Introduction
The impact of household human capital on agricultural production decisions has drawn considerable scholarly attention. Existing research predominantly focuses on current agricultural labor endowments, such as age, education, health, farming experience, and their effects on long-term investment behaviors (Foster and Rosenzweig, Reference Foster and Rosenzweig2010; Abdulai and Huffman, Reference Abdulai and Huffman2014; Zhang et al., Reference Zhang, Chen, Huang and Lai2022; Li et al., Reference Li, Lu, Luo, Li and Gao2024). While this perspective offers valuable insights, it overlooks the dual dimensions of investment decisions: current resource availability and forward-looking expectations (Huffman, Reference Huffman1985). For example, stable land tenure expectations drive land investments by enhancing security, collateral, and liquidity effects (Cao, Bai and Zhang, Reference Cao, Bai and Zhang2020; Senda et al., Reference Senda, Robinson, Gachene and Kironchi2022), underscoring the importance of long-term resource accessibility in decision-making (Zhou, Lu and Zou, Reference Zhou, Lu and Zou2023). However, a comparable focus on labor remains limited. Future labor availability, mainly reflected by offspring gender composition, holds potential implications for long-term agricultural strategies but remains underexplored.
Throughout China's extensive history of agrarian production, families served as both kinship networks and economic units, embedding enduring family-centered ideologies (Hansen, Jensen and Skovsgaard, Reference Hansen, Jensen and Skovsgaard2015; Fredriksson and Gupta, Reference Fredriksson and Gupta2023). Male offspring, valued for their physical advantages, emerged as central figures in agricultural labor, enabling households to expand arable land and enhance bargaining power within village communities (Sheridan et al., Reference Sheridan2023). These roles facilitated access to external resources and bolstered the household's economic status (Boserup, Tan and Toulmin, Reference Boserup, Tan and Toulmin2007). Consequently, this gender-based division of labor reinforced cultural norms that positioned women within the domestic sphere while men engaged in external economic roles (Becker, Reference Becker1985; Alesina, Giuliano and Nunn, Reference Alesina, Giuliano and Nunn2013). Furthermore, ideologies emphasized the imperatives of ‘continuing the family line’ (‘Continuing the family line’ refers to the perpetuation of a family's lineage, traditionally through the birth of sons who can carry on the family name and heritage across generations.) and ‘raising sons for old age’ (‘Raising sons for old age’ signifies the traditional belief that raising sons ensures support and care in one's old age. It encompasses raising children to provide security and support during later years.), underscoring the sociocultural importance of male offspring, entrenched male offspring as primary inheritors of family assets, particularly land and property (Alesina and Giuliano, Reference Alesina and Giuliano2010; Breitenbach and Foguesatto, Reference Breitenbach and Foguesatto2023; Edafe et al., Reference Edafe, Osabuohien, Matthew, Osabohien and Khatoon2023; Bavorová et al., Reference Bavorová, Ullah, Garcia and Cavicchioli2024). In contrast, female offspring, perceived as a labor loss due to marital relocation—encapsulated by the saying ‘a married daughter is like spilt water’ (The saying ‘a married daughter is like spilt water’ signifies that once a daughter is married, she is no longer considered part of her natal family. This idiom compares a married daughter to water spilt onto the ground, which spreads and seeps into the soil, becoming irretrievable. This metaphor highlights the traditional belief in the severance of ties between a daughter and her birth family after marriage, underscoring the notion that she no longer has any connection to her natal family.)—were often excluded from substantial inheritance (Fei, Hamilton and Zheng, Reference Fei, Hamilton and Zheng1992). This entrenched framework of gendered resource allocation highlights household expectations for male and female offspring, raising critical questions about its implications for long-term agricultural investment.
The role of offspring gender composition in shaping household agricultural investment decisions
When households have male offspring, they gain heirs to inherit family assets and continue the family lineage (Glauben et al., Reference Glauben, Petrick, Tietje and Weiss2009; Alesina, Giuliano and Nunn, Reference Alesina, Giuliano and Nunn2013). This reduces decision-makers' preference for immediate returns, encouraging them to prioritize long-term strategies, such as organic fertilizer application, that support intergenerational survival (Arends-kuenning et al., Reference Arends-Kuenning, Kamei, Garcias, Romani and Shikida2021; Breitenbach and Foguesatto, Reference Breitenbach and Foguesatto2023). Meanwhile, male heirs are pivotal in building household social capital through active participation in village collective affairs and labor-sharing activities during peak agricultural seasons (Alesina, Giuliano and Nunn, Reference Alesina, Giuliano and Nunn2013). Stronger social capital enables households to mitigate the labor, credit, and information constraints that often hinder long-term agricultural investments (Shiferaw, Okello and Reddy, Reference Shiferaw, Okello and Reddy2009; Magnan et al., Reference Magnan, Spielman, Lybbert and Gulati2015; Gao et al., Reference Gao, Liu, Yu, Yang and Yin2019). By contrast, households with only female offspring face constraints in long-term labor availability, which weaken their intergenerational planning and reinforce their preference for immediate returns (Wheeler et al., Reference Wheeler, Bjornlund, Zuo and Edwards2012). These constraints exacerbate challenges, coupled with insufficient social capital, further limiting their ability to pursue long-term agricultural investments (Shiferaw, Okello and Reddy, Reference Shiferaw, Okello and Reddy2009; Gao et al., Reference Gao, Liu, Yu, Yang and Yin2019; Xie et al., Reference Xie, Yang, Wang and He2024).
The current study
The purpose of this study is to examine whether and how offspring gender composition influences household long-term agricultural investment behavior, focusing on commercial organic fertilizer adoption (Gao, Sun and Huang, Reference Gao, Sun and Huang2017; Zhou, Cheng and Zhang, Reference Zhou, Cheng and Zhang2022). Analyzing cross-sectional data from 4090 rice-growing households across 10 counties in Hubei province, China (2021–2023), ordinary least-squares (OLS) regressions reveal a significant negative relationship between the absence of male offspring and long-term investments. Instrumental variable (IV) techniques are employed to address potential endogeneity. We examine two key pathways through which offspring gender composition affects investment decisions: psychological and decision-making tendencies, reflected in time preferences, and external resource acquisition, mediated by social capital. The analysis also highlights heterogeneity, distinguishing between internal factors, such as the gender of the agricultural decision-maker, and external conditions, including the level of land marketization.
The contributions of this paper are twofold. First, it extends the existing literature by shifting the focus from the short-term availability of human capital to forward-looking expectations, as reflected in the offspring gender composition. This perspective highlights how differences in the long-term availability of offspring labor influence household long-term agricultural investments. Second, the paper identifies both the mechanisms through which offspring gender composition impacts investment behavior and the heterogeneity of these effects, offering valuable guidance for policymakers.
Theoretical framework and research hypothesis
Offspring gender composition and long-term investment behavior in rural households
China's principles of ‘no increase in land with population growth, no decrease with population decline’ (‘Increase in population without an increase in land, decrease in population without a decrease in land’ is a significant institutional arrangement in rural China. Under this system, rural households are regarded as the primary entities holding the contractual management rights to land. As a result, the death or migration of one or more family members does not affect the household's entitlement to possess, use, and benefit from the contracted land. If members listed on a household registration migrate out of the household but remain within the village collective, they may negotiate with the original household head to claim a portion of the land rights; however, they are not eligible to obtain additional land. When all the natal family's hukou members pass away, the village collective recovers the contracted land.) aim to clarify land management rights and stabilize farmers' expectations. This policy strengthens farmland's property attributes, providing greater security and predictability (Zhou, Lu and Zou, Reference Zhou, Lu and Zou2023). Despite these efforts, land redistribution within farming households remains challenging, particularly during ‘family division’ when offspring marry. (‘Family division’ refers to the process in which relatives who previously cohabited divide their shared family assets, establishing separate households and living independently. Upon marriage and forming a new household, a male offspring relocating from the original household registration while remaining within the same village collective can negotiate with the original household head to claim a share of the family's contracted land. However, additional land cannot be obtained. He retains inheritance rights to the family's contracted land if he stays registered with the original household.) This process often exposes gender-based disparities, with male offspring typically inheriting family assets (Breitenbach and Foguesatto, Reference Breitenbach and Foguesatto2023; Edafe et al., Reference Edafe, Osabuohien, Matthew, Osabohien and Khatoon2023; Bavorová et al., Reference Bavorová, Ullah, Garcia and Cavicchioli2024). The preference for male heirs is driven by three primary factors.
First, male offspring are seen as the bearers of the family lineage. They inherit the Y chromosome from their ancestors, and the father's chromosomes determine the sex of the offspring. Passing on the Y chromosome results in a male child, while passing on the X chromosome results in a female child (Jobling and Tyler-Smith Reference Jobling and Tyler-Smith2003). Furthermore, male offspring increase the family population through marriage, integrating their wives into the family and having children (Fei, Hamilton and Zheng, Reference Fei, Hamilton and Zheng1992). Thus, male offspring are considered capable of ‘continuing the family line’ and ensuring the ‘prosperity of the family’. Historically, China has followed the tradition of ‘women are not included in genealogy’, meaning only males are listed in family records as they are believed to continue the family bloodline.
Second, male offspring are traditionally viewed as the guardians of family property. In rural families, this concept extends beyond financial assets to include land. Rooted in the concept of ‘obedience to the husband after marriage’ (‘Obedience to the husband after marriage’ is a principle rooted in the Three Obediences and Four Virtues, constituting a foundational set of moral principles for women in Confucianism. These doctrines dictate that a virtuous woman must adhere to the authority of the male figures in her life: obeying her father before marriage, her husband during marriage, and her son if widowed.), marriage customs typically require female offspring to relocate to their husband's household (Zhang, Reference Zhang2019; Du, Xiao and Zhao Reference Du, Xiao and Zhao2021). This practice reduces the natal family's labor force and increases the risk of the property being transferred to the husband's family, potentially leading to a net outflow of assets (Fei, Hamilton and Zheng, Reference Fei, Hamilton and Zheng1992). Furthermore, marriage typically creates a separation between women and the land, compounded by cultural perceptions of women as less capable of managing agriculture, thereby weakening their ability to safeguard the natal family's contracted land (Edafe et al., Reference Edafe, Osabuohien, Matthew, Osabohien and Khatoon2023; Bavorová et al., Reference Bavorová, Ullah, Garcia and Cavicchioli2024). (A more severe scenario arises when female offspring transfer their household registration (hukou) to their husband's village or region upon marriage, losing the collective membership necessary to inherit their natal family's contracted land. As a result, when all original household members pass away, the land reverts to collective ownership.) In contrast, when male offspring marry, their spouses' integration and children's birth enhance the family's labor force and property retention capacity.
Third, male offspring are viewed as the primary bearers of filial responsibility. This perception arises from both physiological and socioeconomic considerations. Specifically, males traditionally serve as the primary labor force in agricultural societies, a role often attributed to their physical strength advantages (Alesina, Giuliano and Nunn, Reference Alesina, Giuliano and Nunn2013; Sheridan et al., Reference Sheridan2023). This physiological distinction, reinforced by cultural norms surrounding gendered labor division, underpins the broader economic and social roles assigned to males, whereas females are typically tasked with domestic responsibilities and childcare (Becker, Reference Becker1985). Thus, rural families prioritize educational and economic investments in male offspring, viewing their labor capacity and income potential as essential for safeguarding family property and ensuring reliable old-age support (Zhang, Reference Zhang2019). (However, in households without male offspring, traditional intergenerational support is limited, and government subsidies provide essential support for such families to sustain their livelihoods and address old-age security needs. Individuals aged 50 or older with agricultural household registration and one or two surviving daughters receive annual subsidies: 600 yuan for ages 50–54, 1200 yuan for single-daughter households aged 55 and above, and 960 yuan for dual-daughter households in the same age range.)
Although daughters possess legal inheritance rights, rural customs and practical constraints often impede transferring family land contracting rights to female offspring (Breitenbach and Foguesatto, Reference Breitenbach and Foguesatto2023; Bavorová et al., Reference Bavorová, Ullah, Garcia and Cavicchioli2024). Therefore, when farming households lack male offspring, they face challenges in succession, asset retention, and economic capacity, which weaken intergenerational demands (Wheeler et al., Reference Wheeler, Bjornlund, Zuo and Edwards2012). Long-term agricultural investments require substantial financial input and extended return cycles (Kansanga et al., Reference Kansanga, Luginaah, Kerr, Dakishoni and Lupafya2021). Organic fertilizer exemplifies this, as it decomposes gradually, releasing nutrients gradually and necessitating continuous application across multiple planting cycles to enhance soil fertility and crop yields (Adekiya et al., Reference Adekiya, Ejue, Olayanju, Dunsin, Aboyeji, Aremu, Adegbite and Akinpelu2020; Zheng et al., Reference Zheng, Wei, Lv, Zhang, Zhang, Zhang, Zhang, Zhu, Ge and Zhang2024). Families without male offspring often question the viability of long-term investments, fearing these efforts may not be inherited or effectively utilized. This concern is amplified by the limited timeframe for aging laborers, whose physical decline diminishes the returns on such investments. (Although families without male heirs may theoretically have a longer decision-making period, the agricultural labor structure and the late emphasis on sustainable agricultural practices limit long-term investment incentives. First, the rural workforce is increasingly dominated by older laborers as younger workers migrate to higher-paying off-farm jobs. Those returning to farming are often near retirement, leaving little time for long-term agricultural investments. Second, sustainable agriculture has only recently become a policy priority in China. Early agricultural policies emphasized production, with little regard for sustainability. Not until 2016, with the Central Government's No. 1 Document, did resource conservation and ecological restoration emerge as core objectives.) Consequently, these families prioritize basic subsistence and shift resources away from long-term investments toward immediate economic needs (Breitenbach and Foguesatto, Reference Breitenbach and Foguesatto2023). In contrast, families with male offspring, driven by the need to preserve land for generational transfer, exhibit greater confidence in the continuity of their investments. This assurance drives them to prioritize enhancing the land's value as an inheritable asset and securing its long-term productivity to meet intergenerational demands (Foguesatto et al., Reference Foguesatto, de Vargas Mores, Kruger and Costa2020). Based on this, the following hypothesis is proposed:
H1: The absence of male offspring inhibits long-term agricultural investment behavior.
Pathways through which offspring gender composition affects long-term investment behavior in rural households
Offspring gender composition influences long-term investment behavior in rural households through two main pathways: changes in internal family psychology and decision-making tendencies, specifically time preference, and variations in external resource acquisition and capabilities, namely social capital.
Offspring gender composition, time preference, and long-term investment behavior
The absence of male offspring influences household time preferences, shaping long-term agricultural investment behavior. These investments, such as organic fertilizer application, require immediate costs with benefits accruing over time, making them classic intertemporal decisions (Zhou, Cheng and Zhang, Reference Zhou, Cheng and Zhang2022; Zheng et al., Reference Zheng, Wei, Lv, Zhang, Zhang, Zhang, Zhang, Zhu, Ge and Zhang2024). Psychological mechanisms, particularly time preference and discount rates, govern these decisions. Time preference refers to the inclination to prioritize immediate utility over delayed utility (Frederick, Loewenstein and O'Donoghue, Reference Frederick, Loewenstein and O'Donoghue2002): a higher time preference indicates a stronger preference for immediate satisfaction, whereas a lower time preference reflects a greater willingness to defer gratification for future rewards (Thaler, Reference Thaler1981; Harris and Laibson, Reference Harris and Laibson2013; Kennedy, Reference Kennedy2020). According to time preference theory, these evaluations of future returns influence households' decisions on long-term investments (Shiferaw, Okello and Reddy, Reference Shiferaw, Okello and Reddy2009; Thaler, Reference Thaler2016).
As previously discussed, the gender composition of offspring profoundly shapes household expectations regarding long-term labor availability, influencing intergenerational inheritance strategies, particularly regarding land (Breitenbach and Foguesatto, Reference Breitenbach and Foguesatto2023; Bavorová et al., Reference Bavorová, Ullah, Garcia and Cavicchioli2024). Households without male offspring, constrained by the post-marriage relocation of female offspring, often face reduced intergenerational inheritance needs (Zhang, Reference Zhang2019; Du, Xiao and Zhao Reference Du, Xiao and Zhao2021). Without reliable mechanisms to transfer land across generations, these households emphasize immediate financial returns, prioritizing short-term utility over long-term productivity (Wheeler et al., Reference Wheeler, Bjornlund, Zuo and Edwards2012). This inclination reflects a higher time preference, favoring immediate gains over investments with delayed returns (Kennedy, Reference Kennedy2020). In contrast, households with male offspring anticipate a stable future labor supply. Male offspring's marriages typically expand household size by integrating spouses and children, ensuring continuity in family (Fei, Hamilton and Zheng, Reference Fei, Hamilton and Zheng1992). This expectation strengthens their commitment to preserving arable land as a durable family asset, prioritizing its long-term generational value over immediate returns. As a result, these households exhibit a lower time preference, aligning their investment strategies with intergenerational continuity (Thaler, Reference Thaler1981).
Time preferences further shape farmers' long-term investment decisions (Thaler, Reference Thaler2016). Farmers with low time preference are more likely to undertake long-term investments characterized by extended return cycles and substantial potential for future appreciation, thereby maximizing intergenerational benefits (Thaler, Reference Thaler1981; O'Donoghue and Rabin, Reference O'Donoghue and Rabin1999). As such, they are more inclined to invest in high-cost, long-term ventures like organic fertilizers and biopesticides, which contribute to long-term agricultural productivity and sustainability (Mao et al., Reference Mao, Zhou, Ying and Pan2021). In contrast, farmers with high time preferences tend to focus on immediate gains and short-term satisfaction (Thaler, Reference Thaler1981). They prefer agricultural inputs with quick paybacks, such as chemical fertilizers and pesticides, and typically avoid long-term investments with delayed returns, even if such investments could improve production efficiency and support sustainable land use in the long run. Based on this, the following hypothesis is proposed:
H2: The absence of male offspring increases time preference, inhibiting long-term agricultural investment behavior in farming households.
Offspring gender composition, social capital, and long-term investment behavior
Offspring gender composition significantly shapes long-term agricultural investment behavior by influencing household social capital. Rural areas in China are often described as ‘relational societies’, where social capital serves as a vital supplement to formal institutions (Gao et al., Reference Gao, Liu, Yu, Yang and Yin2019). Social capital, defined as resources obtained through social interactions such as networks, trust, and reciprocity (Coleman, Reference Coleman1988; Putnam, Reference Putnam1993), underpins many aspects of rural life. Within this context, offspring gender composition directly impacts the construction and accumulation of household social capital (Boserup, Tan and Toulmin, Reference Boserup, Tan and Toulmin2007).
Families with male offspring typically hold a stronger voice in village collective affairs. These families frequently contribute more labor to communal projects, such as repairing canals and roads (Alesina, Giuliano and Nunn, Reference Alesina, Giuliano and Nunn2013), traditionally assigned to men due to gender-based labor division (Becker, Reference Becker1985). By contributing labor, the family enhances its social influence within the village, accumulating social capital (Putnam, Reference Putnam2000). Moreover, the seasonal nature of agricultural work further underscores the importance of male offspring. During peak agricultural periods, families with male offspring are better equipped to participate in labor-sharing activities (Krishna, Reference Krishna2001). These arrangements reinforce reciprocal ties and broaden social networks by fostering cooperative household relationships. Over time, such interactions enable these families to accumulate higher levels of social capital (Fredriksson and Gupta, Reference Fredriksson and Gupta2023). In contrast, families without male offspring limit their engagement in collective affairs and communal interactions. Consequently, they tend to form weaker social networks, gain less access to shared resources, and build relatively lower levels of social capital.
Long-term investments are often hindered by deficiencies in labor, credit, and information markets (Shiferaw, Okello and Reddy, Reference Shiferaw, Okello and Reddy2009). Social capital, as an informal institution, mitigates these issues by coordinating labor, facilitating access to financial resources, and promoting information flow (Gao et al., Reference Gao, Liu, Yu, Yang and Yin2019). However, its erosion weakens this capacity. First, the decline in social capital worsens labor shortages by disrupting labor-sharing arrangements that enable farmers to share workforce resources during peak agricultural seasons (Krishna, Reference Krishna2001). Second, diminished social capital restricts access to informal financial networks, heightening liquidity constraints for cash-strapped households (Wossen, Berger and Di Falco, Reference Wossen, Berger and Di Falco2015). This limitation restricts farmers' ability to adopt long-term agricultural investments, which require high upfront costs and offer returns only over extended periods (Gao, Sun and Huang, Reference Gao, Sun and Huang2017). Third, reduced social capital disrupts information dissemination, widening information asymmetries and increasing the transaction costs of adopting new technologies (Rogers, Singhal and Quinlan, Reference Rogers, Singhal and Quinlan2008; Xie et al., Reference Xie, Yang, Wang and He2024). Farmers without access to solid peer networks struggle to acquire reliable knowledge about the benefits and proper usage of investments (Kassie et al., Reference Kassie, Jaleta, Shiferaw, Mmbando and Mekuria2013). Empirical evidence indicates that farmers consistently regard peers as their most trusted and reliable sources of information (Magnan et al., Reference Magnan, Spielman, Lybbert and Gulati2015). When these networks weaken, uncertainties regarding new technologies grow, and trial-and-error costs rise, further discouraging investment (Conley and Udry, Reference Conley and Udry2010). Based on this, the following hypothesis is proposed:
H3: The absence of male offspring reduces household social capital, inhibiting long-term agricultural investment behavior.
In summary, this paper constructs a theoretical framework illustrating how the offspring gender composition affects long-term agricultural investment behavior (Fig. 1).
Figure 1. Theoretical framework for the impact of offspring gender composition on long-term investment behavior.
Materials and methods
Data collection
The data for this study were collected in a cross-sectional household survey in 2021, 2022, and 2023 in Hubei province's primary rice-producing regions. Hubei province is a central grain-producing region in China with high agricultural production levels. Selecting Hubei reflects central China's agricultural production and management patterns. Additionally, its central location offers a unique context for examining family cultural characteristics and gender roles. Rice occupies the largest planting area in Hubei, approximately 28.02% (source of data: Hubei Provincial Bureau of Statistics), followed by wheat and corn. Thus, focusing on rice farmers allows this study to gain deeper insights into agricultural investment behavior and decision-making mechanisms.
All respondents in this study provided informed consent by signing a consent form that outlined the risks and benefits of participation, confidentiality clauses, and the incentives offered. Each respondent received an incentive of 15 yuan, regardless of the completeness of their questionnaire responses. The regions surveyed included Zhijiang city, Qianjiang city, Xiantao city, Honghu city, Wuxue city, Jingshan city, Zhongxiang city, Shayang county, Qichun county, and Xinzhou district. The survey employed a multi-stage random sampling process: two to three townships or streets were selected in each area, followed by two to three administrative villages within those townships or streets, and finally, at least 25 rice-farming households in each village were surveyed.
This paper investigates how offspring gender composition shapes long-term agricultural investment behavior in farming households. The analysis includes only farming households with at least one second-generation offspring, excluding non-farming households, as they do not engage in agricultural investment decisions. Households without children are also excluded for two reasons. First, as the study centers on offspring gender composition, the presence of children is essential. Second, including childless households, which consist of newly formed families and established families without children, could distort the findings by diluting the focus on the latter. The final sample comprises 4090 households (1246 from 2021, 1469 from 2022, and 1375 from 2023) after excluding 383 non-farming households (7.77%) and 459 households without offspring (10.09%) from the initial pool.
Variable selection and measurement
Dependent variable
The dependent variable in this study is long-term agricultural investment behavior, measured by the expenditure on commercial organic fertilizer per mu (667 m2) of land in the previous year (Zhou, Cheng and Zhang, Reference Zhou, Cheng and Zhang2022). This measure is appropriate because organic fertilizers provide lasting benefits beyond the current growing season, making them a recognized proxy for long-term agricultural investments (Jacoby, Li and Rozelle, Reference Jacoby, Li and Rozelle2002; Gao, Sun and Huang, Reference Gao, Sun and Huang2017; Zhou, Cheng and Zhang, Reference Zhou, Cheng and Zhang2022). Unlike chemical fertilizers, which primarily offer immediate nutrients for crop growth, organic fertilizers improve soil structure and water retention capacity, supporting sustainable productivity over time (Zheng et al., Reference Zheng, Wei, Lv, Zhang, Zhang, Zhang, Zhang, Zhu, Ge and Zhang2024).
Organic fertilizers are broadly divided into commercial and homemade types. However, this study focuses exclusively on commercial organic fertilizers due to the significant decline in the use of homemade fertilizers in rural China. This decline is attributed to rising incomes, stricter environmental regulations, and changing rural lifestyles (Lui, Huang, and Zikhali, Reference Liu, Huang and Zikhali2014). Policies such as the Regulations on Pollution Prevention and Control of Livestock and Poultry Farming and the Animal Husbandry Law have further accelerated the separation of livestock rearing from crop cultivation at the household level. Consequently, the proportion of rural households simultaneously engaging in both activities fell from 71% in 1986 to 12% in 2017, with this trend continuing (Jin et al., Reference Jin, Zhang, Wu, Han, Hu, Ren, Zhang, Wei, Wu, Mol, Reis, Gu and Chen2021). Additionally, the Rural Living Environment Improvement Action Plan (2021–2025), which mandates replacing traditional dry toilets with flushing systems, has further reduced the availability of human and animal waste for fertilizer production.
Core independent variable
Following Li and Qian (Reference Li and Qian2021), we use the presence or absence of male offspring to represent gender composition. Specifically, if a household has no male offspring in the second generation, the variable is assigned a value of 1; if the household has male offspring in the second generation, the variable is assigned a value of 0.
In our sample, the average expenditure on commercial organic fertilizer per mu significantly differs between households with and without male offspring, which is statistically significant at the 1% level. Households with male offspring have an average commercial organic fertilizer expenditure per mu of 12.547, whereas households without male offspring have an average expenditure of 6.806 (Table 1). This suggests that the presence or absence of male offspring may influence long-term agricultural investment behavior.
Table 1. Long-term investment behavior by offspring gender composition
Note: The difference represents the mean difference in commercial organic fertilizer expenditure per mu between households with and without male offspring.
Significance levels are indicated as follows: ***, **, and * are significant at 1, 5 and 10%, respectively.
Mediator variable
This study considers two mediator variables: time preference and social capital. Referring to the research by Frederick, Loewenstein and O'Donoghue (Reference Frederick, Loewenstein and O'Donoghue2002), two methods exist to estimate an individual's time preference. The first method indirectly infers time preference by analyzing individuals' decision-making and planning activities in daily life, using these behaviors as proxy variables. The second method asks respondents about their intertemporal choices to estimate time preference. Due to the indirect inference method's complexity and susceptibility to external factors, this study adopts the second, more direct approach by asking farmers, ‘Do you prefer to choose crop varieties with shorter growth cycles?’ Farmers with solid time preferences are likelier to select fast-maturing crop varieties because they provide quicker economic returns, satisfying their preference for immediate benefits.
This study measures social capital using three dimensions: social networks, trust, and norms, following the methodology of Guo, Chen and Zhang (Reference Guo, Chen and Zhang2022). Precisely, social networks are measured by the number of relatives and friends visited during the previous Chinese New Year. Social trust is assessed by the respondent's satisfaction with neighborhood relations (ranging from 1 = very dissatisfied to 5 = very satisfied). Social norms are measured by the number of civil disputes in the village yearly. Considering the impact of the overall social capital level of the village on individual social capital—since the effect of a given amount of individual social capital can vary depending on the village's total social capital—we measure the relative advantage of individual social capital within the village's social capital, based on the study by Li, Lin and Jin (Reference Li, Lin and Jin2019). The specific process is as follows: first, the scores for social networks, trust, and norms K ij are assigned weights W j determined by the entropy method, and each farmer's individual social capital S i is calculated using a weighted sum (Equation (1)). Second, the total social capital of the village S is obtained by summing the individual social capital of all farmers (Equation (2)). Finally, the ratio of a farmer's individual social capital S i to the village's total social capital S is used as the farmer's available social capital SCA i within the village (Equation (3)).
In the above formulas, K ij denotes the score of the i-th farmer on the j-th dimension. W j is the weight assigned to the j-th dimension. S i represents the individual social capital of the i-th farmer (where i = 1, 2, 3, …, n). S signifies the total social capital at the village level. SCA i indicates the available social capital for the i-th farmer within the village, and n is the total number of farmers in the sample.
Control variable
To minimize potential omitted variable bias, this study incorporates controls for several essential factors. Individual characteristics of the agricultural decision-maker—such as age, education, political affiliation, and health—are included to capture personal characteristics influencing decision-making (Foster and Rosenzweig, Reference Foster and Rosenzweig2010; Guo, Chen and Zhang, Reference Guo, Chen and Zhang2022). Household-level characteristics, including the number of agricultural laborers, total annual income (including wage income, net business income, net property income, and net transfer income), household savings rate, hukou migration, and cooperative membership, control for variations in household resources that may affect long-term investment behavior (Wei and Zhang, Reference Wei and Zhang2011; Guo, Chen and Zhang, Reference Guo, Chen and Zhang2022). Land-related variables, such as the total area of cultivated paddy fields, number of field plots, soil fertility, distance from residence to fields, and land tenure certification status, account for differences in land endowments (Zhou, Cheng and Zhang, Reference Zhou, Cheng and Zhang2022). Beyond these intrinsic factors, village-level controls capture external influences on long-term investment behavior, such as farmland consolidation projects and the village's economic conditions (Corsi, Reference Corsi2009; Pun, Joshi and Pun, Reference Pun, Joshi and Pun2024). The model incorporates town-level fixed effects to account for regional variations, capturing factors such as climate, ecological conditions, availability of agricultural inputs, access to technology, and input prices (Bar-Shira, Finkelshtain and Simhon, Reference Bar-Shira, Finkelshtain and Simhon2006). Additionally, time-fixed effects are incorporated to account for macroeconomic and seasonal fluctuations.
Table 2 reports the definitions, assignments, and descriptive statistics of the variables.
Table 2. Descriptive statistics of variables
Note: Descriptive statistics for log-transformed variables are based on the original values; log transformation is applied in model estimation.
Model setting
Baseline regression model
To investigate the impact of offspring gender composition on the long-term investment behavior of rural households, we specify the following baseline regression model:
In Equation (4), LIB it represents the long-term investment behavior of household i in year t, measured as a continuous variable. The OLS method is employed for estimation. OGC it denotes the absence of male offspring in household i in year t, Control it includes individual characteristics, household factors, land features, and village-level conditions. θ r and γ t represent regional and temporal fixed effects, respectively, to account for unobserved heterogeneity across regions and time periods. a 0 is the intercept, a 1 and a 2 are coefficients to be estimated, and $\varepsilon _{it}$
is the random error term.
Mechanism effect test
Following Jiang (Reference Jiang2022), we further identify the causal relationship between the core explanatory variable and the mediator variables to explore how offspring gender composition affects the long-term investment behavior of rural households. The specific model is set as follows:
In Equation (5), M it represents the mediator variables, including time preference and social capital. OGC it denotes the absence of male offspring, and Control it includes the same control variables as in Equation (4). θ r and γ t capture regional and temporal fixed effects, respectively. β 0 is the intercept, β 1 and β 2 are coefficients to be estimated, and $\varepsilon _{it}$
is the random error term. If β 1 is statistically significant, it suggests that the absence of male offspring significantly influences the mechanism variable, thereby suggesting a causal pathway.
Results and discussion
Baseline results
Table 3 presents the regression results on the impact of offspring gender composition on long-term agricultural investment behavior. The estimates in columns (1) and (2) show that, after controlling for individual characteristics of the agricultural decision-maker, household characteristics, land endowment features, village-level conditions, as well as regional and temporal fixed effects, the absence of male offspring has a significant negative effect on long-term agricultural investment behavior. This effect is significant at the 1% level, confirming H1: the absence of male offspring significantly inhibits long-term agricultural investment behavior in households.
Table 3. Baseline regression results for the impact of offspring gender composition on long-term investment behavior
Note: ***, **, and * are significant at 1, 5 and 10%, respectively.
The control variables also provide important insights. The age of the agricultural decision-maker exhibits a significant negative effect, suggesting that older decision-makers are less inclined to pursue long-term investments. Hukou migration similarly has a statistically significant negative impact, reflecting how labor mobility can hinder such investments. Conversely, decision-makers who are members of the Communist Party of China are more likely to make long-term investments. Land-related variables, such as the total field area and the number of field plots, show negative effects, implying that larger operational scales may lead to management challenges that deter long-term investments, which are consistent with the findings of Cao, Bai and Zhang (Reference Cao, Bai and Zhang2020). In addition, greater distances between residences and fields significantly hinder investment, underlining the critical role of geographic proximity in agricultural production. Land tenure certification positively influences long-term investments, supporting Zhou, Cheng and Zhang (Reference Zhou, Cheng and Zhang2022), who emphasized the role of clear ownership rights in promoting long-term agricultural investments. Furthermore, village farmland consolidation significantly enhances investment behavior, aligning with the findings of Pun, Joshi and Pun (Reference Pun, Joshi and Pun2024).
Endogeneity discussion
One major issue in this study is the potential endogeneity of offspring gender composition. The gender composition of offspring might be influenced by unobserved factors that could also affect long-term agricultural investment behavior, leading to omitted variable bias. To address this issue, this study employs the number of children as an IV to identify the absence of male offspring. Theoretically, as offspring size increases, the probability of having at least one male child rises significantly. This relationship is particularly pronounced in son-preference cultures, where parents often continue childbearing until a male child is born (Congdon Fors and Lindskog, Reference Congdon Fors and Lindskog2023). Hence, the instrument satisfies the relevance criterion. Regarding exogeneity, the number of children predominantly reflects factors such as family fertility preferences, sociocultural influences, and policy environment (Huber, Zahourek and Fieder, Reference Huber, Zahourek and Fieder2017; Zhang, Reference Zhang2019), without directly affecting household decisions on long-term agricultural investments. Accordingly, this variable also fulfills the exogeneity requirement, as it does not exert a direct effect on the dependent variable.
Table 4 presents the estimation results of the two-stage least squares method. In the first stage, shown in column (1), the number of offspring exhibits a statistically significant negative coefficient for the absence of male offspring, confirming the instrument's relevance. In the second-stage results, presented in column (2), the absence of male offspring has a statistically significant negative effect on long-term agricultural investment behavior, consistent with the baseline regression findings. Additionally, the Kleibergen–Paap rk Wald F statistic of 767.521 far exceeds the Stock–Yogo critical value of 16.38 at the 10% maximal IV size threshold, further validating the instrument's strength. These findings confirm the robustness of the IV approach in addressing endogeneity concerns and substantiate the causal relationship between offspring gender composition and long-term agricultural investment behavior.
Table 4. Endogeneity discussion—instrumental variable regression results
Note: Standard errors are in brackets; control variables are the same as in Table 3.
***, **, and * are significant at 1, 5 and 10%, respectively.
Robustness test
We conducted several robustness tests to verify the stability of our model estimation results. First, following Brodeur, Cook and Heyes (Reference Brodeur, Cook and Heyes2020), we randomly selected 80% of the sample for regression analysis to evaluate the impact of sample selection on the estimation results (column (1)). Columns (2) and (3) use alternative dependent variables to test the robustness of the results. In column (2), the dependent variable is replaced with household investment in agricultural machinery, measured as the logarithm of the total value of machinery owned (yuan), following Liu, Li and Xu (Reference Liu, Li and Xu2024). In column (3), long-term agricultural investment is proxied by adopting improved water conservancy facilities, including irrigation and drainage infrastructure. This variable is coded as 1 if such improvements are adopted and 0 otherwise, based on Cao, Bai and Zhang (Reference Cao, Bai and Zhang2020). In column (4), we expand the scope of the independent variable by including households without second-generation offspring. This adjustment tests the generalizability of the findings beyond households with second-generation offspring. Column (5) introduces the proportion of female offspring as an alternative proxy for offspring gender composition. This measure captures the family's overall gender structure, addressing the limitations of the binary variable. Lastly, considering that some households may have zero long-term agricultural investments, we employed a Tobit model to examine the consistency and stability of our results under different model specifications (Zhou, Cheng and Zhang, Reference Zhou, Cheng and Zhang2022). The Tobit model results are displayed in column (6), reinforcing the robustness of our findings.
Table 5 summarizes the results of these robustness tests. Across varying samples, alternative dependent variables, and model specifications, the findings consistently demonstrate that the absence of male offspring has a significant negative impact on long-term agricultural investment. These results confirm the robustness of the baseline regression and reinforce the validity of the study's conclusions.
Table 5. Robustness test results for the impact of offspring gender composition on long-term investment behavior
Note: Standard errors are in brackets; control variables are the same as in Table 3.
***, **, and * are significant at 1, 5 and 10%, respectively.
Mediation analysis
We conducted a mediation analysis to explore how offspring gender composition affects long-term agricultural investment behavior, focusing on time preference and social capital. As shown in Table 6, after controlling for individual characteristics of the agricultural decision-maker, household characteristics, land endowment features, village-level conditions and regional and temporal fixed effects, column (1) shows that the absence of male offspring significantly increases household time preference. This finding suggests that families without male offspring are more inclined toward immediate benefits, which inhibits long-term agricultural investment behavior, thereby verifying H2. The results in column (2) show that the absence of male offspring significantly reduces household social capital. Families without male offspring have weaker social networks and limited resource access, inhibiting long-term agricultural investment behavior, thereby verifying H3.
Table 6. Mediation analysis of the effects of offspring gender composition on long-term investment behavior
Note: Standard errors are in brackets; control variables are the same as in Table 3.
***, **, and * are significant at 1, 5 and 10%, respectively.
Heterogeneous results
The impact of offspring gender composition on long-term agricultural investment behavior is not uniform. Internal factors like the continuation of male offspring within the family and external factors such as land marketization levels can influence it. In our analysis, we define the gender of the agricultural decision-maker as a binary variable (‘0’ for female, ‘1’ for male), while the level of land marketization is measured by the proportion of households that lease land through monetary transactions relative to the total number of households engaged in land transfers within the village (China's land system separates ownership rights, contract rights, and management rights. The collective retains ownership rights, while farmers hold contract and management rights, which they can transfer through two primary methods. Non-monetary transfers include informal arrangements such as social rentals, verbal agreements, and short-term leases, leading to higher transaction costs due to their non-market nature. Conversely, monetary transfers involve formal contracts and long-term leases, resulting in lower transaction costs due to their market-oriented approach.) (Liu et al., Reference Liu, Cao, Yan and Wang2016). This analytical framework allows us to examine how offspring gender composition interacts with these internal and external factors to shape long-term agricultural investment behavior across varied family and market environments.
This study uses interaction term regressions to explore the heterogeneous effects of the agricultural decision-maker's gender on how offspring gender composition influences long-term agricultural investment behavior. Column (1) shows that the interaction between the absence of male offspring and the decision-maker's gender is statistically significant at the 5% level, with a negative coefficient. This finding suggests that when the decision-maker is male, the absence of male offspring further discourages long-term agricultural investments. This may reflect the reinforcement of traditional gender roles within the family, particularly through the intergenerational transmission of labor divisions that prioritize males (Daniele, Reference Daniele2024). For instance, resource allocation rights within the family typically default to male members, particularly in long-term economic planning and significant decisions where male opinions often predominate. Consequently, when the decision-maker is male, and there are no male offspring, their conservative attitude toward future agricultural investment may heighten due to insufficient expectations of continuing the family lineage (Duesberg, Bogue and Renwick, Reference Duesberg, Bogue and Renwick2017). In contrast, female agricultural decision-makers are more likely to prioritize fairness in resource allocation and inclusivity in decision-making (Fertő and Bojnec Reference Fertő and Bojnec2024). Their focus on equitable development encourages balanced investment strategies within the household, regardless of offspring gender composition (Fernández, Fogli and Olivetti, Reference Fernández, Fogli and Olivetti2004). This inclusive approach can help mitigate the negative effects of the absence of male offspring on long-term agricultural investment.
Column (2) shows that the interaction between the absence of male offspring and land marketization is statistically significant at the 5% level, with a positive coefficient. This finding suggests that higher levels of land marketization mitigate the negative effect of not having male offspring on long-term agricultural investment behavior. Increased land marketization reduces land transfer transaction costs, enabling farmers to adjust their land holdings more flexibly and decreasing reliance on male labor (Liu et al., Reference Liu, Cao, Yan and Wang2016). A market-driven land transfer mechanism allows farmers to lease land as needed, optimizing the match between land resources and household labor. For example, farmers can lease out excess land beyond their operational capacity, intensively reallocating resources to cultivate within their means. This flexibility improves land and labor use efficiency, enhancing agricultural productivity. Furthermore, higher levels of land marketization make long-term land investments more easily monetizable. Even with limited long-term family labor, an efficient land transfer market allows farmers to earn rent or capital gains, providing direct economic compensation for long-term investments. By reducing the uncertainty of future returns, land marketization increases farmers' incentives for long-term agricultural investments and offsets the negative influence of offspring gender composition on investment behavior (Table 7).
Table 7. Heterogeneity analysis of the effects of offspring gender composition on long-term investment behavior
Note: Control variables are the same as in Table 3.
***, **, and * are significant at the level of 1, 5 and 10%, respectively.
Limitations and future research
This study has several limitations: first, although mixed cross-sectional data provide a diverse horizontal perspective, they do not control for individual fixed effects. This limitation makes it challenging to address time-invariant omitted variables (Gao, Sun and Huang, Reference Gao, Sun and Huang2017). Second, the research is confined to Hubei province, China. While this region's agricultural economic environment and policy background offer some representativeness, its unique local characteristics may limit the external validity of the findings. Third, this study uses expenditures on commercial organic fertilizer to indicate long-term investment behavior. While this reflects farmers' concern for maintaining soil fertility (Zhou, Cheng and Zhang, Reference Zhou, Cheng and Zhang2022), and robustness tests incorporate variables such as household investment in agricultural machinery and adoption of improved water conservancy facilities, these measures do not fully represent the breadth of long-term agricultural investments. Finally, the study focuses primarily on rice farmers. The production methods and investment behaviors of rice farmers may differ significantly from those of farmers growing economic crops like citrus and tea. Therefore, the applicability of the findings to other agricultural sectors may be limited.
Policy implications
The findings suggest several policy implications. In the short term, alternative human capital solutions, such as assistance with hiring machinery or outsourcing services, should be provided to counteract high-time preference and establish positive production expectations for households without male offspring (Qian et al., Reference Qian, Lu, Gao and Lu2022). Strengthening rural community organizations and enhancing mutual assistance mechanisms among villagers are crucial. Community activities, like cultural and sports events, can foster interaction and communication among farmers (Mcmanus et al., Reference McManus, Walmsley, Argent, Baum, Bourke, Martin, Pritchard and Sorensen2012). Local governments can support these efforts by organizing training sessions and technical exchange meetings, helping farmers expand their social networks and build robust cooperative frameworks (Skaalsveen, Ingram and Urquhart, Reference Skaalsveen, Ingram and Urquhart2020). Additionally, transparent management and fair resource distribution systems can enhance trust among villagers. Establishing and promoting village regulations can standardize behavior, encouraging mutual assistance and collective development. Local governments should introduce incentive policies to reward farmers who actively participate in and contribute to community cooperation, thereby setting examples of mutual support. These measures will increase the availability of social capital through broader social interactions, enhancing farmers' willingness and capacity for long-term investment.
In the long term, fostering gender equality in rural areas by encouraging equal opportunities and resource allocation between male and female offspring is crucial for promoting overall family welfare and development (Liu et al., Reference Liu, Cao, Yan and Wang2016). Besides, advancing land market reforms can enhance economic incentives for farmers by simplifying administrative procedures, reducing land transfer transaction costs, and improving the flexibility and efficiency of land use and allocation (Fei, Lin and Chunga, Reference Fei, Lin and Chunga2021). Establishing a scientific land evaluation system to accurately reflect the long-term investment value of land will boost farmers' confidence and capacity for long-term and effective land investments. Addressing these factors can drive sustainable agricultural development and improve the quality of agricultural investments.
Conclusions
This paper examined the impact of offspring gender composition on long-term agricultural investment behavior, using household commercial organic fertilizer expenditure as an indicator. Based on survey data from 4090 grain farming households in 10 cities (counties) of Hubei province, China, collected from 2021 to 2023, the study reveals that the absence of male offspring significantly reduces long-term agricultural investment behavior. This effect is mediated by increased time preference and decreased social capital. Moreover, the inhibitory impact is more substantial when the current agricultural decision-maker is male and land marketization levels are low.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1742170524000358.
Data availability statement
The data underlying the findings of this study can be accessed by contacting the corresponding author. Public availability of the data is restricted due to privacy and ethical considerations.
Acknowledgments
The authors gratefully acknowledge financial support from the National Natural Science Foundation of China (NSFC grant numbers 42071157 and 72473049). We sincerely thank the farmers who generously participated in this study and shared their valuable insights. Our gratitude also goes to the anonymous reviewers for their constructive feedback and to the editors for their dedication. We appreciate Qingmeng Tong and Zhihui Liang for their suggestions on the paper's structure, and Hongli Li and Shengpeng Sun for their encouragement, and guidance throughout the study's development.
Author contributions
Pengyu Chen: idea, conceptualization, writing—original draft, data processing, and editing. Lu Zhang: writing revision, supervision, project administration, and funding acquisition. Jorge Ruiz-Menjivar: writing—review and editing.
Funding statement
This work was supported by the National Natural Science Foundation of China (Grant No. 42071157).
Competing interests
None.
Open access
