosexual relationship status and self-reported interest in heterosexual romantic relationships in nationally representative data.

Methods: We used data from seven rounds of the National Fertility Survey (1987–2015) and included adults aged 18–39 years (18–34 years in the 1987 survey; sample size 11,683–17,675). Current heterosexual relationship status (married; unmarried but in a relationship; single) was estimated by sex, age group and survey year, with singles further categorized into those reporting interest vs. no interest in heterosexual romantic relationships. Information about same-sex relationships were not available.

Results: Between 1992 and 2015, the age-standardized proportion of 18-39-year-old Japanese adults who were single had increased steadily, from 27.4 to 40.7% among women and from 40.3 to 50.8% among men. This increase was largely driven by decreases in the proportion of married women aged 25–39 years and men aged 30–39 years, while those in a relationship had increased only slightly for women and remained stable for men. By 2015, the proportion of single women was 30.2% in those aged 30–34 years and 24.4% in those aged 35–39 years. The corresponding numbers for men were 39.3% and 32.4%. Around half of the singles (21.4% of all women and 25.1% of all men aged 18–39 years) reported that they had no interest in heterosexual romantic relationships. Single women and men who reported no interest in romantic relationships had lower income and educational levels and were less likely to have regular employment compared to those who reported such an interest.

Conclusions: In this analysis of heterosexual relationships in nationally representative data from Japan, singlehood among young adults had steadily increased over the last three decades. In 2015 around one in four women and one in three men in their thirties were unmarried and not in a heterosexual relationship. Half of the singles reported no interest in romantic relationships and these women and men had lower income and educational levels and were less likely to have regular employment.

Discussion

Intimate relationships constitute an important component of human fertility and life satisfaction and are associated with better health outcomes [6, 13–16]. As such, a purportedly decreasing interest in sex and romantic relationships has brought young Japanese adults under national scrutiny and fostered unease in light of the country’s rapidly ageing population, low birth rates, and large number of individuals living in social isolation [25, 26]. In this study, which is the first to investigate the phenomenon of so-called “herbivore” young adults in nationally representative data in Japan, we assessed changes in, and sociodemographic factors associated with, heterosexual relationship status and reporting of no interest in heterosexual romantic relationships.

Between 1992 and 2015, the age-standardized proportion of 18-39-year-old Japanese adults who were single (unmarried and not in a heterosexual relationship) had increased steadily, from 27.4 to 40.7% among women and from 40.4 to 50.8% among men, with these numbers corresponding an increase of 2.2 million single women and 1.7 million single men in the investigated age range. The increase was largely driven by increases in singlehood among women aged 25–39 years and men aged 30–39 years. By 2015, around one in four women and one in three men in their thirties were single. Our findings show that the decline in marriage among young adults [9] indeed reflects an increase in singlehood as the proportion who were unmarried and in a relationship rose only slightly among women and remained stable among men. Notably, being in a relationship became increasingly uncommon with higher age and among adults in their thirties, this group was so small that the proportion who were unmarried largely equaled the proportion who were single.

Around one in five women and men aged 18–39 years in 2015 were single while reporting that they had no interest in heterosexual romantic relationships. It should be assumed that a proportion of these individuals have interest, or are currently in, non-heterosexual sexual or romantic relationships. Moreover, the term “kousai” which was used in the question about interest in romantic relationships was not defined in the National Fertility Survey. This term is commonly used to refer to the act of engaging in a relationship with a romantic/sexual partner, but as with the word “dating,” the level of commitment to such a relationship is uncertain; it is possible that a proportion of those who answered that they had no interest in a romantic relationship may have had interest in sexual or romantic pursuits but preferred not to be in a stable relationship at the time of survey participation. However, over half of women and men who reported no interest in heterosexual romantic relationships had no heterosexual experience, indicating that many of these individuals are indeed sexually inactive.

While interpretation of the results on income and employment status among women is made difficult by the relatively large proportion of married women who are housewives [27], we note that single women without interest in romantic relationship had lower educational levels and income than single women with such an interest. For men, there was a pronounced gradient in income and occupational status across categories of relationship status and interest in relationships. For example, the proportion with regular employment was highest among married men (85.8%) and decreased stepwise for those in a relationship (63.2%), single with interest in relationships (53.2%), and single with no interest (42.6%). In age-adjusted analyses, married men were twice as likely to have an annual income of JPY 5 million as compared to those in a relationship and those who were single with an interest in relationship and more than three times as likely to have such an income compared to those without interest in romantic relationships.

The herbivore stereotype has typically been portrayed as a timid individual (most of the time male), seemingly indifferent to sex and romance; in other words, “herbivore” is largely described as a type of personality or a set of behaviors secondary to sexual interests [7, 8, 28]. This notion is challenged by the association of lower education, low income, and a lack of regular employment with reporting no interest in relationships among single women and men. It is likely that aspirations and outcomes in the labor and partner market have common determinants such as personality, values, physical appearance and physical and mental health status, and the associations may not represent causality. Nonetheless, high and stable incomes are important predictors of appeal in the partner market, particularly for men [10, 29–31], indicating that individuals with lower income may have difficulties in finding romantic partners. In addition, one could speculate that precarious employment and financial insecurity may contribute to decreased resources, motivation, and opportunities for pursuing romantic relationships [18]. Indeed as shown in our study and in previous analyses [32], income is strongly associated with marriage among Japanese men. Notably, in a study using longitudinal data (2002–2006) from a nationally representative sample of women [33], those with higher earnings (before marriage) were more likely to marry at any age than those with lower earnings. Moreover, women with a university or junior college degree were more likely to be married at the age of 38 years as compared with those with a high school degree or less, indicating that, perhaps contrary to popular belief [34], high earning potential and education constitute advantages in the partner market also for Japanese women. As such, it could be hypothesized that for some women and men, the reporting of no interest in romantic relationships may represent an adjustment of their expectations, or resignation, given their current lack of access to romantic partners and inability to pursue them, while they would be otherwise interested if they were to meet suitable partners. In support of this hypothesis, we found that around two thirds of the women and men reporting no interest in romantic relationships answered that they wished to get married in their lifetime and that around one third reported that a major reason for not getting married was that they had not met a suitable partner. It should also be noted that the proportion who reported no interest in romantic relationships in 2015 decreased with age and the proportions were largely similar in 2010 vs. 2015, indicating that some individuals get married regardless of their reported interest in romantic relationships.

Although there are other types of social connections which may provide benefits to health and life satisfaction [35–39], and while some individuals may prefer to live without romantic relationships, the large proportion of young adults that has lost interest, given up, or find it hard to form romantic relationships may have important implications for public health and fertility. Our findings indicate a possibility that policy measures aimed at improving employment opportunities and addressing economic disadvantage may also lead to increased interest in romantic relationships and marriage. Moreover, income and employment status are considered as important spouse-selection criteria in Japan, especially for women who evaluate potential husbands [10]. Given that most unmarried individuals wish to get married and the most common reason for staying single is that no suitable partner is available [10], it could be speculated that some individuals are reluctant to form romantic relationships with partners who do not fulfill their partner-selection criteria for marriage (e.g. with respect to income and employment status) but who are otherwise of romantic and sexual interest. This hypothesis is supported by our finding that almost no individuals in their late thirties were unmarried and in a romantic relationship, implying that the promotion of marriage as the most socially acceptable form of relationship between adults and a necessary step for co-habitation and family formation could constitute a barrier to forming romantic relationships in Japan.

While the decline in marriage rates has been established in many high-income countries [40], data on relationship status among unmarried individuals is scarce. A situation similar to that in Japan may be ongoing in South Korea, where a national survey found that 68% of unmarried women and 74% of unmarried men aged 20–44 years were not in heterosexual relationships and that young adults listed lack of money as an important reason for giving up dating [41, 42]. In the US and Europe, higher earnings and education have been associated with a higher likelihood of marriage among both women and men [43–45], although it is not known how these factors influence interest in romantic relationships. We found that the proportion of Japanese women who were single was 65.6% (18–24 years), 41.9% (25–29 years), 30.2% (30–34 years), and 24.4% (35–39 years), and the proportion of men who were single was 75.9% (18–24 years), 55.1% (25–29 years), 39.3% (30–34 years), and 32.4% (35–39 years). Our analyses of nationally representative survey data from Britain (Natsal-3, 2010–2012) and the US (General Social Survey, 2012–2018) showed that singlehood tends to be less common in these countries. In Britain, the proportion of women who were single (not married/living together with partner and not in a stable relationship) was 41.5% (18–24 years), 23.6% (25–29 years), 16.3% (30–34 years), and 14.0% (35–39 years). The corresponding numbers for men were 52.6% (18–24 years), 32.5% (25–29 years), 14.7% (30–34 years), and 11.8% (35–39 years). In the US, the proportion of women who were single was 62.6% (18–24 years), 25.2% (25–29 years), 20.0% (30–34 years), and 16.6% (35–39 years). The corresponding numbers for men tended to be closer to those we observed in Japan: 81.4% (18–24 years), 55.8% (25–29 years), 35.9% (30–34 years), and 22.0% (35–39 years). Of note, however, are the slightly different definitions of single used (S1 File), and that the estimates for Britain and the US accounted for any stable relationship, regardless of the gender of the partner, while only heterosexual relationships were considered in Japan.

Our study has limitations. First, as data on relationship status and interest in romantic relationships was self-reported, findings may have been affected by social desirability bias [46]; the risk of such a bias, however, may have been mitigated by the survey’s use of self-administered questionnaires [47]. Second, although the response rate in the National Fertility Survey was high (70.0–83.8% among unmarried individuals and 85.7–92.5% among married couples) and the sample was weighted so that it was representative of the Japanese population in terms of sex, age, and marital status, non-response might have introduced bias in our results. Third, the large sample size may have resulted in small differences in the investigated variables being statistically significant and interpretation of the findings should consider the absolute differences. Fourth, because the data were cross-sectional, we could not assess temporality of the associations of sociodemographic variables with relationship status and interest in romantic relationships. Finally, as the questions asked in the National Fertility Survey were limited to heterosexual relationships, we could not assess other types of relationships. Data on gay, lesbian, bisexual, and transgender groups in Japan is scarce [48], and questions targeted to the experiences of these groups should be included in future surveys.