Introduction

In the background of global agricultural restructuring, promoting rural economic prosperity and sustainable development has become an important issue. Under the influence of technological progress, market liberalization and other factors, the sustainability of traditional small-scale farming has increasingly declined. As a result, more and more rural households, particularly small farmers have begun to engage in diversified entrepreneurial activities, including rural tourism, agro-processing, rural e-commerce and the agricultural service industry (Edilegnaw et al., 2021; Hassink et al., 2016), which play a vital role in stimulating endogenous growth and innovation. Therefore, the international community, especially China, has pursued government policies actively, such as digital village construction and inclusive financial support (Hasan et al., 2021; Razzaq and Yang, 2023; Yang et al., 2023), to improve farmers’ entrepreneurial decision-making (EDM). Farmers’ entrepreneurship has a positive impact on society and individuals. On the one hand, entrepreneurial activities create employment opportunities, absorb local labor, and contribute to the prosperity of the rural economy. On the other hand, entrepreneurship is beneficial to improving lifestyle and empowering individuals to realize the value of lives.

Currently, research on EDM is gradually increasing, and its positive impact on individuals and society is also progressively being confirmed. EDM is influenced by three major factors. The first is the individual characteristics of entrepreneurs, which affect whether they will start a business, such as age, self-efficacy, educational background, and information skills (Maalaoui et al., 2020; Syed et al., 2024; Civelek et al., 2025; Zulfiqar et al., 2021). The second relates to household-level attributes, such as the number of available workforces, intergenerational structure, and family member support (De-Clercq et al., 2023; Hu et al., 2021; Song et al., 2025). The third involves external environmental factors, including institutional policies, infrastructure, and social norms (Qing and Chen, 2024; Vargas-Zeledon and Lee, 2024; Emilio and Mercedes, 2017; Su et al., 2023). Another very important factor is resilience (Williams et al., 2013; Billiet et al., 2021; Kabbara et al., 2025; Uduji and Okolo-Obasi, 2022). Although there has been extensive research on the relationship between resilience and EDM, there is still a lack of research from the micro-level perspective of farming households, particularly regarding the mechanisms between household financial resilience and entrepreneurial behavior. Moreover, existing studies often lack a functionally systematic measurement of financial resilience. Most studies rely on a single indicator, such as household assets or use credit constraints as a proxy (Özsuca, 2024), overlooking the combined impact of key financial indicators such as cash flow liquidity, risk protection capacity, and asset return rates. Finally, in entrepreneurship research, differences in financial conditions between groups are often simplified—such as those differentiated by gender—where identification and subgroup comparative analysis remain insufficient. So, our study aims to fill these gapes.

This study makes the following important contributions: firstly, compared with previous studies, we construct a more precise measure of financial resilience from the functional perspective that better reflects the actual financial situation of farmers, focusing on the following three aspects - household cash flow liquidity, insurance coverage, and investment returns, enriching the research related to financial resilience. Secondly, through empirical analysis, we investigate the relationship between financial resilience and farmers’ entrepreneurial decisions, identified two paths: risk-taking (RT) capacity and social capital (SC), which enriches the research on micro-level behavioral decision-making. Finally, by utilizing micro-level field survey data from rural areas, we capture gender heterogeneity more effectively than national survey data, so as to provide a scientific basis for inclusive policy design, enriching the gender research.

This paper is structured as follows: Section “Literature Review and Theoretical Hypothesis” constructs a literature review and proposes the research theoretical hypothesis. Section “Data, Model Specification, and Variables Description” introduces the data sources, empirical models, and variable selection. Section “Empirical Results” discusses the empirical results, robustness tests, and heterogeneity analysis. Section “Conclusion and Policy Recommendations” presents the conclusions and limitation.

Literature review and theoretical hypothesis

Financial resilience and entrepreneurial decision-making

Under the shock of the COVID-19 pandemic, a growing number of studies have investigated the impact of financial resilience (Kulshreshtha et al., 2023; Nykiforuk et al., 2023). As a core concept reflecting a household’s capacity to cope with economic shocks, scholars have found that financial resilience benefits adolescents’ academic performance (Liu and Chen, 2024; Seely and Mickelson, 2019), members’ health status (Chipunza, 2023; Hamilton et al., 2019) and quality of life (Tahir et al., 2022). In addition, financial resilience also has an impact on production and management behaviors. For example, Bojnec and Fertő (2024) found that financial resilience can expand the scale of farming and improve green production behavior (Wang et al., 2025). These findings suggest that financial resilience may also play a significant role in shaping entrepreneurial investment behaviors.

EDM refers to an individual’s identification of entrepreneurial opportunities and choice of whether to carry out entrepreneurial activities (Iionen et al., 2018), which based on owner resources, information, and subjective cognition. Its characteristics include high uncertainty and resource constraints. So, from the perspective of household finance, enhancing farmers’ EDM is an urgent and important practical issue that needs to be addressed. Due to the important context of household financial status, we start with three indicators: liquidity management, insurance coverage, and investment management, and use an equal-weighted aggregation approach to measure financial resilience. This approach is mainly based on the functions of the three dimensions, which are all indispensable and important components for realizing financial resilience, with independent roles. Meanwhile, equal-weighted assignment is a common practice in related studies at home and abroad (Arrigoni et al., 2022; Greco et al., 2019; Wang and Zhang, 2024; Zhuang et al., 2025).

Specifically, firstly, liquidity assets represent the most flexible financial resources and are a crucial initial capital source for entrepreneurship. In the early instability stage of entrepreneurship, adequate liquidity plays a buffer role and increase farmers’ willingness to try entrepreneurship. Secondly, Cid-Aranda and López-Iturriaga (2025) supposed rural households usually face various risks such as illness, accidents, or natural disasters. By enabling ex-ante risk transfer, insurance system reduces the potential volatility caused by unexpected expenditures (Zhang et al., 2025). A good insurance coverage increases the probability of entrepreneurial opportunities without being overly concerned about the unforeseen events consequences. Thirdly, Duchek (2018) proposed that investment performance plays a supportive role in enhancing entrepreneurial capacity and confidence, acting as a guarantee for rural business. Thus, these three dimensions form a robust conceptual framework for understanding financial resilience and its positive influence on EDM is evident. Based on the above analysis, we propose the following hypothesis.

Hypothesis 1. Financial resilience promotes farmers’ EDM.

The impact mechanism of financial resilience on entrepreneurial decision-making

Financial resilience improves RT capacity, which in turn affects EDM. RT refers to the capacity to response with risk, encompassing both subjective and objective dimensions. First of all, based on prospect theory, it is known that sufficient liquid assets, comprehensive insurance coverage and financial gains can reduce the “risk psychological barrier” in farmers’ entrepreneurial choices, enabling farmers to perceive risks more accurately without overestimating potential obstacles (Brüggen et al., 2017). Next, financial resilience raises “the risk tolerance threshold” of households, liquidity provides a financial buffer, while investment performance shows the potential for long-term returns (Daadmehr, 2024; Hsu and Wang, 2023), all of which significantly increase the tolerance for losses, market volatility, and other negative outcomes. Especially in the post-epidemic era, where instability becomes the norm, Zhao et al. (2024) noted that financially resilient farmers are more likely to initiate new ventures and contribute to entrepreneurial happenings. Based on these theoretical insights, we propose the following hypothesis.

Hypothesis 2. Financial resilience promotes farmers’ EDM by enhancing their RT capacity.

Financial resilience increases SC, which in turn has an impact on EDM. Entrepreneurship is not only a reorganization of resources at the household level, but also a social behavior that is deeply relied on local social networks (Hechavarría and Brieger, 2022; Zulfiqar et al., 2021). Specifically, financial resilience improves farmers’ social sociability and economic credibility. In the acquaintance society, financial stability can be translated into social credit (Lucarelli et al., 2025; Zhao and Li, 2021), Guariglia et al. (2021) proposed that it will help enhance farmers’ bargaining power and voice within the local network. Moreover, financially resilient households tend to expand their social boundaries and build broader relationship networks, which increases their access to potential entrepreneurial opportunities (Tahir et al., 2022). These also enhance the reciprocity of farm households, which in turn enhances the possibility of entrepreneurial cooperation. Only with strong SC can farmers effectively maximize their information-sharing, resource-sharing, and joint RT, all of which indirectly promote EDM. Based on the above analysis, we propose the following hypothesis.

Hypothesis 3. Financial resilience promotes farmers’ EDM by enhancing SC.

The heterogeneity of gender groups

Gender differences also play a significant role in such household decision-making. In traditional rural societies, men are generally responsible for the livelihood of the family and are more engaged with external markets (Cabeza-García et al., 2019), while women are more involved in the internal affairs of the family. Many scholars have pointed out that women often bear a heavily burden of unpaid household labor (Asongu et al., 2020; Sakyi-Nyarko et al., 2022), including childcare and family caregiving (Kellard et al., 2024), which imposes time and energy constraints that limit their ability to pursue entrepreneurial activities. At the same time, as mothers or family caregivers, women tend to prioritize household stability, safety, and consumption smoothing in decision-making (Huang and Lin, 2022; Kellard et al., 2024), rather than engaging in proactive business behavior. Consequently, the effect of financial resilience on EDM may be more pronounced among male-headed households. However, Wang et al. (2025) noted that entrepreneurship represents a critical pathway for women’s economic empowerment. To promote gender equality and sustainable development, targeted policy is needed to reduce the barriers when women faced in starting entrepreneurship. So, we propose the following research hypothesis. The research theory logic of financial resilience and EDM are shown in Fig. 1.

Fig. 1
figure 1

Research logical framework.

Full size image

Hypothesis 4. The impact of financial resilience on EDM exhibits gender heterogeneity.

Data, model specification, and variables description

Data source

The data used in this paper were collected from a field survey targeting rural households, the survey was conducted by our research team in 2021 in Shaanxi Province. Shaanxi Province, located in west-central China, is one of the largest agricultural provinces. It features a representative “hills–plains–mountain” composite agro-geographical structure, which covering traditional agriculture and new agricultural entities, epitomizing the rural economy of developing countries (Xi et al., 2024). In recent years, Shaanxi has actively responded to China’s rural returnee entrepreneurship policies by offering strong support measures, such as entrepreneurship subsidies and incubation parks. The entrepreneurial atmosphere and environment are favorable (Ma et al., 2022), and it provides a solid foundation for observing entrepreneurial behaviors, which is of great significance for other rural areas in China and developing countries. Therefore, we chose Shaanxi Province as a typical investigation site to analyze the impact of financial resilience on farmers’ EDM.

We selected three major cities based on the representative landforms of “hills–plains–plateaus”, and further selected counties according to the “high–medium–low” principle of economic development. Specifically, in the northern Shaanxi Loess Plateau region, we chose Fuxian, Luochuan, and Huanglong in Yan’an City. In the central Shaanxi region, the selection included Yijun County in Tongchuan City, Fuping County in Weinan City, and Fufeng County in Baoji City. In the southern Shaanxi hills region, the selections encompass Zhashui County in Shangluo City, Hanyin County in Ankang City, and Yang County in Hanzhong City. Using the same method, we further identified 3 representative townships in each selected county and subsequently pinpointed 3 typical villages within each representative township. Subsequently, the sampling strategy was continually employed for selecting respondents in 20–30 sample households in each village, who were the decision-makers. In addition, the provincial and city-level distribution of the samples are shown in Fig. 2a, b.

Fig. 2: Distribution of the sample.
figure 2

a The distribution map of research area at the provincial level. b The distribution map of research area at the city level.

Full size image

Thus, the sample for the following analysis is highly representative after a series of rigorous and scientific sampling processes. After removing invalid questionnaires, a total of 942 questionnaires were obtained, covering 10 counties and 7 cities. The frequency distribution across counties is presented in Table 1, the average proportion of the surveyed sample in each county is around 10%, indicating a relatively balanced distribution.

Table 1 The frequency distribution across county.
Full size table

Model specification

To investigate the influence of financial resilience on farmers’ EDM, the following equation is formulated for the following empirical analysis:

$${{EDM}}_{{ij}}={\alpha }_{0}+{\alpha }_{1}\times {{FR}}_{{ij}}+{\alpha }_{2}\times {X}_{i}+{\alpha }_{3}\times {X}_{j}+{\varepsilon }_{{ijt}}$$
(1)
$${Prob}\left({{EDM}}_{{ij}}=1,|,{{FR}}_{{ij}},{X}_{i},{X}_{j}\right)=\Phi \left({\beta }_{0}+{\beta }_{1}\times {{FR}}_{{ij}}+{\beta }_{2}\times {X}_{i}+{\beta }_{3}\times {X}_{j}+{\varepsilon }_{{ijt}}\right)$$
(2)
$$\begin{array}{l}{M}_{{ij}}={\rho }_{0}+{\rho }_{1}\times {{FR}}_{{ij}}+{\rho }_{2}\times {X}_{i}\\\qquad+\,{\rho }_{3}\times {X}_{j}+{\mu }_{{ijt}}\end{array}$$
(3)
$$\begin{array}{l}{Prob}\left({{EDM}}_{{ij}}=1,|,{{FR}}_{{ij}},{X}_{i},{X}_{j}\right)\\\qquad\qquad\qquad=\Phi \left({\gamma }_{0}+{\gamma }_{1}\times {{FR}}_{{ij}}+{\gamma }_{2}\times {M}_{{ij}}+{\gamma }_{3}\times {X}_{i}+{\gamma }_{4}\times {X}_{j}+{\mu }_{{ijt}}\right)\end{array}$$
(4)

In this equation: \({{EDM}}_{{ij}}\) is the explained variable, standing for the EDM. \({{FR}}_{{ij}}\) is the explanatory variable, standing for the financial resilience. \({M}_{{ij}}\) represents RT (\({{RT}}_{{ij}}\)) and SC (\({{SC}}_{{ij}}\)). \({X}_{i}\) and \({X}_{j}\) are family and county characteristics. Equations (1) and (2) is basic regression model, Eqs. (3) and (4) are transmission intermediary models.

Variables description

  1. (1)

    Explained variables

    In this study, the explained variable is EDM. Combining the previous studies with the farmers’ realities, we define EDM as a binary variable, using whether they make a non-agricultural entrepreneurial decisions as a judgment criterion, assigning a value of 1 for yes and 0 for no.

  2. (2)

    Core explanatory variables

    The core explanatory variable is the financial resilience. To assess the level of financial resilience, a comprehensive measurement approach considering three key dimensions is employed, including household cash liquidity, insurance coverage, and investment performance. In this context, cash liquidity score is evaluated based on the ratio of current assets to monthly expenses (x), we set the passing threshold at 3 (equivalent to 60 points), with the mean of all samples denoted as x̅. Specific scoring criteria can be found in Table 3. Insurance coverage score is assessed based on whether there is sufficient insurance coverage, represented by the insurance coverage rate denoted as x. The mean insurance coverage rate for all samples is x̅, and the scoring criteria can be found in Table 3. Investment management score is measured according to the ratio of the total value of investment management to total assets, families in the study sample that utilize financial tools have all gained some returns. These three scores calculated above are equally weighted to finally determine the overall financial resilience level.

  3. (3)

    Mechanism variables

    In this study, there are two mechanism variables: RT and SC. To measure RT, we used the level of farmers’ ability to response risks, which is presented in the form of a five-point Likert. In measuring SC, we examine three key dimensions: social networks, social trust, and social reputation, including six items. According to the validity test results, the KMO value for the six items was 0.637, and the significance level of Bartlett’s test of sphericity was 0.000, which indicates that there is a good correlation between the measurement items. The specific items presented in Table 2.

    Table 2 Index system of social capital.
    Full size table
  4. (4)

    Control variables

    Control variables are chosen from individual-level, household-level, and regional-level factors separately. At the individual level, variables include age, educational level, marital status. Household-level factors encompass the arable land area, agricultural labors force, the property value, and the agricultural policy subsidies. Regional-level variables consist of the digital inclusive finance level. The detailed definitions, value assignment, and descriptive statistics for each variable are detailed in Table 3.

    Table 3 Variable descriptions and descriptive statistics.
    Full size table

Empirical results

Baseline results: financial resilience and entrepreneurial decision-making

The Probit model is employed to estimate the impact of financial resilience on farmers’ EDM, the regression results are presented in Table 4. It can be seen that after controlling the other factors at the individual level, household level and regional level, the coefficient of marginal effect of financial resilience on farmers’ EDM is significant at 1% confidence level, which is 0.329, that is, for every one-unit increase in the level of financial resilience of farmers, the likelihood of making an entrepreneurial decision rise by 32.9%. Hypothesis 1 has been validated. In terms of economic significance, this result reflects that farmers with greater financial resilience are more relax in the face of uncertainty in entrepreneurship, thereby being more inclined to start businesses. In addition, the control variables also exert significant effects on EDM. Specifically, the probability of farmers engaging in entrepreneurship increases with age up to a certain point and then declines, indicating an inverted U-shaped relationship; the area of arable land, the agricultural labor force, and the development of digital financial inclusion also have a significant impact on entrepreneurial decisions.

Table 4 Regression analysis results.
Full size table

Mechanism results: transmission of risk-taking and social capital

Risk-taking effects

To examine whether financial resilience can increase farmers’ RT capacity, the results are shown in Table 5. In column (2), it is evident that financial resilience is significantly positively correlated with farmers’ RT. For every one percentage point increase in financial resilience, the RT capability rises by 67.2%. That is to say, farmers with better financial resilience can withstand high-risk production and business projects easily. In column (3), it demonstrates a significant positive influence of RT on EDM, and compared with column (1) to add the mechanism variables, the estimated coefficients of financial resilience on EDM become smaller, suggesting that there is a part of mediation effect (McCorriston and MacLaren, 2024; Zhang and Liu, 2022), with the mediation effect estimated at 0.094 (0.672×0.140). The mediating effect of RT capacity as a mechanism variable is 28.59% (0.094/0.329). Therefore, the study posits that financial resilience can promote EDM by enhancing farmers’ RT capacity, confirming the existence of the causal mechanism. Hypothesis 2 has been validated.

Table 5 The risk-taking mechanism results.
Full size table

Social capital effects

To examine whether financial resilience can increase farmers’ SC, the results are shown in Table 6. In column (2), it is evident that financial resilience is significantly positively correlated with farmers’ SC. For every one percentage point increase in financial resilience, the SC improves by 54.4%. That is to say, financial resilience can enhance trust-based relationships (An et al., 2025; Crowley and Barlow, 2022; Deng et al., 2020), expand access to resources, and facilitate information exchange. In column (3), it demonstrates a significant positive influence of SC on EDM, and compared with column (1) to add the mechanism variables, the estimated coefficients of financial resilience on EDM become smaller, suggesting that there is a part of mediation effect, with the mediation effect estimated at 0.080(0.544 × 0.148). The mediating effect of RT capacity as a mechanism variable is 24.31% (0.080/0.329). By comparing the mediating effects, we find that the effect of RT plays a greater mediating role in the relationship between financial resilience and EDM. Therefore, the study posits that financial resilience can promote EDM by improving farmers’ SC, confirming the existence of the causal mechanism. Hypothesis 3 has been validated.

Table 6 The social capital mechanism results.
Full size table

Endogeneity tests and robustness tests

Endogeneity tests

To address endogeneity issues arising from reverse causation, omitted variables, measurement errors, and other unobservable factors, the study employs the instrumental variable (IV) method for two stages least squares estimation. The selection of instrumental variables must satisfy two conditions, including relevance and exogeneity. We choose the instrumental variable “Whether the household experienced the major natural disasters” because disasters are random in nature and do not directly determine entrepreneurship (de Blasio et al., 2021), but their experience of disasters makes households more wary of future uncertainty (Matsukawa et al., 2024), which in turn makes them more inclined to build up safety buffers, increase their savings rate, insurance coverage, and so on, and thus affects financial resilience. Consequently, it is calculated as a more suitable instrument variable for farmers’ financial resilience.

Table 7 presents the results of IV-Probit estimation. In the results of the first-stage regression, the F-statistic for the instrumental variable is 24.66, exceeding the critical value 10. This confirms that the selected instrumental variable does not suffer from a weak instrument problem (Wen et al., 2023). Moreover, the instrumental variable demonstrates a positive relationship with financial resilience. Specifically, the disaster shock experience is associated with a higher financial resilience for the respondent. This relationship is statistically significant at the 1% level, indicating a robust correlation between the selected instrumental variable and the explanatory variable. In the second-stage estimation results, the coefficient for financial resilience influencing farmers’ EDM is significantly positive. This indicates that even after accounting for endogeneity issues, the positive relationship between financial resilience and EDM is still prominent.

Table 7 IV-Probit analysis results.
Full size table

Robustness tests

(1) Reconstructing the Dataset. We have reconstructed a new dataset by excluding the top 5% and bottom 5% of the financial resilience levels. This step aims to verify the robustness of the above results. The specific results are shown in column (1) of Table 8, after excluding the highest level of 5% and the lowest level of 5%, the effect of financial resilience on the EDM of farmers is still significantly positive, which is consistent with the previous analysis.

Table 8 Robustness test results.
Full size table

(2) Replacing the Empirical Model. In the earlier text, the study employed Probit model for empirical analysis. To further enhance the robustness of the empirical findings, Logit model is used for additional empirical investigation. The specific results, as shown in column (2) of Table 8, demonstrate that using the Logit model, the influence of financial resilience on EDM remains significantly positive, ratio \({\rm{OR}}=\exp \left({\rm{\beta }}\right)=1.870 > 1\), it is clear that financial resilience exerts a positive influence on the likelihood of EDM, providing additional support for the robustness of the baseline regression findings in this study.

Heterogeneity analysis

During the field survey, we observed that a significant number of households in rural areas have female economic decision-makers. Therefore, we further analyzed the data by gender, so that to determine whether the impact of financial resilience on EDM is consistent between female decision-makers and male decision-makers, or if there are notable differences. The test results are presented in Table 9. Columns (1) and (2) of Table 9 present the results for the male group. Column (1) shows results without controlling for individual, household, and regional characteristics, while Column (2) includes these control variables. It can be observed that regardless of whether control variables are included, financial resilience has a significant positive effect on EDM. Males, as economic decision-makers, are more inclined to pursue entrepreneurship when the financial situation of the household is relatively stable (Chen et al., 2023; Coronel-Pangol et al., 2024; Guzman and Kacperczyk, 2019).

Table 9 Heterogeneity analysis results.
Full size table

Columns (3) and (4) show the results for the female subgroup. Compared to the male group, the coefficient of financial resilience is significantly smaller for the female group compared to the male group. This suggests that while financial resilience positively influences entrepreneurship for both genders, its role is relatively less pronounced among women. So, hypothesis 4 has been validated. Women usually tend to be more conservative compared to men, preferring to maintain financial stability. And women may place more emphasis on non-financial factors (Yuan et al., 2024), such as work-life balance or family responsibilities, when making entrepreneurial decisions, thereby reducing the relative entrepreneurial weight of financial resilience in their choices. However, women’s entrepreneurship is important for women as individuals, for their families and for society (Ahl et al., 2024; Harrison et al., 2020), so in the following we propose some measures to improve women’s entrepreneurship.

Conclusion and policy recommendations

Key conclusions

Based on survey data collected in Shaanxi Province in 2021, the study establishes a comprehensive financial resilience measurement framework encompassing three dimensions: household liquidity management, insurance coverage and investment management. It explores the mechanisms between financial resilience and EDM based on the core essence of financial resilience. The research findings indicate a significant positive impact of financial resilience on farmer’ EDM. In other words, the higher the level of financial resilience, the higher the probability of engaging in entrepreneurial activities. After solving the potential endogeneity issues of models, the research conclusion remains robust. The mechanisms between the two are as follows: Financial resilience can improve farmers’ RT ability and SC, thus facilitating EDM. A comparison of the two contributions shows that the RT effect plays a larger role than the SC effect.

Building upon these insights, an analysis of the heterogeneous impact of financial resilience on entrepreneurial decisions across different gender was conducted. The study reveals that, compared to men, women as household economic decision-makers are less likely to engage in entrepreneurial decisions under the same conditions of financial resilience.

Policy recommendations

Based on the above findings from theoretical and empirical analysis, the following policy recommendations are proposed:

Firstly, there is a pressing need to enhance farmers’ financial literacy and underscore the significance of financial resilience. Specifically, a collaborative mechanism should be established, spearheaded by local governments and financial institutions, with the participation of village-level organizations, to build a long-term framework for financial education. Furthermore, it is crucial to develop a comprehensive plan to promote risk management awareness among farmers. Local governments should take a leading role in actively exploring specialized financial literacy initiatives and tracking farmers’ financial resilience through scientific monitoring systems. Financial institutions should intensify efforts to develop community-level financial service points, innovate financial service models, and provide farmers with more targeted financial products. Of course, it is also important to develop gender-differentiated financial support policies. This will encourage farmers to make informed decisions regarding household asset allocation, thereby fostering wealth accumulation within the farming community, and bridge the entrepreneurial gender gap.

Secondly, there is an urgent imperative to augment financial investment in agricultural insurance and strengthen the rural social security system. Given the pivotal role of risk protection in bolstering financial resilience, rural social security organizations should expand the coverage of social insurance to enhance financial support. Moreover, tailored assistance measures and diversified elderly care services should be formulated. By ensuring farmers’ basic material needs and exploring avenues to enhance their endogenous development capacity. Lastly, the sustainable capacity building of farmers through training, policy support, and other innovative means should also be proactively developed. Only by fully leveraging the function of the entire social safety net can farmers feel confident in venturing into entrepreneurship and innovation activities.

Thirdly, to promote EDM among rural women, policies should aim at mitigate barriers and support their participation. Firstly, targeted financial support, such as microcredit and grants, should be provided to encourage women to pursue entrepreneurship while maintaining household financial stability. Secondly, entrepreneurial training and skill development for women can build confidence and practical expertise. Additionally, implementing family support initiatives, such as affordable childcare services, can alleviate caregiving responsibilities, enabling women to engage in entrepreneurial activities. Lastly, fostering community networks and mentorship opportunities can cultivate SC, motivating rural women to explore and sustain personal value.

Of course, this study has some limitations. First, we plan to expand the sample scope, and extend the research to multiple regions as an important direction. The further comparison with other provinces in China is also an important future study. Second, this study uses cross-sectional data, so we intend to use panel data or conduct longitudinal surveys to capture temporal dynamic effects. Finally, in the future, further analysis can be done around how to provide help for women’s entrepreneurship, exploring the barrier to women’s entrepreneurship through a combination of qualitative or case studies. Moreover, it is important to examine the broader significance of women’s entrepreneurial activities, particularly in terms of their economic, social, and empowerment impacts. Those are some areas that further investigation.