Hidden markets, visible gains: informal village-level farmland transfer platforms and rural household income

Abstract

Farmland transfer platforms serve as crucial vehicles for land circulation. However, empirical evidence of how informal village-level farmland transfer platforms affect farmers’ incomes remains unclear. Based on the data of 2,501 households in 12 provinces (districts), this paper analyzes in detail the impact of farmland transfer platform on farmers’ income and its mechanism. The study found that: First, the agricultural land transfer platform has a significant positive impact on farmers’ agricultural operating income and non-agricultural income. Second, heterogeneity analysis shows that for agricultural operating income, agricultural land transfer platform for high education level farmers and training Farmers has a greater impact. For non-agricultural income, the impact of farmland transfer platforms on low-education and non-training farmers is significantly positive. Third, further analysis of the mechanism shows that the agricultural land transfer platform affects the agricultural operation income and non-agricultural income of the farmers through the transfer area and the number of migrant workers, and the transfer area has a partial mediation effect, and the number of migrant workers has a complete mediating effect.

1 Research background

Effectively increasing farmers’ income is a core challenge for developing countries, essential for sustainable economic growth and social stability. It also plays a critical role in achieving balanced urban–rural development, reducing poverty, and advancing the comprehensive rural revitalization strategy (Davis et al., 2024; Wang et al., 2026). As the most significant resource owned by farmers, land plays a pivotal role in boosting their income by granting them more robust property rights (Bu and Liao, 2022; Li et al., 2025). On the one hand, land transfer-in enables the concentration of land toward farmers with higher productivity, thereby optimizing resource allocation (Chen et al., 2023; Jiang et al., 2023; Li and Wang, 2026). On the other hand, land transfer-out allows farmers to gain more free time for non-agricultural employment (Wang, J. et al., 2020; Yang and Wen, 2020). Meanwhile, households that transfer in land can also facilitate the non-agricultural labor transfer of surrounding farmers by consolidating land (Chen et al., 2021; Li et al., 2022). Thus, activating the land factor market is central to increasing farmers’ income (Ji et al., 2025; Talvi and Végh, 2005; Zhang et al., 2018b).

The Chinese government has continuously advanced farmland system reforms, providing a critical institutional foundation for the development of land transfer platforms. The 2014 “No. 1 Central Document” formally proposed the “three-right separation” reform of farmland ownership rights, contract rights, and management rights. While upholding collective ownership of rural land, this reform separated contract rights from management rights, clearing property rights obstacles for the orderly transfer of management rights. The 2018 revision of the Rural Land Contract Law further confirmed the legality of management rights transfers at the legislative level and protected the rights and interests of all parties involved. As of the end of 2018, the nationwide program for the registration and certification of rural land contract and management rights was largely completed, clarifying contractual relationships and laying the property rights cornerstone for the healthy development of the land transfer market.

Farmland transfer platforms, as carriers of land transfer, play a crucial role in revitalizing rural “dormant” resources and stimulating endogenous development vitality (Kan, 2021; Li and Ito, 2021). By enhancing information services and creating an information market, these platforms provide farmers with effective land transfer information and use contracts to regulate and safeguard the rights and obligations of both parties involved in the transfer (Li et al., 2017; Wu et al., 2015). Currently, many regions in China have established various forms of farmland transfer platforms to meet farmers’ land transfer needs. From the perspective of initiating entities, these platforms can be broadly categorized into two types: (1) informal farmland transfer platforms established by village collectives or agricultural associations, and (2) formal farmland transfer platforms set up by government agencies or related functional departments (Ge et al., 2018; He and Huang, 2024; Zhang et al., 2023). The first type, rooted in rural areas as self-organized informal entities, lacks independent legal person status and relies on village rules, customs, and trust embedded in acquaintance networks for its operation, thus functioning as endogenous intermediaries. Within this category, platforms led by village collectives are anchored to administrative villages and exhibit a stronger public character, whereas those led by cooperatives or agricultural associations primarily serve their members and carry a mutual-benefit orientation. Both types, however, remain outside the formal regulatory system and constitute informal institutional arrangements. For instance, in Nanlizhuang Village, Wenshang County, Jining City, Shandong Province—where we conducted fieldwork—the village collective established a dedicated farmland transfer platform. This platform not only assists farmers in selecting high-quality contractors but also facilitates land transfer-in for contractors while mediating land disputes between parties. The second type, typically located in urban areas, is registered with industrial and commercial or civil affairs authorities, possesses independent legal person status, operates under the supervision of higher-level competent authorities, and must comply with explicit transaction rules, thus functioning as exogenous intermediaries. Examples include the Rural Comprehensive Property Rights Transaction Center in Haicheng City, Liaoning Province, and the Rural Property Rights Transaction Center in Yulin City, Guangxi Province. Compared to formal urban platforms, informal rural platforms are deeply embedded in rural communities, possessing a better understanding of local farmland area, quality, and farmers’ transfer intentions. Moreover, within the “acquaintance society” of rural areas, these platforms are more likely to gain farmers’ trust. Additionally, informal platforms are more flexible, allowing them to quickly adjust rules and services based on local conditions, better meeting the specific needs of smallholder farmers and promoting the optimization and development of agricultural production (Fang et al., 2024). The above analysis suggests that activating land production factors impacts farmers’ income; however, whether informal rural farmland transfer platforms contribute to income growth requires further empirical validation.

In recent years, research on the impact of farmland transfer platforms on farmers’ income has been increasing. A substantial body of evidence has documented the broader benefits of farmland transfer. For instance, found that farmland transfer can effectively alleviate farmland abandonment in mountain regions by promoting the reallocation and optimal utilization of land resources (Shao et al., 2016). Further demonstrated a significant effect of farmland transfer on agricultural economic growth in China, with agricultural total factor productivity serving as a mediating channel in this relationship (Kuang et al., 2022). Scholars generally agree that land transfer is a vital pathway to improving agricultural productivity and farmers’ income (Fei et al., 2021; SHI, 2024; Zhang and Zhu, 2017). However, during the land transfer process, issues such as information asymmetry, high transaction costs, and low willingness of farmers to participate have constrained the development of the land transfer market (Chen et al., 2022; Li et al., 2020; Wu et al., 2023). To address these challenges, farmland transfer platforms have emerged as an essential mechanism for facilitating land transfer and increasing farmers’ income. In developed countries, farmland transfer platforms are relatively mature. For instance, in the United States and the European Union, well-established platforms have promoted agricultural modernization and large-scale operations by providing efficient transaction channels, which not only offer farmers convenient trading opportunities but also foster agricultural scale and modernization (Azadi et al., 2011; Gorgan and Hartvigsen, 2022). However, in China, the development of farmland transfer platforms is still in its early stages, lacking a mature operational mechanism and widespread application, particularly in rural areas. Informal platforms, due to their flexibility and proximity to grassroots communities, are prevalent but face numerous challenges, including insufficient policy support, lack of legality and standardization, information asymmetry, high transaction costs, and low farmer participation willingness. These factors collectively limit the platforms’ scalability and their effectiveness in promoting the land transfer market (Babu et al., 2015). Furthermore, informal platforms encounter operational challenges such as the absence of unified regulation, frequent transaction disputes, and difficulty competing with formal platforms, which hinders their potential to facilitate land transfer, optimize resource allocation, and increase farmers’ income (Jiali et al., 2021). These issues necessitate improvements through policy guidance, institutional enhancements, and informatization measures (Yang, 2024). Although informal village-level platforms are more likely to gain farmers’ trust within rural acquaintance societies, their legality and standardization still need strengthening. Current research predominantly focuses on the impact of government-led formal platforms on land transfer, which typically benefit from comprehensive legal and regulatory support and policy guidance, ensuring transparency in transactions and protecting farmers’ rights. However, empirical studies on the impact of informal rural platforms on farmers’ income remain relatively scarce. These informal platforms, often operating based on local customs and social networks, may better align with farmers’ actual needs and flexibly adapt to market changes. Exploring informal village-level platforms can enable more smallholder farmers to participate in broader land markets, reduce transaction costs, and increase trading opportunities, thereby generating greater economic benefits for farmers.

In summary, this study aims to empirically examine the impact of informal village-level farmland transfer platforms on farmers’ income, providing theoretical and policy insights for improving the land transfer market and increasing farmers’ income. To this end, the study employs micro-level household survey data to analyze the impact and mechanisms of informal village-level platforms on farmers’ income. The study extends existing research in the following aspects: First, it provides new micro-level evidence on the income effects of informal village-level platforms, an underexplored but important institutional arrangement in rural land markets, by disaggregating farmers’ income into agricultural income and non-agricultural income to offer a more holistic view of their economic effects. Second, to address the endogeneity issues of informal platforms, this study integrates the Instrumental Variable (IV) method, Propensity Score Matching (PSM), and Treatment Effect Model, ensuring the robustness and reliability of causal inference. Third, by introducing a mediation effect model, the study further explores the mechanisms through which informal platforms affect farmers’ income, revealing their potential impact pathways via optimizing resource allocation, reducing transaction costs, and facilitating non-agricultural employment. The analysis provides new theoretical and empirical support for policy optimization of informal village-level platforms.

The structure of this paper is as follows: Section 2 provides a theoretical analysis of the impact of informal village-level farmland transfer platforms on farmers’ income; Section 3 details the data sources, variable selection, descriptive statistical analysis, and model specification; Section 4 presents the empirical results and analysis, including baseline regression, heterogeneity analysis, robustness checks, endogeneity issues, and mechanism analysis; Section 5 offers conclusions and policy implications.

2 Mechanism analysis of the impact of village-level farmland transfer platforms on farmers’ income

Property rights serve as the foundation for transactions, while platforms ensure their efficiency and smooth execution. Currently, with the implementation of the “three-right separation” reform and land titling in China, farmland property rights have become more clearly defined, and the land transfer rate has been steadily increasing. According to statistics from the Ministry of Agriculture and Rural Affairs, by the end of 2023, China’s land transfer area reached 680 million mu (approximately 45.33 million hectares), with a land transfer rate of 36%. Consequently, informal village-level farmland transfer platforms have proliferated, making it both theoretically and practically significant to evaluate their impact on farmers’ income in a timely manner. Given the diversification of rural income sources, and based on field survey findings, the primary income sources for rural residents mainly include agricultural income and non-agricultural income. This study focuses on analyzing the impact of informal village-level farmland transfer platforms on farmers’ agricultural operating income and non-agricultural income.

This study constructs an economic theoretical model to analyze the impact of informal village-level farmland transfer platforms on farmers’ income, as illustrated in Figure 1. First, we assume the existence of a perfect land market where information is fully symmetric and farmland transaction costs are zero. In this scenario, the market supply curve is S0, the demand curve is D0, and the equilibrium point is E0. However, in reality, farmland markets are characterized by low transaction volumes and a high number of small-scale transactions, compounded by the endowment effect of farmers’ land ownership. As a result, farmland transaction costs are often high, causing the supply curve to shift upward from S0 to S1, the demand curve to shift downward from D0 to D1, and the corresponding equilibrium point to move to E1. Consequently, the farmland market transaction volume decreases from Q0 to Q1, and the transaction price rises from P0 to P1. To reduce farmland transaction costs and mitigate information asymmetry, farmland transfer platforms have emerged. Informal village-level farmland transfer platforms provide a venue for both supply and demand sides of farmland transactions, regulating and standardizing the behavior of both parties. This leads to an increase in both farmland supply and demand, with the supply curve shifting downward from S1 to S2, the demand curve shifting upward from D1 to D2, and the corresponding equilibrium point moving to E2. At this point, the farmland market transaction volume increases from Q1 to Q2, and the transaction price decreases from P1 to P2. In other words, the emergence of informal village-level farmland transfer platforms reduces farmland transaction prices and increases transaction volumes, thereby facilitating both land transfer-in and transfer-out. From the perspective of farmers, land transfer-out provides more time for non-agricultural employment, while land transfer-in enables the expansion of operational scale, achieving economies of scale.

Figure 1

The impact of informal village-level farmland transfer platforms on farmers’ income primarily operates through two channels: land transfer-in and off-farm employment. On the one hand, due to the fragmentation of land among smallholder farmers, in the absence of transfer platforms, households seeking to transfer in land to expand their operational scale face significant search, information, bargaining, decision-making, monitoring, and default costs, which hinder land transfer and prevent optimal resource allocation (He and Huang, 2024). Informal village-level farmland transfer platforms provide transfer-in households with more farmland information sources, reducing their information search and screening costs. Moreover, these platforms standardize pricing, contract signing, monitoring, and default processes during land transfer, enabling transfer-in households to acquire land with greater confidence while mitigating risks such as default or opportunistic behavior (e.g., price gouging) by transfer-out households (Wang and Guo, 2011). Under conditions of clear land property rights and regulated platforms, farmers are more likely to expand production scales and make long-term agricultural investments (Huang and Ji, 2012; Zhang and Zhu, 2017). Therefore, farmland transfer platforms can increase farmers’ agricultural operating income by facilitating larger land transfer-in areas and encouraging long-term investments. On the other hand, informal village-level farmland transfer platforms provide transfer-out households with more transfer information, reducing their information search and screening costs (Kowalczyk et al., 2019; Zhang et al., 2023). In China’s rural society, characterized by “acquaintance networks,” while such networks facilitate land transfer-out, they also lead to transfer-out households being unable to secure genuine market-based land transfer prices due to social obligations (e.g., “face” considerations). This phenomenon is evident in the “zero-rent” and hierarchical patterns observed in land transfers by many scholars (Jiang et al., 2023). Informal village-level farmland transfer platforms offer standardized transfer contracts for potential transfer-out households, reflecting true market land prices while reducing monitoring and default costs (Zhang et al., 2017). Consequently, these platforms enable farmers who are reluctant to transfer land due to transfer risks, low transfer prices, or high information search and screening costs to do so with greater confidence. Furthermore, farmland transfer platforms allow farmers to transfer out land on a long-term basis, providing more leisure time for off-farm employment, reducing the opportunity cost of staying in rural areas versus migrating to urban areas, and ultimately increasing their non-agricultural income. Additionally, in rural areas, land transfer-in typically prioritizes neighboring areas, generating positive spillover effects: by transferring in land from surrounding farmers, it enables those farmers to gain more leisure time for off-farm employment.

In summary, informal village-level farmland transfer platforms activate the land transfer market by reducing transaction costs and standardizing land transfer practices, thereby increasing farmers’ income through land transfer-in or off-farm employment opportunities.

3 Data, variables, and model

3.1 Data sources

The data for this study were collected during the winter vacation survey conducted by the National Institute of Agricultural and Rural Development, China Agricultural University, from January to February 2018. The survey team primarily consisted of undergraduate, master’s, and doctoral students from various disciplines at China Agricultural University, where all programs are related to agriculture. This background ensured the surveyors’ familiarity with agricultural contexts, thereby guaranteeing the quality of the questionnaires. Prior to the survey, participants underwent specialized training to address difficulties and ambiguities in the questionnaire design. The survey included household questionnaires and village questionnaires, with each village requiring a random sample of 10–20 households. This effort yielded 2,553 household questionnaires and 159 village questionnaires. During data processing, the household and village questionnaires were matched, and samples with inconsistent matches or significant missing values were excluded, resulting in a final dataset of 2,501 household questionnaires. These questionnaires were sourced from 12 provinces (autonomous regions), 136 counties (districts), and 159 villages. The 12 provinces (autonomous regions)—including eastern Inner Mongolia, Jilin, Sichuan, Anhui, Shandong, Jiangsu, Jiangxi, Hebei, Henan, Hubei, Hunan, and Heilongjiang—cover China’s eastern, central, western, and northeastern regions, offering strong representativeness. The geographical distribution of the sample regions is illustrated in Figure 2.

Figure 2

3.2 Variable selection and descriptive statistics

3.2.1 Dependent variable—farmers’ income

Farmers’ income sources have become diversified, encompassing agricultural operating income, non-agricultural income (primarily wage income), pensions and subsidies, property income, minimum living allowance income, and rural pension insurance income, among others. Given that farmland transfer platforms most directly influence agricultural operating income and non-agricultural income, this study selects these two components as indicators of farmers’ income.

3.2.2 Core independent variable—informal village-level farmland transfer platform

Farmland transfer platforms reduce transaction costs, mitigate risks and uncertainties in land transfer, and thereby promote land circulation. These platforms can be provided by village collectives, cooperatives, or other intermediary organizations. In the questionnaire design, respondents were directly asked whether their village had a land transfer platform. Among the 159 villages, 33 had transfer platforms, while 126 did not, corresponding to 2,501 household samples, with 544 households from villages with platforms and 1,957 from villages without platforms.

3.2.3 Mediating variables

Since farmland transfer platforms affect the area of land transferred in and the non-agricultural transfer of labor, this study selects land transfer-in area and labor non-agricultural transfer as mediating variables. Given that labor non-agricultural transfer is significantly influenced by the number of family laborers, following the approach of Hong W. and Hu X (Hong and Hu, 2019), labor non-agricultural transfer is defined as the number of off-farm workers divided by the number of family agricultural laborers.

3.2.4 Control variables

This study controls for other variables affecting farmers’ income, including age, health status, education level, training participation, scope of operations, whether the village is classified as a well-off village, distance to highways, water resource security, economic level, and topographic features. The descriptive statistical analysis of the dependent variables, core independent variable, and control variables is presented in Table 1.

3.3 Model specification

3.3.1 OLS estimation and Tobit model

Since farmers’ agricultural operating income and non-agricultural income are continuous variables, the basic Ordinary Least Squares (OLS) regression is initially employed. The specific form is as follows:

Where represents farmers’ income (either agricultural operating income or non-agricultural income), denotes the presence of an informal village-level farmland transfer platform, is a vector of control variables (including age, health status, education level, training participation, scope of operations, whether the village is classified as a well-off village, distance to highways, water resource security, economic level, and topographic features), represents regional dummy variables, is the constant term, is the coefficient of the informal village-level farmland transfer platform, is the coefficient vector of control variables, is the coefficient of regional dummy variables, and is the random error term.

Given that agricultural operating income and non-agricultural income may include zero values, the Tobit model is further adopted to ensure the robustness of the results. The specific forms are presented in Equations 2, 3 as follows:

Where represents the latent agricultural operating income and non-agricultural income, and the other variables remain consistent with those described above.

3.3.2 Propensity score matching and treatment effect model

In real-world studies, data bias and confounding variables often lead to estimation bias. Propensity Score Matching (PSM) effectively addresses these issues by constructing counterfactuals. First, a logit model is used to estimate the probability of a village having an informal farmland transfer platform and to calculate the propensity score, as shown in Equation 4:

Next, matching methods—including nearest neighbor matching, kernel matching, local linear matching, radius matching, and Mahalanobis distance matching—are applied to pair the treatment group (villages with platforms) and the control group (villages without platforms). Finally, based on the matched samples, the average difference in farmers’ income between the treatment and control groups is calculated to evaluate the impact of informal village-level farmland transfer platforms on farmers’ income, as shown in Equation 5 and Table 2.

Where ATT represents the Average Treatment Effect on the Treated (ATT), D is a binary variable (1 for the treatment group, 0 for the control group), P(x) is the propensity score, and Y₁ and Y₀ represent the estimated outcomes for farmers with and without informal farmland transfer platforms, respectively.

Since PSM relies heavily on observable variables for matching, the selection of these variables significantly affects the results. If too few variables are chosen or if they lack strong relevance, the results may be biased. To address this, Maddala (Maddala, 1983) proposed the Treatment Effect Model based on the Heckman model, which accounts for unobservable variables in sample matching. The specific equations are as shown in Equations 6, 7:

Among them, Equation 7 is the treatment equation, is the latent variable, > 0, and 0 otherwise. includes at least one variable not present in ; specifically, two additional variables—“road proportion” and “number of cooperatives” in the village—are included in .

Assuming the covariance of the error terms and is zero, they follow a bivariate normal distribution, as shown in Equation 8:

In Equation 1, is the correlation coefficient between , with the variance of standardized to 1. If , the model exhibits endogeneity; if , no endogeneity exists.

3.3.3 Mediation effect model

Since informal village-level farmland transfer platforms may influence farmers’ income through specific pathways or mechanisms, this study adopts the mediation effect model proposed by Baron and Kenny (Baron and Kenny, 1986) and employs a stepwise regression approach for testing. Given that informal platforms directly affect farmers’ off-farm employment and land transfer-in area, “labor non-agricultural transfer” and “land transfer-in area” are selected as mediating variables. The model consists of Equations 9–11. First, farmers’ income is regressed on the presence of an informal village-level farmland transfer platform, as shown in Equation 9. Second, the platform is regressed on the mediating variables, as shown in Equation 10; if the coefficient is significant, the analysis proceeds to the next step; if not, the test is terminated. Finally, farmers’ income is regressed on both the platform and the mediating variables, as shown in Equation 11. Following the testing procedure outlined by Wen et al. (2004), the analysis proceeds as follows: (1) Test in Equation 9; if significant, proceed to the next step; if not, terminate the analysis. (2) Test and in Equations 10, 11; if both are significant, proceed to the next step; if at least one is not significant, move to step four. (3) Test in Equation 11; if not significant, this indicates a full mediation effect, meaning the platform’s impact on farmers’ income is entirely through the mediating variables; if significant, this indicates a partial mediation effect, meaning only part of the platform’s impact on income operates through the mediating variables. (4) Conduct the Sobel test; if the result is significant, a mediation effect exists; if not, the test concludes.

Where represents the mediating variables (“labor non-agricultural transfer” and “land transfer-in area”), and the control variables and regional dummy variables are consistent with those described above.

4 Empirical results and analysis

The primary objective of this study is to analyze the impact of informal village-level farmland transfer platforms on farmers’ income. The analysis proceeds as follows: First, OLS estimation is used to examine the effect of informal village-level farmland transfer platforms on farmers’ agricultural operating income and non-agricultural income. Second, robustness tests are conducted using variable substitution, sample adjustment, model substitution, and village-level data. Third, the heterogeneity of the impact of informal village-level farmland transfer platforms on farmers’ income is analyzed from the perspectives of education level and training participation. Fourth, the Instrumental Variable (IV) method, Propensity Score Matching (PSM), and Treatment Effect Model are employed to address endogeneity issues. Fifth, a mediation effect model is utilized to further explore the mechanisms through which informal village-level farmland transfer platforms affect farmers’ income.

4.1 Impact of informal village-level farmland transfer platforms on farmers’ income

Prior to regression, a multicollinearity test is conducted, revealing a Variance Inflation Factor (VIF) of 1.88, indicating no multicollinearity. The regression results are presented in Table 3. Column (1) includes only the core variable, Column (2) adds control variables, and Column (3) further incorporates control variables and regional dummy variables. The results show that informal village-level farmland transfer platforms have a significant positive impact on both agricultural operating income and non-agricultural income. This suggests that farmers in villages with farmland transfer platforms have significantly higher incomes than those in villages without such platforms. Further comparison reveals that the impact coefficient of informal platforms on agricultural operating income is higher than on non-agricultural income, with significance at the 1 and 5% levels, respectively. This indicates a stronger effect on agricultural operating income, likely because informal platforms primarily facilitate land transfer, reallocating land from farmers unwilling or unsuitable for farming to those capable and motivated to improve productivity, thereby achieving efficient land resource allocation and directly enhancing economies of scale in agricultural production.

Regarding control variables, the coefficient for age is negative, indicating that older farmers have lower incomes. The health status coefficient is negative, suggesting that poorer health reduces agricultural operating income. The education level coefficient is negative in the agricultural operating income equation but positive in the non-agricultural income equation, possibly because higher education levels lead farmers to prefer non-agricultural employment over farming. Education not only enhances farmers’ occupational choice capabilities but also improves their awareness and adaptability to non-agricultural job opportunities (Zhang et al., 2018a; Zhu et al., 2024). The training coefficient in the agricultural operating income equation is positive, indicating that farmers who receive training have higher agricultural operating income. Agricultural training significantly improves production efficiency and economic returns by imparting modern agricultural techniques, crop management methods, and market knowledge (Mgendi et al., 2022; Wonde et al., 2022). Compared to pure crop farmers, those engaged in integrated crop-livestock farming or crop-livestock-tourism operations have higher agricultural operating income. The coefficient for well-off villages is negative, possibly because, compared to well-off villages, farmers in non-well-off villages are more likely to engage in agricultural or non-agricultural work. Due to greater economic pressure and resource constraints, farmers in non-well-off villages often adopt diversified livelihood strategies to mitigate risks and increase income (Peng et al., 2022). The distance to highways coefficient is negative, indicating that more remote locations lead to lower non-agricultural income. Adverse geographic conditions often result in insufficient agricultural income to sustain households, prompting farmers to migrate to cities or other areas for non-agricultural job opportunities (Hao and Tang, 2023). The water resource security coefficient is negative in the agricultural operating income equation but positive in the non-agricultural income equation, possibly because regions with better water resources are more economically developed, leading farmers to engage more in non-agricultural employment and less in farming. In reality, water-rich regions can support diverse economic activities, including industry, manufacturing, and services, which typically offer higher wages and more stable employment opportunities than agriculture (Cui et al., 2025; Liu et al., 2024). The economic level coefficient is negative, indicating that lower local economic development results in lower farmers’ income. Compared to plains, farmers in mountainous areas have lower agricultural operating income, while those in mountainous and other regions have higher non-agricultural income, likely because harsher geographic conditions force farmers to seek off-farm employment. Adverse geographic conditions, such as poor soil, water scarcity, or rugged terrain, significantly reduce agricultural productivity, compelling farmers to pursue alternative income sources through non-agricultural employment (Liu et al., 2022; Lyu et al., 2019).

4.2 Heterogeneity analysis

The OLS estimation analyzed the impact of informal village-level farmland transfer platforms on farmers’ income; however, these results reflect only the average effect and do not account for the heterogeneity of this impact. This section examines heterogeneity from two perspectives: farmers’ education level and training participation. In the survey, education levels were categorized into six levels. Considering the rural context, this study classifies farmers with a junior high school education or below as having a low education level, and those with a high school education or above as having a high education level. The results are presented in Tables 4, 5.

From the perspective of education level, farmers with different levels of human capital may differ in their efficiency in acquiring and utilizing land transfer information, which in turn shapes their income growth pathways. For the agricultural operating income equation, informal village-level farmland transfer platforms have a significant positive impact on both low- and high-education farmers. However, the impact coefficient is larger for high-education farmers, possibly because they are more adept at managing agricultural operations and can better leverage transfer platforms to enhance agricultural operating income. High-education farmers typically possess stronger technology adoption capabilities and market insights, enabling them to optimize land allocation and production strategies more effectively, thus achieving higher agricultural income (Ji et al., 2025; Liu et al., 2022; Zhang et al., 2024). For the non-agricultural income equation, informal platforms have a significant positive impact on low-education farmers but not on high-education farmers. This may be because low-education farmers can transfer out land through the platform, enabling them to engage in non-agricultural employment, whereas high-education farmers already have greater capabilities and opportunities for non-agricultural employment, rendering the platform’s impact on their non-agricultural income negligible (Yu and Sloan, 2017).

From the perspective of training participation, skill accumulation is directly linked to farmers’ production efficiency and their capacity to make use of farmland transfer platforms, and may therefore moderate the effects of such platforms on income. For the agricultural operating income equation, informal village-level farmland transfer platforms have a significant positive impact on both trained and untrained farmers. However, the impact coefficient for trained farmers is significantly higher than for untrained farmers, indicating that training has a more pronounced effect on agricultural operating income. Agricultural training enhances farmers’ technical skills and production efficiency, improving their ability to utilize transfer platforms and significantly boosting agricultural income (Gebresilasse, 2023; Mgendi et al., 2022). Training typically includes knowledge of modern agricultural techniques, crop management, and market dynamics, enabling farmers to better integrate land resources and optimize production (Nakano et al., 2018). For the non-agricultural income equation, the platform’s impact on the non-agricultural income of trained farmers is not significant, while it has a significant positive impact on untrained farmers. This may be because training primarily focuses on agricultural production, thus having little effect on non-agricultural income. In contrast, untrained farmers are more likely to be primarily engaged in non-agricultural employment and thus do not participate in agricultural training, while the platform facilitates land transfer-out, further promoting their non-agricultural employment.

4.3 Endogeneity issues

4.3.1 Instrumental variables

While the preceding analysis demonstrates a significant positive impact of informal village-level farmland transfer platforms on farmers’ income, it does not account for the potential endogeneity of these platforms. Generally, when farmers in a region have higher agricultural income, there may be greater demand for land transfer-in, and when non-agricultural income is higher, there may be increased demand for land transfer-out, both of which can stimulate the development of local farmland transfer markets. This creates a reverse causality issue, leading to endogeneity. Failure to address endogeneity can result in biased estimates. To tackle this, this study selects “whether the area is suburban” and “road proportion” as instrumental variables for informal village-level farmland transfer platforms. Theoretically, rural development levels differ between suburban and non-suburban areas, affecting the demand for informal farmland transfer platforms. Suburban areas, which typically have better economic development, tend to have more farmland transfer markets. Current policies also prioritize regions with higher economic development as pilot areas to promote farmland transfer markets. Similarly, road proportion reflects the level of regional economic development to some extent; a higher road proportion increases the likelihood of a region having a farmland transfer market. However, whether informal village-level farmland transfer platforms exhibit endogeneity and whether “whether the area is suburban” and “road proportion” are valid instrumental variables require further empirical validation.

As shown in Tables 2, 6, for both agricultural operating income and non-agricultural income, the Hausman test is significant at the 10% level, and the heteroskedasticity-robust Durbin–Wu–Hausman (DWH) test is significant at the 5% level, indicating the presence of endogeneity in informal village-level farmland transfer platforms. For the weak instrument test, Table 2 shows that the F-statistic is 21.502, significant at the 1% level and exceeding the critical value under a 15% bias threshold, suggesting no weak instrument issue. To ensure robustness, the Limited Information Maximum Likelihood (LIML) method is further applied, and the results are similar to those of Two-Stage Least Squares (2SLS), further confirming the absence of weak instruments. Similarly, Table 6 shows that for both agricultural operating income and non-agricultural income equations, the F-statistic is 13.405, significant at the 1% level and exceeding the critical value under a 15% bias threshold, indicating no weak instrument issue. LIML estimation yields results consistent with 2SLS, reinforcing the absence of weak instruments.

Considering the presence of heteroskedasticity, Generalized Method of Moments (GMM) is more efficient than 2SLS. This study further employs GMM and iterative GMM estimation. Regardless of the estimation method used, informal village-level farmland transfer platforms exhibit a significant positive impact on both agricultural operating income and non-agricultural income, with estimated coefficients that are relatively consistent, further confirming the robustness of the results.

4.3.2 Propensity score matching

Although the instrumental variable approach can address endogeneity issues such as reverse causality, the sample often suffers from data bias and confounding variables. To mitigate these issues, Propensity Score Matching (PSM) is employed by constructing counterfactuals, as shown in Table 7.

This study applies nearest neighbor matching, kernel matching, local linear matching, radius matching, and Mahalanobis distance matching to estimate the Average Treatment Effect on the Treated (ATT) for farmers with and without informal village-level farmland transfer platforms (treatment and control groups). For both agricultural operating income and non-agricultural income equations, the ATT is significantly positive, indicating that informal village-level farmland transfer platforms significantly increase farmers’ income. Further examination of the ATT coefficients reveals that the ATT for the agricultural operating income equation is significant at the 1% level and generally higher than that for the non-agricultural income equation, suggesting a greater impact on agricultural operating income.

4.3.3 Treatment effect model

Since PSM heavily relies on observable variables for matching, insufficient or weakly correlated variables may lead to biased results. To address this, the Treatment Effect Model is further employed, as shown in Table 8.

This study adopts two estimation methods: the Two-Step method and Maximum Likelihood Estimation (MLE). The Two-Step method’s hazard-lambda test (λ) and the MLE’s athrho test () are both significant at the 1% level, indicating that the informal village-level farmland transfer platform is an endogenous dummy variable and that the Treatment Effect Model outperforms OLS estimation. In terms of coefficients, for both agricultural operating income and non-agricultural income equations, the coefficient of the informal platform is significantly positive at the 1% level, consistent with the results from OLS, PSM, and instrumental variable methods.

4.4 Robustness tests

To ensure the reliability of the findings, this section conducts robustness tests through variable adjustments, sample adjustments, model adjustments, and village-level data analysis.

Variable Adjustment: The study aggregates agricultural operating income and non-agricultural income, as well as total household income (excluding pensions, rural pension insurance income, etc., due to their limited relevance to transfer platforms), as dependent variables for analysis. The results, presented in Table 9, show that the coefficient of informal village-level farmland transfer platforms is significantly positive at the 1% level.

Sample Adjustment: Given the distinct geographic environment, agricultural production conditions, and national agricultural policy priorities for Northeast China compared to the eastern, central, and western regions, the study excludes Northeast samples (Jilin and Heilongjiang). The results, shown in Table 10, indicate that the coefficient of informal village-level farmland transfer platforms remains significant at the 1% level. Further excluding western region samples (Inner Mongolia and Sichuan), the conclusions remain robust.

Model Adjustment: Since agricultural operating income and non-agricultural income include zero values (i.e., censored dependent variables), the Tobit model is employed for regression. As shown in Table 11, the coefficient of informal village-level farmland transfer platforms is significantly positive at the 5% level, consistent with the earlier findings.

To further ensure robustness, village-level data are analyzed, as presented in Table 12. Villages with informal farmland transfer platforms exhibit significantly higher per capita net income than those without, alongside higher village-level land transfer ratios, non-agricultural employment proportions, numbers of large-scale farmers, maximum farm sizes, and numbers of cooperatives. This further confirms that farmland transfer platforms increase farmers’ income, promote land transfer, and facilitate non-agricultural employment.

4.5 Mechanisms of the impact of informal village-level farmland transfer platforms on farmers’ income

The preceding analysis has demonstrated a significant positive impact of informal village-level farmland transfer platforms on farmers’ income. This section explores the mechanisms through which these platforms influence farmers’ income—that is, why they have such an impact—using a mediation effect model.

Logically, in villages with informal farmland transfer platforms, farmers can more easily transfer in land, thereby increasing agricultural operating income. These platforms facilitate land transfer-in, enabling farmers to expand their operational scale, benefit from economies of scale, and enhance agricultural productivity and income (Fang et al., 2024; Feng et al., 2010; Wang, L. et al., 2020). Additionally, farmers can transfer out land through these platforms, freeing up more time for non-agricultural employment and increasing non-agricultural income. Therefore, this study selects “land transfer-in area” and “labor non-agricultural transfer” as mediating variables, with detailed results presented in Table 13. Regarding the mechanism of the impact on agricultural operating income, according to Equation 9, informal village-level farmland transfer platforms have a significant positive impact on farmers’ operating income. According to Equation 10, the coefficient of the platform on land transfer-in area is 0.430, significant at the 1% level. According to Equation 11, the coefficients of both the platform and land transfer-in area are positive and significant at the 5% level, indicating a partial mediation effect. This suggests that the platform’s impact on agricultural operating income is partially realized through the land transfer-in area. Regarding the mechanism of the impact on non-agricultural income, according to Equation 9, the platform has a significant positive impact on non-agricultural income. According to Equation 10, the coefficient of the platform on labor non-agricultural transfer is 0.129, significant at the 1% level. According to Equation 11, the platform’s coefficient is positive but not significant, while the coefficient of labor non-agricultural transfer is positive and significant at the 5% level, indicating a full mediation effect. This suggests that the platform’s impact on non-agricultural income is fully realized through labor non-agricultural transfer.

To further validate these findings, the number of off-farm workers is used as a substitute for labor non-agricultural transfer, and total household income replaces agricultural operating income to ensure robustness. The results, shown in Table 14, confirm that the platform’s impact on total household income is partially mediated through land transfer-in area, while its impact on non-agricultural income is mediated through labor non-agricultural transfer, further demonstrating the robustness and reliability of the conclusions.

5 Conclusions and policy implications

5.1 Conclusion

Using data from 2,501 households across 12 provinces (autonomous regions), 136 counties (districts), and 159 villages, collected by the National Institute of Agricultural and Rural Development at China Agricultural University in 2018, this study employs OLS estimation, Tobit models, Propensity Score Matching (PSM), Treatment Effect Models (TEM), and mediation effect models to investigate the impact of informal village-level farmland transfer platforms on farmers’ agricultural operating income and non-agricultural income.

The results show that the findings are as follows: First, informal village-level farmland transfer platforms have a significant positive impact on both agricultural operating income and non-agricultural income, with a stronger effect on agricultural operating income. Accounting for the endogeneity of these platforms, the Instrumental Variable (IV) method, Propensity Score Matching (PSM), and Treatment Effect Model (TEM) results all confirm a significant positive impact on farmers’ income. Second, robustness tests through variable adjustments, sample adjustments, model adjustments, and village-level data analysis demonstrate the consistency of the results. Third, heterogeneity analysis reveals that for agricultural operating income, informal platforms have a greater impact on farmers with higher education levels and those who have received training, while for non-agricultural income, the platforms significantly benefit farmers with lower education levels and those without training, but the effect is not significant for farmers with higher education levels or those who have received training. Fourth, mechanism analysis indicates that informal village-level farmland transfer platforms influence agricultural operating income and non-agricultural income through land transfer-in area and labor non-agricultural transfer, respectively, with land transfer-in area showing a partial mediation effect and the number of off-farm workers exhibiting a full mediation effect.

5.2 Policy implications

Based on the above analysis, this study provides a robust theoretical and practical foundation for understanding the role of informal village-level farmland transfer platforms in enhancing farmers’ income, offering policy recommendations in three key areas.

First, the construction of farmland transfer platforms should be advanced in a classified and sequential manner. The findings demonstrate that informal village-level farmland transfer platforms significantly increase farmers’ income; however, most villages currently lack such platforms. Therefore, it is essential to accelerate the construction of farmland transfer platforms. Pilot projects should be initiated in villages with relatively high levels of economic development and strong farmer willingness to transfer land. In these pilots, village collectives or cooperatives should take the lead in establishing village-level land transfer service stations, with three clearly defined basic functions: information dissemination, contract verification, and dispute mediation. Based on the pilot experience, such stations should be gradually extended to surrounding villages, ultimately forming a land transfer service network covering the village, township, and county levels. For villages with a weaker economic base, land transfer service functions can be embedded into existing cooperatives or village committees at lower cost, thereby avoiding redundant institutional construction.

Second, a differentiated institutional safeguard system should be established. Given that informal farmland transfer platforms are operated by diverse entities such as village collectives and cooperatives, classified management policies should be introduced. Specifically, platforms led by village collectives should be registered with the township government, which should conduct regular inspections of contract compliance, while platforms led by cooperatives should be required to report their transfer contracts to the village committee for record. Meanwhile, county-level agricultural and rural affairs departments should formulate a unified model contract for farmland transfers, specifying core clauses such as transfer duration, price adjustment mechanisms, and default handling, to standardize the transaction practices of various types of platforms and reduce the performance risks arising from oral agreements.

Third, accessible mechanisms for dispute mediation and risk prevention should be established. As the scale of land transfer expands, transfer-related disputes are on the rise. A land transfer dispute mediation window should be set up in township judicial offices, jointly staffed by judicial mediators, village cadres, and villager representatives, to accept and resolve transfer disputes on the spot. Simultaneously, a land transfer risk guarantee fund, jointly financed by county and township governments and village collectives, should be established to advance transfer fees to transfer-out households in cases where their interests are harmed due to the withdrawal of the operating entity, thereby effectively protecting the rights and interests of smallholder farmers.

References

• 1

  AzadiH.HoP.HasfiatiL. (2011). Agricultural land conversion drivers: a comparison between less developed, developing and developed countries. Land Degrad. Dev.22, 596–604. doi: 10.1002/ldr.1037

  • CrossRef
  • Google Scholar

• 2

  BabuS. C.HuangJ.VenkateshP.ZhangY. (2015). A comparative analysis of agricultural research and extension reforms in China and India. China Agric. Econ. Rev.7, 541–572. doi: 10.1108/CAER-05-2015-0054

  • CrossRef
  • Google Scholar

• 3

  BaronR. M.KennyD. A. (1986). The moderator–mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J. Pers. Soc. Psychol.51, 1173–1182. doi: 10.1037/0022-3514.51.6.1173,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4

  BuD.LiaoY. (2022). Land property rights and rural enterprise growth: evidence from land titling reform in China. J. Dev. Econ.157:102853. doi: 10.1016/j.jdeveco.2022.102853

  • CrossRef
  • Google Scholar

• 5

  ChenL.ChenH.ZouC.LiuY. (2021). The impact of farmland transfer on rural households’ income structure in the context of household differentiation: a case study of Heilongjiang Province, China. Land10:362. doi: 10.3390/land10040362

  • CrossRef
  • Google Scholar

• 6

  ChenJ.XuJ.ZhangH. (2022). Explanation of the phenomenon “different prices on the same land” in the farmland transfer market—evidence from China’s farmland transfer market. Agriculture12:2087. doi: 10.3390/agriculture12122087

  • CrossRef
  • Google Scholar

• 7

  ChenJ.YangM.ZhangZ.WangZ.ZhangJ. (2023). Can farmland transfer reduce vulnerability as expected poverty? Evidence from smallholder households in rural China. Front. Sustain. Food Syst.7:1187359. doi: 10.3389/fsufs.2023.1187359

  • CrossRef
  • Google Scholar

• 8

  CuiX.LaiW.LinT. (2025). Long-distance water infrastructure, rural development and urban growth: evidence from China. J. Urban Econ.146:103736. doi: 10.1016/j.jue.2025.103736

  • CrossRef
  • Google Scholar

• 9

  DavisB.de la O CamposA. P.FarraeM.WintersP. (2024). Whither the agricultural productivity-led model? Reconsidering resilient and inclusive rural transformation in the context of agrifood systems. Glob. Food Secur.43:100812. doi: 10.1016/j.gfs.2024.100812

  • CrossRef
  • Google Scholar

• 10

  FangT.ZhouY.WangL.ShiD.DuanX. (2024). The impact of multiplex relationships on households’ informal farmland transfer in rural China: a network perspective. J. Rural. Stud.112:103419. doi: 10.1016/j.jrurstud.2024.103419

  • CrossRef
  • Google Scholar

• 11

  FeiR.LinZ.ChungaJ. (2021). How land transfer affects agricultural land use efficiency: evidence from China’s agricultural sector. Land Use Policy103:105300. doi: 10.1016/j.landusepol.2021.105300

  • CrossRef
  • Google Scholar

• 12

  FengS.HeerinkN.RubenR.QuF. (2010). Land rental market, off-farm employment and agricultural production in Southeast China: a plot-level case study. China Econ. Rev.21, 598–606. doi: 10.1016/j.chieco.2010.06.002

  • CrossRef
  • Google Scholar

• 13

  GeD.LongH.ZhangY.MaL.LiT. (2018). Farmland transition and its influences on grain production in China. Land Use Policy70, 94–105. doi: 10.1016/j.landusepol.2017.10.010

  • CrossRef
  • Google Scholar

• 14

  GebresilasseM. (2023). Rural roads, agricultural extension, and productivity. J. Dev. Econ.162:103048. doi: 10.1016/j.jdeveco.2023.103048

  • CrossRef
  • Google Scholar

• 15

  GorganM.HartvigsenM. (2022). Development of agricultural land markets in countries in Eastern Europe and Central Asia. Land Use Policy120:106257. doi: 10.1016/j.landusepol.2022.106257

  • CrossRef
  • Google Scholar

• 16

  HaoP.TangS. (2023). Migration and consumption among poor rural households in China. Habitat Int.137:102832. doi: 10.1016/j.habitatint.2023.102832

  • CrossRef
  • Google Scholar

• 17

  HeL.HuangJ. (2024). Social capital, government guidance and contract choice in agricultural land transfer. PLoS One19:e0303392. doi: 10.1371/journal.pone.0303392,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 18

  HongW.HuX. (2019). Farmland rights stability and labor non-agriculture migration. Econ. Rev.02, 34–47. doi: 10.19361/j.er.2019.02.03

  • CrossRef
  • Google Scholar

• 19

  HuangJ.JiX. (2012). The verification of the right to use farmland and farmers' long-term investment in farmland. J. Manag. World09, 76–81+99+187-188. doi: 10.19744/j.cnki.11-1235/f.2012.09.007

  • CrossRef
  • Google Scholar

• 20

  JiX.WangY.YangL.LiC.ChenL. (2025). The impact of cropland transfer on rural household income in China: the moderating effects of education. Land Use Policy148:107399. doi: 10.1016/j.landusepol.2024.107399

  • CrossRef
  • Google Scholar

• 21

  JialiW.YanfangL.XiaolingZ. (2021). Conflict in informal rural construction land transfer practices in China: a case of Hubei. Land Use Policy109:105573. doi: 10.1016/j.landusepol.2021.105573

  • CrossRef
  • Google Scholar

• 22

  JiangF.JiangY.PengJ.LvY.WangW.ZhouZ. (2023). Effects of rural collective economy policy on the common prosperity in China: based on the mediating effect of farmland transfer. Front. Environ. Sci.11:1302545. doi: 10.3389/fenvs.2023.1302545

  • CrossRef
  • Google Scholar

• 23

  KanK. (2021). Creating land markets for rural revitalization: land transfer, property rights and gentrification in China. J. Rural. Stud.81, 68–77. doi: 10.1016/j.jrurstud.2020.08.006

  • CrossRef
  • Google Scholar

• 24

  KowalczykC.NowakM.ŹróbekS. (2019). The concept of studying the impact of legal changes on the agricultural real estate market. Land Use Policy86, 229–237. doi: 10.1016/j.landusepol.2019.05.012

  • CrossRef
  • Google Scholar

• 25

  KuangY.-p.YangJ.-l.AbateM.-C. (2022). Farmland transfer and agricultural economic growth nexus in China: agricultural TFP intermediary effect perspective. China Agric. Econ. Rev.14, 184–201. doi: 10.1108/CAER-05-2020-0076

  • CrossRef
  • Google Scholar

• 26

  LiY.DuR.LiL.JiangG.FanZ. (2020). Influences of the transaction intention of farmland transfer under information asymmetry: an empirical study of 1100 questionnaires from China. Sustainability12:3739. doi: 10.3390/su12093739

  • CrossRef
  • Google Scholar

• 27

  LiX.ItoJ. (2021). An empirical study of land rental development in rural Gansu, China: the role of agricultural cooperatives and transaction costs. Land Use Policy109:105621. doi: 10.1016/j.landusepol.2021.105621

  • CrossRef
  • Google Scholar

• 28

  LiL.KhanS. U.GuoC.HuangY.XiaX. (2022). Non-agricultural labor transfer, factor allocation and farmland yield: evidence from the part-time peasants in loess plateau region of Northwest China. Land Use Policy120:106289. doi: 10.1016/j.landusepol.2022.106289

  • CrossRef
  • Google Scholar

• 29

  LiX.LiT.ZhangY. (2017). The calculation of the trading index of the rural property rights trading platform in the central region. Stat. Decis.11, 101–104. doi: 10.13546/j.cnki.tjyjc.2017.11.026

  • CrossRef
  • Google Scholar

• 30

  LiQ.WangX. (2026). Can agricultural cooperatives improve livestock farm profitability? Evidence from the beef cattle sector in China. Appl. Econ. Lett.1-5, 1–5. doi: 10.1080/13504851.2026.2615175,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31

  LiC.XieH.HuangZ.GaiQ. (2025). Property rights, resource reallocation and welfare effects: evidence from a land certification programme. Land Use Policy154:107562. doi: 10.1016/j.landusepol.2025.107562

  • CrossRef
  • Google Scholar

• 32

  LiuJ.LiX.LiuS.RahmanS.SriboonchittaS. (2022). Addressing rural–urban income gap in China through farmers’ education and agricultural productivity growth via mediation and interaction effects. Agriculture12:1920. doi: 10.3390/agriculture12111920

  • CrossRef
  • Google Scholar

• 33

  LiuM.ZhouX.HuangG.LiY. (2024). The increasing water stress projected for China could shift the agriculture and manufacturing industry geographically. Commun. Earth Environ.5:396. doi: 10.1038/s43247-024-01560-y

  • CrossRef
  • Google Scholar

• 34

  LyuH.DongZ.RoobavannanM.KandasamyJ.PandeS. (2019). Rural unemployment pushes migrants to urban areas in Jiangsu Province, China. Palgrave Commun.5:92. doi: 10.1057/s41599-019-0302-1

  • CrossRef
  • Google Scholar

• 35

  MaddalaG. S. (1983). Limited-Dependent and Qualitative Variables in Econometrics. Cambridge: Cambridge University Press.

  • Google Scholar

• 36

  MgendiB. G.MaoS.QiaoF. (2022). Does agricultural training and demonstration matter in technology adoption? The empirical evidence from small rice farmers in Tanzania. Technol. Soc.70:102024. doi: 10.1016/j.techsoc.2022.102024

  • CrossRef
  • Google Scholar

• 37

  NakanoY.TsusakaT. W.AidaT.PedeV. O. (2018). Is farmer-to-farmer extension effective? The impact of training on technology adoption and rice farming productivity in Tanzania. World Dev.105, 336–351. doi: 10.1016/j.worlddev.2017.12.013

  • CrossRef
  • Google Scholar

• 38

  PengW.RobinsonB. E.ZhengH.LiC.WangF.LiR. (2022). The limits of livelihood diversification and sustainable household well-being, evidence from China. Environ. Dev.43:100736. doi: 10.1016/j.envdev.2022.100736

  • CrossRef
  • Google Scholar

• 39

  ShaoJ. a.ZhangS.LiX. (2016). Effectiveness of farmland transfer in alleviating farmland abandonment in mountain regions. J. Geogr. Sci.26, 203–218. doi: 10.1007/s11442-016-1263-6

  • CrossRef
  • Google Scholar

• 40

  ShiC.-l. (2024). Impact of land transfer on high-quality agricultural development: analysis based on the green TFP perspective. J. Nat. Resour.39, 1418–1433. doi: 10.31497/zrzyxb.20240610

  • CrossRef
  • Google Scholar

• 41

  TalviE.VéghC. A. (2005). Tax base variability and procyclical fiscal policy in developing countries. J. Dev. Econ.78, 156–190. doi: 10.1016/j.jdeveco.2004.07.002

  • CrossRef
  • Google Scholar

• 42

  WangY.GuoX. (2011). The transfer of land contract management rights: bilateral transactions and centralized transactions. J. Agrotechnical Econ.10, 46–54. doi: 10.13246/j.cnki.jae.2011.10.006

  • CrossRef
  • Google Scholar

• 43

  WangA.MaJ.HeK. (2026). Citizenization and urban–rural income inequality: evidence from the reform of hukou system in China. World Dev.199:107238. doi: 10.1016/j.worlddev.2025.107238

  • CrossRef
  • Google Scholar

• 44

  WangJ.XinL.WangY. (2020). How farmers’ non-agricultural employment affects rural land circulation in China?J. Geogr. Sci.30, 378–400. doi: 10.1007/s11442-020-1733-8

  • CrossRef
  • Google Scholar

• 45

  WangL.YangR.WuB. (2020). A study on total factor productivity of agricultural production of rural households in China. J. Manag. World36, 77–93. doi: 10.19744/j.cnki.11-1235/f.2020.0185

  • CrossRef
  • Google Scholar

• 46

  WenZ.ChangL.HauK.-T.LiuH. (2004). Testing and application of the mediating effects. Acta Psychol. Sin.36:614. https://journal.psych.ac.cn/acps/EN/Y2004/V36/I05/614

  • Google Scholar

• 47

  WondeK. M.TsehayA. S.LemmaS. E. (2022). Training at farmers training centers and its impact on crop productivity and households’ income in Ethiopia: a propensity score matching (PSM) analysis. Heliyon8:e09837. doi: 10.1016/j.heliyon.2022.e09837,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  WuY.DuanK.ZhangW. (2023). The impact of internet use on farmers’ land transfer under the framework of transaction costs. Land12:1855. doi: 10.3390/land12101855

  • CrossRef
  • Google Scholar

• 49

  WuL.WangS.ZhuD.HuW.WangH. (2015). Chinese consumers’ preferences and willingness to pay for traceable food quality and safety attributes: the case of pork. China Econ. Rev.35, 121–136. doi: 10.1016/j.chieco.2015.07.001

  • CrossRef
  • Google Scholar

• 50

  YangZ. (2024). Measuring transaction costs in China’s rural land rental market: ex-ante bargaining associated with ex-post contract-violation uncertainty and location specificity. J. Econ. Behav. Organ.224, 688–716. doi: 10.1016/j.jebo.2024.06.022

  • CrossRef
  • Google Scholar

• 51

  YangZ.WenF. (2020). From employees to entrepreneurs: rural land rental market and the upgrade of rural labor allocation in non-agricultural sectors. J. Manag. World36, 171–185. doi: 10.19744/j.cnki.11-1235/f.2020.0109

  • CrossRef
  • Google Scholar

• 52

  YuY.SloanF. A. (2017). Trends in elderly health by cohort: evidence from China. China Econ. Rev.44, 282–295. doi: 10.1016/j.chieco.2017.05.001

  • CrossRef
  • Google Scholar

• 53

  ZhangL.DongY.LiuC.BaiY. (2018a). Off-farm employment over the past four decades in rural China. China Agric. Econ. Rev.10, 190–214. doi: 10.1108/CAER-11-2017-0212

  • CrossRef
  • Google Scholar

• 54

  ZhangL.FengS.HeerinkN.QuF.KuyvenhovenA. (2018b). How do land rental markets affect household income? Evidence from rural Jiangsu, P.R. China. Land Use Policy74, 151–165. doi: 10.1016/j.landusepol.2017.09.005

  • CrossRef
  • Google Scholar

• 55

  ZhangJ.FengS.ZhuP. (2017). Does government intervention on farmland rental market Lead to rural income inequality? Journal of. Public Management14104-116+158-159. doi: 10.16149/j.cnki.23-1523.2017.01.009

  • CrossRef
  • Google Scholar

• 56

  ZhangJ.MishraA. K.MaX. (2023). Mechanism of Chinese farmers’ land rental participation: the role of invisible markets and public intervention. Food Policy117:102453. doi: 10.1016/j.foodpol.2023.102453

  • CrossRef
  • Google Scholar

• 57

  ZhangC.NieK.ZhuY. (2024). Participation in land rental and labor markets and agricultural economic performance of banana farmers in China. Land Use Policy146:107307. doi: 10.1016/j.landusepol.2024.107307

  • CrossRef
  • Google Scholar

• 58

  ZhangJ.ZhuP. (2017). The effect of different farmland transfer patterns on household agricultural productivity based on surveys of four counties in Jiangsu Province. Resour. Sci.39, 629–640. doi: 10.18402/resci.2017.04.05

  • CrossRef
  • Google Scholar

• 59

  ZhuS. e.YangR.WangL.WuB. (2024). Rural land transfer and the change of agricultural production mode in China. J. Manag. World40, 76–88+106+189-190. doi: 10.19744/j.cnki.11-1235/f.2024.0011

  • CrossRef
  • Google Scholar