Clan Networks, Spatial Selection, and Farmland Transfer Contracts: Evidence from China

Abstract

Contracts play a crucial role in the reform of land markets and the process of farmland transfer. This study examines how spatial distance and clan networks impact the choice of farmland transfer contracts based on micro-level survey data from farmer households in China. Our research findings offer valuable insights into the role of contracts as a governance tool in land market reform and provide important implications for policymakers and stakeholders. In this study, we reveal that spatial distance significantly influences the selection of farmland transfer contracts. When farmers face long spatial distances, they tend to prefer written contracts to regulate the transfer relationship. This preference helps to mitigate information asymmetry and cooperation risks, ensuring a more secure and efficient transfer process. Additionally, our findings show that clan networks also play a significant role in the choice of farmland transfer contracts. Strong clan networks in high-density areas often have well-defined social norms and codes of conduct. As a result, farmers in these areas are more likely to opt for written contracts, which provide a formalized framework for governing farmland transfers. Furthermore, the density of the clan network acts as a moderator in the relationship between spatial distance and contract choice. A dense clan network intensifies the influence of spatial distance on contract choice, especially when dealing with long spatial distances. This suggests that social networks and community dynamics play a crucial role in shaping farmers’ contract preferences in farmland transfer. In conclusion, this study highlights the importance of contracts as a governance tool in land market reform and provides insights into the influence of spatial distance and clan networks on the choice of farmland transfer contracts. Policymakers and stakeholders involved in land market reforms should consider the findings of this study when designing policies and interventions. By understanding the dynamics surrounding farmland transfer, policymakers can develop more effective strategies to promote secure and efficient land transactions in the context of market-oriented reforms.

1. Introduction

The continuous development of the market economy has highlighted the importance of the market in resource allocation. A crucial aspect of this development is the establishment of a robust land market system [1]. Creating a land market that facilitates the mobility and flexibility of land resources is essential for optimizing resource allocation and improving production efficiency [2]. In this process of land marketization, farmland transfer plays a vital role in promoting optimal resource allocation and efficiency [3]. The choice of contract by farmers within the land transfer process is of great significance as it ensures smooth transactions and balances the interests of all parties involved. Contracts serve as the economic and legal foundation for agricultural land transfers, stipulating the rights and obligations of both parties and regulating transfer behavior [4]. Therefore, it is crucial to study the factors and mechanisms that influence the selection of farmland transfer contracts in order to optimize the functioning of the farmland transfer market and enhance transfer efficiency.

In the context of farmland transfer, two influential factors are the clan network and spatial distance. The clan network, as a social network, represents a form of social capital formed through connections and cooperation among relatives, friends, and fellow villagers [5]. It plays a vital role in information transmission, trust establishment, and risk sharing in farmland transfer contracts [6]. The presence of a strong and dense clan network facilitates the flow of information among its members, which is essential for making informed decisions during the farmland transfer process. Additionally, the network fosters trust among its members, making them more willing to engage in farmland transfer transactions with each other. The risk-sharing aspect of the clan network ensures that if a transfer does not go as planned, the network members collectively bear the consequences, reducing the overall risk for individual farmers.

On the other hand, spatial distance, as a geographic factor, affects various aspects of farmland transfer transactions [7]. Longer spatial distances between farmers and potential transfer objects can lead to higher transaction costs, as more resources are required for transportation, communication, and coordination. These higher costs can influence the choice of contract, with farmers preferring written contracts to ensure a clear and formalized agreement, given the potential uncertainties associated with long-distance transactions. Additionally, spatial distance affects resource utilization efficiency, as closer proximity allows for easier access to and management of farmland, leading to more efficient production processes.

This study aims to explore how the clan network and spatial distance influence the choice of farmland transfer contracts and analyze their effects to deepen our understanding of the operational principles of the farmland transfer market and farmers’ decision-making behavior. By examining the impact of the clan network and spatial distance on contract choice, policymakers can gain valuable insights to optimize the farmland transfer market, promote efficient resource allocation, and foster sustainable development in rural economies.

To conduct the study, field survey data collected from March to April 2023 were utilized to investigate the relationship between clan networks, spatial distance, and the selection of farmland transfer contracts in rural areas. Several innovations were introduced in this research. Firstly, an analysis framework was introduced that incorporated spatial distance, clan networks, and the choice of farmland transfer contracts, enriching the theoretical understanding of contract choice behavior by examining these factors from multiple dimensions. This comprehensive approach allowed for a more nuanced understanding of the factors influencing contract choices among farmers. Secondly, the research took a unique perspective by analyzing the selection of farmland transfer contracts through the lens of clan networks and spatial distance, expanding our knowledge and comprehension of contract choices in the context of farmland transfers. This multi-dimensional analysis provided valuable insights into the interplay between social networks and geographic factors in contract decision-making. Lastly, the use of up-to-date data collected in 2023 ensured timeliness and empirical value, contributing to the study of farmland transfer contract selection. The use of recent data enhanced the reliability and accuracy of the research results, enabling comprehensive theoretical investigations and empirical analyses of the current land market and contract choice behavior.

In conclusion, these innovations contribute to a deeper understanding of the dynamics between clan networks, spatial distance, and the choice of farmland transfer contracts. By shedding light on the interplay between social networks and geographic factors, this study provides valuable information for policymakers aiming to optimize the farmland transfer market, promote efficient resource allocation, and foster sustainable development in rural economies. This study also lays the foundation for more robust theoretical investigations and empirical analyses of the land market and contract choice behavior in the present period.

2. Literature Review and Theoretical Analysis

2.1. Literature Review

Farmland transfer contracts have become a significant subject of study in institutional economics research. Different scholars have emphasized various aspects of contract choices. Marx focused on the consideration of long-term lease contracts versus tenancies at will, highlighting the importance of contract duration [8]. Zhang Wuchang highlighted the role of risk diversification benefits and transaction costs in contract choices, with transaction costs influencing decisions to pursue higher income [9]. Contracts are regarded as a means of addressing shared concerns and solving problems through voluntary and equal communication between parties [10,11]. They help reduce fraud and ensure market operations, but their incompleteness may require adjustments based on real-world conditions [12]. Considering various influencing factors is crucial to ensure the effectiveness and practicality of contracts, especially in the context of farmland transfer.

In the realm of farmland transfer contracts, clan networks and spatial distance emerge as significant influencing factors. Clan networks facilitate communication and cooperation among farmers, enhancing willingness and cooperation in farmland transfer contracts [13]. Empirical research has shown that clan networks impact innovation and entrepreneurship, influencing the competitiveness of farmers’ entrepreneurial enterprises and resource acquisition limitations [14,15]. However, clan networks can also have adverse effects on village-level collective economies by promoting labor outflows and reducing the supply of public goods [16]. Their impact on resource allocation efficiency in land transfers varies depending on the local market conditions [17].

Spatial distance, as a geographic factor, also plays a significant role in the choice of farmland transfer contracts. Studies have shown that spatial distance affects consumer demand, industrial upgrading, and collective decision-making processes [18]. Longer spatial distances may lead to reduced consumer psychological ownership and willingness to share, impacting decision-making [19]. Additionally, spatial distance between parties involved in transfers affects contract choices and participation in public affairs, which can be influenced through various communication channels and social networks [20].

In conclusion, understanding the factors influencing the choice of farmland transfer contracts, such as clan networks and spatial distance, is vital for policymakers and researchers. By considering these factors, policymakers can design more effective land market policies, while researchers can deepen their understanding of the dynamics of farmland transfer and contract choices in rural economies.

2.2. Analytical Framework

Harvey’s perspective emphasizes that the concept of space is derived from human experience, and geographers focus on studying selected phenomena within space rather than studying everything encompassed by it [21]. Loesch highlights the importance of spatial factors in economic research, particularly in the context of farmland transfer [22].

When there is a long spatial distance between parties involved in farmland transfer and asymmetric information exists, identifying the relevant attributes of the other party becomes challenging, leading to higher transaction costs. These attributes directly influence the choice of transfer contracts. The objective and accurate identification of attributes, such as the relationship network of the other party, often depends on the degree of intimacy in transactions or communication, which is closely related to spatial distance. Differences in spatial distance between the residences of the parties result in variations in transaction frequencies or exchanges [23]. Therefore, spatial distance becomes an important explanatory variable in farmers’ contract choices.

From the perspective of spatial distance, a greater spatial distance between the parties in a transfer implies higher transaction costs and more effort needed to address incomplete and asymmetric information [24]. Firstly, communication becomes limited. Longer spatial distances reduce opportunities for face-to-face interaction and communication between farmers and transferees. Language environment, cultural differences, and time–space disparities may further hinder communication, affecting the understanding and negotiation of transfer matters. Secondly, there is an increased risk of default. When significant interests and risks are involved, monitoring and controlling contract performance become necessary. A written contract can provide a clear binding and monitoring mechanism, allowing for tracking and recording contract performance, facilitating the resolution of potential disputes in the future. Conversely, when the spatial distance between the parties is relatively close, transaction costs and the effort required to address incomplete and asymmetric information are reduced [25].

In rural “acquaintance societies”, where interpersonal relationships are typically close and neighbors often share long-standing kinship and trust foundations, farmers may be more inclined to choose verbal contracts for land transfers because they believe the other party will honor their commitments. Moreover, oral contracts are simpler and more flexible than written contracts. Farmers can directly communicate and negotiate face-to-face with nearby transfer objects, avoiding the cumbersome preparation and signing of contract documents and the associated costs [26].

As shown in Figure 1, from the perspective of “social spatial distance”, the clan network, as a close social network, facilitates information transmission and sharing. In the process of farmland transfer, farmers can acquire information about the land being transferred through the clan network, including details about land quality, the reliability of transfer objects, and fair prices [27]. This information helps farmers make more informed decisions [28].

Additionally, the clan network serves as a channel for information transmission and communication Illustrated in Figure 1, enabling farmers to gather information about land transfers and assess the credibility and reliability of transfer objects. Moreover, the clan network provides a mechanism for risk sharing in agricultural land transfers. Through the clan network, farmers receive support and assistance from relatives, sharing risks throughout the transfer process. In addition to tangible human, financial, and institutional resources, the clan network represents an influential cultural force [29].

Cultural values not only directly influence individual actions but also shape actors’ skills, styles, habits, and abilities. Thus, clan network variables play a significant role in the choice of farmland transfer contracts. This paper distinguishes between clan network strength and clan network density variables. High density in the clan network indicates closer and more efficient connections among members, leading to faster and more effective information transmission. During the process of farmland transfers, farmers can obtain valuable information about potential transfer objects through the clan network, including details about transaction history [30] and reputation [31]. This information exchange within the clan network plays a crucial role in facilitating the decision-making process for farmers, enabling them to make informed choices regarding their land transfer activities [30,31]. This information sharing and transmission deepen their understanding of the transfer objects, instilling greater confidence and dependence. High clan network strength implies closer and more stable relationships among members. Within farmland transfers, clan members have established long-term mutual trust and cooperative relations. The presence of such relationships enhances the willingness of both parties to cooperate and the reliability of contract execution [32].

3. Materials and Methods

3.1. Sample Selection

The data in this article were collected through a field survey conducted by the research team in four counties of Guizhou Province during March and April 2023. To ensure the representation of the research areas, we carefully selected counties based on criteria such as the level of economic development, geographical distribution characteristics, and endowment characteristics of agricultural land resources.

The selected research areas include Meitan County, which is one of the first batches of experimental areas for the reform of the agricultural property rights system determined by the state after the reform and opening up. Guanling County, with Dingyun Community, known as “the first village of China’s rural land reform”, is also among the selected areas. Sansui County was chosen for its contract documents collected from Miao Village and Dong Village in the river basin, which hold significant research value for re-establishing contract awareness and promoting the spirit of contract. Additionally, Pan County, the first in Guizhou Province to carry out the “three changes” reform in rural areas, was included in the study.

A total of 1250 questionnaires were distributed to the targeted participants, and after excluding farmers who had not experienced land transfer, 1101 valid questionnaires remained, resulting in an effective response rate of 88.80%. The high response rate indicates a good level of participation and engagement from the respondents, enhancing the reliability and representativeness of the data collected for the research. With a large sample size, the research findings can provide valuable insights into the factors influencing the choice of farmland transfer contracts in rural areas.

3.2. Model Selection

The basic model of this paper is set as follows:

$$Contract={\alpha }_0+{\alpha }_{1}Dis\tan ce+{\alpha }_{2}X+{\epsilon }_i$$

(1)

$$Contract={\beta }_0+{\beta }_{1}Clan+{\beta }_{2}X+{\epsilon }_i$$

(2)

$$Contract={\lambda }_0+{\lambda }_{1}Dis\tan ce+{\lambda }_{2}Clan+{\lambda }_{2}X+{\epsilon }_i$$

(3)

Among them, Contract represents the choice of contract for farmland transfer, Distance represents the spatial distance, Clan represents the clan network, and X represents the control variables, namely individual characteristics of farmers, family endowment characteristics, land cognitive characteristics and village environment characteristics. Model (1) examines the influence of spatial distance on the choice of farmland transfer contract, model (2) examines the influence of clan network on the choice of farmland transfer contract, and model (3) examines the relationship between clan network and the choice of farmland transfer contract in spatial distance and the regulating effect. ${\alpha }_0~{\alpha }_2$, ${\beta }_0~{\beta }_2$, ${\gamma }_0~{\gamma }_2$, ${\lambda }_0~{\lambda }_2$ represent the sample regression coefficient, ${\epsilon }_i$ is a random perturbation term.

3.3. Variable Selection

The incorporation of spatial distance and clan networks as key factors in analyzing the choice of farmland transfer contracts is a significant contribution of this research. By examining the impact of different spatial distances on contract selection, this study aims to shed light on the influence of geographic factors on farmers’ decision-making behavior. Additionally, by investigating the role of clan networks, the research seeks to explore the social capital aspect of contract choices and its implications for efficient resource allocation in farmland transfers.

To ensure a comprehensive analysis, this study considers various other factors that may affect contract choice. Individual farmers’ characteristics, family endowment, land cognition, and village environment are considered in the analytical model. By including these variables, the research aims to capture the multifaceted nature of contract choices and provide a more holistic understanding of the factors influencing farmers’ decision-making in the context of farmland transfer.

The integration of these different factors into the analytical model allows for a more robust examination of the determinants of contract selection. Further, it provides a comprehensive framework that considers both individual and contextual factors, enabling a deeper understanding of the complex dynamics involved in farmland transfer decisions. Through this approach, our research contributes valuable insights to the existing literature on farmland transfer contract choices and provides a solid foundation for policymakers and researchers to optimize the functioning of the farmland transfer market and promote sustainable development in rural economies.

3.3.1. The Variable to Be Explained

The choice of farmland transfer contract serves as the dependent variable in this research. It is assessed based on whether a written contract or an oral contract was signed between the transfer parties. The categorization follows the approach taken by previous studies, such as [18,31,33], and other scholars in the field. In this study, a value of “0” represents an oral contract, while a value of “1” represents a written contract.

The average value of the contract choice variable is 0.530, indicating that, on average, a slightly higher proportion of farmers have opted for written contracts in the context of farmland transfers. This finding suggests that there has been an increase in the prevalence of written contracts over time, likely driven by the government’s regulation of the land transfer market and the implementation of land transfer policies. These policies may have prompted farmers to adopt written contracts as a means to ensure legal protection and clarity in their transfer agreements. The prevalence of written contracts may also reflect a growing recognition among farmers of the importance of formalizing their transfer agreements to reduce transaction risks and enhance the security of their land transactions.

3.3.2. Main Explanatory Variables

The main independent variables in this paper are the clan network and spatial distance. Drawing on the research of Jiang, X., Ma, X., et al. (2022) and Wang, A., He, K. et al. (2022), the clan network is divided into two components: clan network intensity and clan network density [16,17].

To measure clan network intensity, the presence of ancestral halls in the village is used as an indicator [34]. Ancestral halls serve as gathering places and centers of activities for clan members, reflecting the strength and cohesion of the clan network. The existence of ancestral halls suggests a higher intensity of the clan network.

Clan network density is measured using the proportion of the population with the largest surname in the village [35]. A higher proportion indicates stronger blood relations and connections among clan members, representing a denser clan network.

As for spatial distance, it is primarily determined by the distance between the farmer and the residence of the transfer object, with reference to the literature of Hong Mingyong (2017) and Hong Mingyong (2018) and the research of various scholars [36,37]. The assessment of spatial distance is based on the responses to questions such as “Where is the home of your land transfer object?” with multiple options provided, including original natural village group, new natural village group, groups outside the village, villages outside the township (town), townships (towns) outside the county (city), and counties (cities) outside. Additionally, the question “What is the distance between your home and the target’s home (in kilometers)?” is used to gather information on the spatial distance.

In this study, the distinction between “groups outside the village” serves as the cutoff point for spatial distance. The distance between farmers and transfer objects is categorized into “0” for near spatial distance and “1” for far spatial distance, representing the level of proximity between the two parties [38]. This categorization allows for a clearer understanding of how spatial distance influences contract choices.

3.3.3. Control Variable

To ensure the accuracy and reliability of the regression results, this study incorporates several control variables that may influence the choice of farmland transfer contracts [39]. The inclusion of these control variables is based on the findings of Hong Mingyong (2017) and Hong Mingyong (2018). The control variables encompass various aspects, including the individual characteristics of farmers, family attribute characteristics, land cognitive characteristics, and environmental characteristics of villages.

The individual characteristics of rural households are considered, and variables such as the head of household’s level of education (Education), health status (Health), and age (Age) are included. These variables capture the individual attributes that may impact contract choices. For instance, higher levels of education may influence farmers’ comprehension of contract terms and their ability to negotiate written contracts.

Family attribute characteristics comprise variables such as whether there are village cadres in the family (Village cadres), the number of migrant workers in the family (Number of migrant workers), and the number of women in the family (Number of women). These variables reflect the family composition and dynamics that could influence contract choices. The presence of village cadres in the family may provide access to information and resources that could influence contract decisions.

Environmental characteristics of the village are represented by variables such as the distance from the village to the nearest expressway entrance (Expressway), the distance from the village committee (Village committee), and the presence of tractor tracks (Tractor track). These variables capture the environmental factors that may affect contract choices. For example, proximity to transportation infrastructure and village administrative centers may influence farmers’ access to information and resources relevant to contract negotiations.

Land cognitive characteristics include variables related to farmers’ perceptions, such as their understanding of land ownership (Land belongs), their perception of land security (Land security), and their awareness of the certificate of title confirmation (Certificate). These variables capture farmers’ cognitive factors that may influence their contract choices. Farmers with a stronger sense of land ownership and perceived land security may be more inclined to choose written contracts for greater protection of their rights.

By incorporating these control variables in the analysis, this study aims to provide a comprehensive examination of the factors influencing the choice of farmland transfer contracts, considering a wide range of individual, family, cognitive, and environmental factors that may shape farmers’ decision-making behavior in the context of farmland transfers.

Table 1. Descriptive statistics of the whole sample.

Variable Name	Variable Definitions	Average	Std.	Min	Max
Explanatory variable
Contract Selection	Divided into written contracts (=1) and oral contracts (=0)	0.5304	0.0150	0.000	1.000
Clan network strength	Refer to whether there is an ancestral hall in the village	0.1035	0.0092	0.000	1.000
Clan network density	Proportion of the largest surname in the village	63.5328	0.5231	7	91.000
Spatial distance	The distance between the places of residence of both parties in circulation (Far space distance = 1, near space distance = 0)	0.9272	0.6282	0.000	1.000
Individual characteristics of farmers
Health	Judge the health status of the head of household (1 is very unhealthy, 3 is average, 5 is very healthy)	3.7639	0.0249	1.000	5.000
Age	Age of head of household	58.0073	0.3709	12.000	93.000
Education	Years of education of the head of household	5.5786	0.1135	0.000	17.000
Family attribute characteristics
Village cadres	Describe the village cadre family	0.1490	0.0107	0.000	1.000
Migrant workers	Number of migrant workers (person)	1.0845	0.0359	0.000	5.000
Number of women	Number of women (person)	2.0173	0.0340	0.000	8.000
Land cognitive characteristics
Land belongs	Judging farmers’ cognition of land ownership	2.2807	0.0257	1.000	3.000
Land security	Judging farmers’ cognition of the social security function of land	0.7920	0.0122	0.000	1.000
Certificate	Determine whether the farmer has the title confirmation certificate	0.5740	0.0149	0.000	1.000
Environmental characteristics of villages
Expressway	Distance from highway intersection	4.8787	0.1363	0.1000	53.000
Village committee	Distance from the village committee	1.3910	0.0322	0.1000	10.000
Tractor track	Judging whether organic farming	0.6312	0.0145	0.000	1.000

presents the processed results, assignments, and descriptive statistics of each variable, providing further details and information for analysis.

4. Results

4.1. Benchmark Model

To begin, Model I was constructed, incorporating variables representing clan network strength and clan network density, aiming to examine their influence on the selection of farmland transfer contracts. Following this, Model II was developed to investigate the impact of spatial distance on farmland transfer contract choices. Subsequently, Model III simultaneously integrated clan network strength, clan network density, and spatial distance to explore their combined effects on farmland transfer contract choices. Additionally, Models IV, V, VI, and VII successively introduced control variables related to household head characteristics, family endowments, land attributes, and village environmental factors to study the collective impact of clan networks and spatial distance on farmland transfer contract choices as these control variables were introduced.

Based on the findings presented in

Table 2. Benchmark regression results.

Variable Name	I	II	III	IV	V	VI	VII
Clan network strength	0.7213 ***		0.8299 ***	0.9064 ***	0.8756 ***	0.9090 ***	0.9660 ***
	(0.2194)		(0.2480)	(0.2512)	(0.2556)	(0.2600)	(0.2644)
Clan network density	0.3714 ***		0.5896 ***	0.5863 ***	0.6027 ***	0.6517 ***	0.6664 ***
	(0.0907)		(0.1070)	(0.1081)	(0.1097)	(0.1187)	(0.1235)
Spatial distance		0.6978 ***	0.7374 ***	0.7377 ***	0.7443 ***	0.7290 ***	0.7443 ***
		(0.0549)	(0.0561)	(0.0564)	(0.0567)	(0.0571)	(0.0585)
Health				−0.1785 **	−0.1504 *	−0.1565 *	−0.1872 **
				(0.0889)	(0.0901)	(0.0908)	(0.0925)
Age				−0.0157 **	−0.0162 **	−0.0159 **	−0.0151 **
				(0.0064)	(0.0065)	(0.0066)	(0.0066)
Education				−0.0436 **	−0.0373 *	−0.0369 *	−0.0245
				(0.0206)	(0.0209)	(0.0210)	(0.0216)
Village cadres					−0.4124 **	−0.4216 **	−0.4363 **
					(0.2057)	(0.2067)	(0.2079)
Migrant workers					−0.1719 ***	−0.1564 ***	−0.1660 ***
					(0.0591)	(0.0593)	(0.0600)
Number of women					−0.0738	−0.0793	−0.0804
					(0.0625)	(0.0630)	(0.0632)
Land belongs						−0.0154	−0.0299
						(0.0918)	(0.0933)
Land security						−0.2658	−0.2667
						(0.1742)	(0.1770)
Certificate						0.3867 ***	0.3595**
						(0.1493)	(0.1507)
Expressway							−0.0119
							(0.0170)
Village committee							−0.0617
							(0.0679)
Tractor track							0.4964 ***
							(0.1551)
Constant	−1.2582 ***	−0.4228 ***	−2.6183 ***	−0.7866	−0.5662	−0.7111	−0.8841
	(0.3233)	(0.0760)	(0.3924)	(0.6893)	(0.7099)	(0.8247)	(0.8519)
N	1101	1101	1101	1101	1101	1101	1101
LR chi2 (2)	34.21	254.03	305.31	316.40	330.29	339.44	351.31
Prob > chi2	0.000	0.000	0.000	0.000	0.000	0.000	0.000
Pseudo R2	0.0225	0.1669	0.2006	0.2079	0.2170	0.2230	0.2308

Note: ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively; the robust standard errors are in brackets.

of Model I, Model II, and Model III, several conclusions can be drawn.

In Model I, the variables representing clan network strength and density show a significant positive relationship with the choice of written contracts. This suggests that the clan network, as an informal governing system, operates independently from formal institutions and plays a crucial role in contract selection. The strength and density of the clan network provide social support, trust, and information flow among clan members, which fosters cooperation and trade. Opting for a written contract enhances transparency, predictability, trust, and stability in cooperation. Additionally, the clan network facilitates the assessment of risks and selection of partners through credit and background information obtained within the network. The presence of the clan network ensures social oversight and constraints, promoting transactional responsibility and commitment.

In Model II, the spatial distance variable is found to be significant. Longer spatial distances increase transaction uncertainty, information asymmetry, and lack of trust, which lead to a higher likelihood of choosing written contracts. Written contracts help to clearly define rights and responsibilities, reducing transaction risks in long-distance transactions. On the other hand, shorter spatial distances facilitate direct interaction, communication, and the establishment of trust-based relationships. In familiar village environments, face-to-face communication helps reduce information asymmetry and enhances cooperation, making oral contracts more likely.

In Model III, when clan network strength, clan network density, and spatial distance are simultaneously included, all variables pass the significance test. This indicates that greater clan network strength and density increase the likelihood of farmers choosing written contracts. The presence of a strong and dense clan network positively influences the choice of farmland transfer contracts, emphasizing the role of clan networks in village governance. Furthermore, even with the inclusion of control variables, the impact of clan network and spatial distance on contract choices remains significant, indicating their robust influence.

Overall, the inclusion of clan network and spatial distance variables, along with the control variables, improves the explanatory power of the model. These findings suggest that both clan networks and spatial distance are important factors in determining the choice of farmland transfer contracts, highlighting the significance of social networks and transactional risks in contract decision-making.

The control variables included in the analysis provide additional insights into the factors influencing the choice of farmland transfer contracts. A summary of the control variables and their potential impact is given as follows:

Health status: Farmers in good health may have higher self-confidence and communication skills, making them more likely to engage in verbal communication and negotiations. This could increase the likelihood of choosing oral contracts.

Age: Older individuals tend to have more experience and wisdom, which can contribute to stronger verbal expression and negotiation skills. Therefore, older farmers may be more inclined to choose oral contracts.

Education level: Farmers with lower levels of education may have limited understanding and application of written texts, making them rely more on written contracts to clarify rights and obligations.

Village cadres: Non-village cadres, such as peasant households, may have limited knowledge and experience in law and contracts. This lack of expertise could lead them to prefer written contracts for greater clarity and legal protection.

Number of migrant workers: Farm households with a larger number of migrant workers may find it easier to communicate and confirm agreements through modern means such as phone calls or text messages. This convenience may reduce their reliance on written contracts.

Certificate of title confirmation: Farmers with a certificate of title confirmation have legal protection and recognition of their land rights and interests. This legal security may enhance their negotiation and transaction capabilities, potentially influencing their contract choice.

Environmental characteristics: The presence of a mechanical farm road in the village signifies established norms and requirements for agricultural production and land management. Farmers in such villages may be more likely to choose written contracts to comply with these standards and requirements.

By including these control variables, the analysis accounts for additional factors that could influence contract choice, providing a more comprehensive understanding of the dynamics involved in farmland transfer decisions.

4.2. Endogeneity Test

Spatial distance is typically considered an exogenous variable, meaning its value is not influenced by other intrinsic factors. The spatial distance between farm households and farmland exists objectively and is unrelated to other factors, eliminating the need for selecting instrumental variables in the analysis.

However, the clan network, as an endogenous variable, may suffer from endogeneity problems. To address this issue and ensure the accuracy and consistency of the estimated results, instrumental variables are used to replace the clan network.

In this study, the “ancestral hall” and “proportion of the largest surname” are employed to measure the clan network density and clan network strength, respectively. These variables effectively avoid endogeneity problems arising from self-selection and simultaneity. Nevertheless, there may still be endogeneity concerns due to measurement errors and omitted variables.

To correct for estimation bias caused by potential endogeneity, we adopt instrumental variables referenced from previous research [40]. Specifically, the “proportion of villages with ancestral temples at the township level to all township villages (VI1)” and “the proportion of villages with the first surname at the township level to all township villages (VI2)” are used as instrumental variables for clan network strength and clan network size. These variables are not directly related to the choice of farmland transfer contracts, but they have a certain association with the clan network. By employing these instrumental variables, this study replaces the impact of clan networks on the choice of farmland transfer contracts and addresses potential estimation bias arising from endogeneity concerns.

By incorporating instrumental variables, we aim to strengthen the validity and robustness of our findings, ensuring that the results accurately reflect the relationships between the variables of interest.

The results presented in

Table 3. Endogeneity test.

Variables	The First Stage	The Second Stage	The First Stage	The Second Stage
	Clan Network Strength		Clan Network Density
Clan network strength		0.260 **
		(0.123)
Clan network density				0.002 **
				(0.001)
IV1	1.003 ***
	(0.070)
IV2			0.002 **
			(0.001)
Control variables	Control	Control	Control	Control
Cragg–Donald Wald F	205.755		153.362
Adj R-squared	(0.157)		(0.122)
Constant	0.401 *	0.505 ***	0.440 ***	0.439 ***
	(0.011)	(0.020)	(0.042)	(0.042)
N	1101	1101	1101	1101

Note: ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively; the robust standard errors are in brackets; the control variables are the same as those in Table 2.

demonstrate that the selection of instrumental variables has successfully passed the validity test. Firstly, the p-values of the underidentification test statistics are all below 0.01, leading to the rejection of the null hypothesis at the 1% significance level. This finding indicates that there is no underidentification problem in the instrumental variables chosen. Secondly, the statistics of the weak instrumental variable test are all greater than the critical value of 10%, resulting in the rejection of the null hypothesis of “there is a weak instrumental variable” at the 5% significance level. These results signify that the instrumental variables used in the analysis are sufficiently strong and robust to address the endogeneity issue.

The estimation results reveal that after appropriately controlling for the endogeneity problem, the coefficients of clan network size and clan network strength remain significantly positive. This outcome suggests that the presence of a clan network significantly influences farmers’ choices in farmland transfer contracts, which is consistent with the findings of the baseline regression results.

By employing instrumental variables and passing the validity tests, the research findings strengthen the credibility and reliability of our conclusions regarding the impact of clan networks on farmers’ decision-making behavior in farmland transfer contract choices. The incorporation of instrumental variables provides a more accurate understanding of the relationships between clan networks and contract choices, contributing to the robustness of our findings.

4.3. Further Analysis

The clan network may play a regulatory role in the relationship between spatial distance and farmland transfer contract choice, and the following equation is constructed:

$$Y=aX+bY+cXM+\epsilon$$

(4)

In the analysis, the dependent variable (Y) represents the choice of farmland transfer contract, specifically distinguishing between written contracts and oral contracts. The independent variable (X) is spatial distance, categorized into long spatial distance and short spatial distance. The moderator variable (M) is the clan network, which plays a moderating role in the relationship between spatial distance and farmland transfer contract choice. To explore the potential interactions, the cross term (XM) of spatial distance and clan network is introduced into the model.

The model development process is as follows: Firstly, the control variables are included in the model to account for other factors that may influence the contract choice. Secondly, the independent variable (spatial distance) and the moderator variable (clan network) are introduced into the model to assess their individual effects. Lastly, the interaction term (spatial distance × clan network) is added to investigate the combined effects of spatial distance and the clan network on farmland transfer contract choices. The results of this analysis are presented in

Table 4. Further analysis results.

Variable	VI	VII
Spatial distance	0.9660 ***	1.3945 ***
	(0.2644)	(0.3142)
Clan network strength	0.6664 ***	0.7674 ***
	(0.1235)	(0.2664)
Clan network density	0.7443 ***	0.7168 ***
	(0.0585)	(0.1241)
Spatial distance × Clan network density		−0.1912 **
		(0.0866)
Spatial distance × Clan network strength		0.2070
		(0.2389)
Control variable	Control	Control
Constant	−0.8841	−3.0828 ***
	(0.8519)	(0.4602)
N	1101	1101
LR chi2 (2)	(351.31)	(310.85)
Prob > chi2	0.000	0.0000
Pseudo R2	0.2308	0.2042

Note: ***, ** indicate significance at the 1%, 5% levels, respectively; the robust standard errors are in brackets; the control variables are the same as those in Table 2.

.

The regression results provide valuable insights into the factors influencing the choice of farmland transfer contracts. The key findings are summarized as follows:

Spatial distance: The coefficient for spatial distance is positive and statistically significant, indicating that greater spatial distance between farmers and transfer objects leads to a higher likelihood of choosing written contracts. This suggests that as the distance between parties involved in the transfer increases, farmers tend to opt for written contracts to mitigate uncertainties and clarify their rights and obligations.

Clan network strength: The interaction term “Spatial distance × Clan network strength” does not show a significant effect on contract choice. This suggests that the strength of the clan network does not significantly influence the relationship between spatial distance and contract choice. Other factors related to clan network density seem to have a more significant impact.

Clan network density: The interaction term “Spatial distance × Clan network density” has a significant effect on contract choice. Clan network density acts as a moderating variable, influencing the relationship between spatial distance and contract choice. When the clan network density is high, farmers are more likely to choose oral contracts, and the impact of spatial distance on contract choice is reduced. A dense clan network facilitates faster and more reliable information transmission, increases trust among members, and provides access to reliable information about transfer objects. As a result, the influence of spatial distance on contract choice is mitigated.

Low clan network density: When clan network density is low, the effect of spatial distance on contract choice may be more pronounced. Greater spatial distances can lead to increased information asymmetry and communication costs, making oral contracts more challenging. In such situations, farmers are more likely to select written contracts to compensate for the lack of information and reduce communication barriers. Therefore, a decrease in clan network density may intensify the impact of spatial distance on contract choice.

In conclusion, both spatial distance and clan network density significantly affect the choice of farmland transfer contracts. The moderating role of clan network density in the relationship between spatial distance and contract choice highlights the importance of social networks in shaping farmers’ decisions. These findings contribute to a deeper understanding of the complexities involved in contract choices and the influence of social networks in farmland transfers.

4.4. Robustness Test

4.4.1. Replacement of Explanatory Variables

In this study, the explanatory variables are replaced with the distance variable (DLOR) between the transfer land and the residence of the transfer object, the genealogy variable (Genealogy), and the proportion of the own surname (Own surname), and the estimations are conducted. This method adjusts the selection of explanatory variables and tests the robustness of the research conclusions. The results presented in Table 4 are consistent with the expectations based on the previous baseline regression results.

4.4.2. Replacement of the Estimation Model

To verify the generalizability of the benchmark regressions, robustness testing using probit models is employed, which estimates the probability of binary categories such as written and oral contracts. The results, as shown in

Table 5. Robustness check.

Variable	(1) Replace Variables	(2) Replace the Metering Model
DLOR	0.6325 **
	(0.3383)
Genealogy	0.1722 ***
	(0.1379)
Own surname	0.1413 ***
	(0.0450)
Spatial distance		0.5871 ***
		(0.1558)
Clan network strength		0.3795 ***
		(0.0719)
Clan network density		0.4221 ***
		(0.0308)
Control variable	Y	Y
Constant	−0.8231 ***	−0.3985
	(0.1298)	(0.5017)
N	1101	1101
LR chi2 (2)	203.13	345.90
Prob > chi2	0.0000	0.0000
Pseudo R2	0.1334	0.2272

Note: ***, ** indicate significance at the 1%, 5% levels, respectively; the robust standard errors are in brackets; the control variables are the same as those in Table 2.

, indicate that as the spatial distance increases and the strength of the clan network increases, farmers are more inclined to choose written contracts. This finding can be attributed to the greater need for written contracts to ensure the reliability and stability of the cooperative relationship and the effective role of the clan network in informal governance. This further confirms the conclusions drawn from the previous baseline regression analysis.

5. Discussion

5.1. Key Findings

Clan networks, essential social structures formed through connections and cooperation among relatives, friends, and fellow villagers, play a crucial role in information transmission, trust establishment, and risk sharing in farmland transfer contracts [40,41,42,43]. Our research findings underscore the positive influence of clan networks on the choice of farmland transfer contracts. In traditional rural societies, clan networks serve as pivotal channels for transmitting vital information about transfer objects, such as land quality, transaction history, and reputation, enabling farmers to make well-informed decisions. These results align with the works of Hong Mingyong, Yang Xuejiao (2021), and Niu Kunzai, Xu Hengzhou et al. (2022), emphasizing the significant role of clan networks in farmland transfer.

Furthermore, this study reveals that spatial distance significantly impacts the choice of farmland transfer contracts. As a geographic factor, spatial distance affects transaction costs, resource utilization efficiency, and information flow [44]. When facing long spatial distances, farmers tend to opt for written contracts, seeking to mitigate information asymmetry and cooperation risks inherent in such transactions. Written contracts offer clearer protection and regulation of rights and interests, enhancing transaction reliability and stability. These findings are consistent with research conducted by Hong Mingyong (2017) and Hong Mingyong (2018). It is important to note that the context of farmland circulation varies, leading to differences in transaction frequency, information flow for evaluating performance, and consequently, contract choices [35,37].

Additionally, this study identifies the moderating role of clan network density between spatial distance and farmland transfer contract choices. Higher clan network density can mitigate the impact of spatial distance on contract choices. Greater network density facilitates faster and more reliable information exchange, fostering higher levels of trust among members. As a result, farmers find it easier to access reliable information about transfer objects and establish strong trust-based relationships. Thus, increasing clan network density can alleviate the influence of spatial distance on contract choices. These findings align with research conducted by Hong Mingyong and Yang Xuejiao et al. (2021), indicating that clan networks play a regulatory role in the allocation of land transfer resources [39].

5.2. Limitations and Future Prospects

Although this study has conducted an in-depth discussion on the relationship between clan networks, spatial distance, and farmland transfer contract choices, there are still some research deficiencies and directions worthy of further exploration.

On the one hand, this study solely considered the influence of clan networks and spatial distance on the choice of farmland transfer contracts, without incorporating other factors that may also impact farmland transfer, such as policy, economic, and cultural factors. Future research should incorporate these additional factors into the analysis and establish a more comprehensive analytical framework to gain a deeper understanding of the operating mechanisms in the farmland transfer market.

On the other hand, this study utilized cross-sectional data for analysis, which limited its ability to capture changes over time. To address this limitation, future research could employ panel data to compare information at different time points, thereby exploring the dynamic relationships among clan networks, spatial distance, and farmland transfer contract choices. This temporal perspective would provide valuable insights into the evolving nature of these relationships and their implications for the farmland transfer process.

6. Conclusions

The conclusions of this study emphasize the significant impact of spatial distance and clan network characteristics on the choice of farmland transfer contracts, providing crucial insights for policymakers and stakeholders in the farmland transfer market.

Firstly, this study emphasizes the importance of establishing standardized and legally sound practices for farmland transfers. Considering the impact of spatial distance on contract choice, it becomes imperative for governments to formulate regulations and policies that ensure the legality and protection of rights and interests in these transactions. By creating a transparent and secure environment, such measures will foster confidence and trust among participants, thereby promoting the healthy development of the market.

Secondly, this study highlights the crucial role of strengthening clan networks and information transmission mechanisms. Recognizing the significance of clan network density and strength in contract selection, efforts should be directed towards supporting and enhancing the cohesion and connectivity of these networks. This can be accomplished by providing information resources and assistance, encouraging connections and collaboration among farmers, and reducing information asymmetry and cooperation risks. Strengthening clan networks will facilitate more reliable information exchange and foster smoother transactions within the farmland transfer market.

Lastly, this study proposes the promotion of diversified forms of farmland transfer contracts. Recognizing the varying needs and circumstances of participants, policymakers can facilitate the flexibility of both oral and written contracts. Through the provision of legal support and tailored guidance to different regions and individuals, participants can opt for the contract format that best aligns with their specific preferences and situations. This approach fosters adaptability and customization in contract selection, ultimately enhancing the effectiveness and efficiency of farmland transfers. By offering a range of contract options, the farmland transfer market can better accommodate the diverse needs of stakeholders and contribute to a more dynamic and inclusive agricultural sector.

In conclusion, implementing the suggested measures can enhance the efficiency, fairness, and overall performance of farmland transfer transactions. This will contribute to the development of a robust and sustainable farmland transfer market, benefiting all involved parties and fostering long-term agricultural growth and stability.