China faces the dual challenges of ultra-low fertility and moderate aging, threatening its development. While social capital is known to influence fertility intentions, empirical research on this relationship remains limited. This study aims to bridge this gap by examining the impact of both offline and online social capital on fertility intentions and exploring the underlying mechanisms. The decline in global fertility rates, falling below replacement levels in many countries, is a significant concern, particularly in China, where birth rates have been below replacement levels for over two decades. This decline is attributed to various factors, including individual characteristics, family factors, socioeconomic status, and the socio-cultural environment. Previous research has explored these factors, such as housing, religious beliefs, socioeconomic factors, fertility policies, and employment, to understand the reasons behind low fertility intentions. However, the impact of social capital, a crucial factor influencing various aspects of life, including academic achievement, health, and economic development, remains under-explored in the context of fertility. This study defines social capital as capabilities, resources, or norms embedded in social network relationships that affect individuals' actions. It categorizes social capital into offline (social trust and networks) and online (frequency of internet access) components, arguing that a combined approach is necessary to understand its impact on fertility intentions. Previous research, mostly conducted in Eastern European countries, has focused on the impact of social networks, a single dimension of offline social capital, without a thorough analysis of the underlying mechanisms. This study aims to address these limitations by combining offline and online social capital analysis to examine their influence on fertility intentions in China and to explore the mechanisms behind this influence, providing empirical evidence to inform strategies aimed at increasing fertility intentions.

The concept of social capital has evolved over time, from Hanifan's initial description of intangible communal resources to Coleman's emphasis on its role in facilitating actions within social structures and Putnam's focus on its macro-level influence on social efficiency. Lin's work further refined the concept by highlighting the resources embedded in social networks. This study leverages these perspectives to define social capital and its offline and online dimensions. Studies have shown that offline social capital, encompassing social trust and social networks, promotes individuals' willingness to have children. High levels of social trust foster reliance on social support systems for childcare and education, assisting women in balancing work and family life. Social networks influence fertility intentions by disseminating new perspectives and values on childbearing and by providing resources that reduce the costs associated with child-rearing. Conversely, the literature suggests that online social capital, often measured through social networking site (SNS) usage, may reduce fertility intentions. SNS usage may spread contraceptive knowledge, reduce time spent with partners, decrease marital satisfaction, increase divorce rates, and exacerbate fertility anxiety by magnifying the perceived negative impacts of childbearing. The increased access to information and opportunities afforded by the internet also increases women’s independence and employment prospects, potentially leading to a higher opportunity cost of childbearing. The increased focus on consumption may also divert resources away from childbearing, further reducing the willingness to have children. Existing studies, predominantly focusing on Eastern European contexts and examining only a limited dimension of offline social capital, lack a comprehensive analysis of the mechanisms through which social capital influences fertility.

This study utilizes data from the Chinese General Social Survey (CGSS) 2017, a nationally representative survey. The sample comprises 7,518 respondents aged 18–60 after excluding missing values and outliers. The dependent variable is the respondents' self-reported ideal number of children without policy restrictions (0, 1, 2, or 3 or more). Independent variables include measures of offline social capital (social trust and social network), and online social capital (frequency of Internet access). Offline social capital is measured using questions on trust in most people and frequency of social interaction in the past year. Online social capital is measured by questions on the frequency of internet use in the past year. Control variables encompass personal characteristics (gender, age, education, hukou, marital status, health, working hours), social insurance coverage, household characteristics (presence of children, son ratio, family income), and regional location. An ordered probit (oprobit) model is employed for parameter estimation due to the ordinal nature of the dependent variable. Marginal effects are calculated to understand the impact of changes in social capital on the probability of different fertility intentions. To address potential endogeneity issues, instrumental variables (self-rated social class, self-rated socioeconomic status, and first online age) are used in a conditional mixed-process (CMP) approach. Robustness checks are performed using an ologit model and OLS regression with a log-transformed dependent variable. Heterogeneity analyses explore the variation in the effects of social capital across gender, region (East vs. Midwest), and age groups (18–29, 30–39, 40–60). Finally, a mechanism analysis is performed using regressions to explore the mediating role of perceptions of pensions (based on the belief that children bear the primary burden of elderly care), filial piety (based on factor analysis of relevant survey items), and social justice (based on respondents' self-rated evaluation of social justice).

The baseline regression results confirm Hypothesis 1, showing a significant positive relationship between offline social capital and fertility intentions. Social trust and social networks are positively associated with higher fertility intentions. Hypothesis 2 is also supported, with online social capital showing a significant negative relationship with fertility intentions. Marginal effects analysis further reinforces these findings, demonstrating that an increase in social trust or social network increases the probability of having two or more children, while increased internet access reduces it. Endogeneity tests using instrumental variables confirm the robustness of the findings, indicating that offline social capital positively and online social capital negatively impacts fertility intentions. Robustness checks using an ologit model and OLS regression with log-transformed dependent variables yield consistent results. Heterogeneity analysis reveals significant differences across subgroups. Women, Eastern region residents, and individuals aged 30–60 are more strongly influenced by offline social capital, with increased offline capital leading to stronger fertility intentions. In contrast, online social capital negatively affects women, Midwesterners, and the 18–29 age group. Mechanism analysis reveals that both offline and online social capital influence fertility intentions by shaping perceptions of pensions, filial piety, and social justice. Offline social capital strengthens traditional views, while online social capital weakens them. The mean fertility intention in the sample is 1.85, below the replacement level of 2.1.

The findings highlight the complex interplay between offline and online social capital in shaping fertility intentions in China. Offline social capital provides crucial social support, reducing the burden of child-rearing and bolstering women's willingness to have children. This effect is particularly prominent among women, older individuals, and residents in the eastern regions of China. Conversely, online social capital's negative impact stems from its role in disseminating diverse and potentially conflicting views on family planning, weakening traditional values that promote larger family sizes. This effect is especially pronounced among younger individuals and those in the mid-western regions, where access to online information may be increasing. The mediating role of perceptions of pensions, filial piety, and social justice further illuminates the mechanisms through which social capital operates. The traditional Chinese emphasis on family support in old age and the cultural importance of filial piety are challenged by access to alternative information and opportunities. The perceived fairness of society also plays a role, indicating that societal inequities may negatively impact fertility intentions. This study contributes to the literature by providing comprehensive evidence on the differentiated effects of offline and online social capital on fertility intentions in China. The findings emphasize the importance of considering both types of social capital, alongside other relevant factors, when designing policies to address low fertility rates.

This study demonstrates the significant influence of offline and online social capital on fertility intentions in China, highlighting the differential impacts across various demographic groups. The findings suggest that policies aimed at enhancing fertility should consider these differentiated effects and mechanisms. Strategies that promote social trust, encourage social interaction, and manage the potentially negative influence of online information on fertility perceptions are crucial. Future research should explore these issues longitudinally, using more comprehensive measures of social capital, and considering the interplay between social capital and other relevant factors to foster a more nuanced understanding of fertility decision-making in China.

This study's cross-sectional design limits the ability to establish causal relationships between social capital and fertility intentions. The measures of offline and online social capital, though based on existing literature, may not capture the full complexity of these concepts. Additionally, the study does not explicitly account for the influence of past restrictive family planning policies on current fertility intentions and social capital dynamics. Future research should address these limitations by employing longitudinal studies, developing more nuanced measures of social capital, and integrating historical contexts of fertility policies.