Study on the Impact of Social Capital on Agricultural Land Transfer Decision: Based on 1017 Questionnaires in Hubei Province

Abstract

In building a market for the transfer of land contracting rights, it is crucial to clarify the influencing factors for farmers’ farmland transfer decisions to promote the orderly transfer of farmland. This article investigates the impact of social capital on farmland transfer and analyzes the moderating effect of marketization of farmland transfer using research data from 1017 farm households in Hubei Province. The results showed that social capital significantly contributes to farmers’ farmland transfer decisions; social networks and social trust have more potent effects on farmland transfer-in than farmland transfer-out and social norms have more substantial effects on farmland transfer-out than farmland transfer-in; and farmland transfer marketization plays a moderating role in the process of social capital’s influence on farmland transfer decisions. In terms of farmland transfer-out, marketization of farmland transfer plays a negative moderating role between social network, social trust, and farmland transfer decision, and a positive moderating role between social norm and farmland transfer decision. In terms of farmland transfer-in, marketization of farmland transfer plays a negative moderating role between social network, social trust, and farmland transfer decision, and no moderating role in social norm and farmland transfer decision. The study concludes that local governments need to strengthen the construction of social norms and the reconstruction of social trust and networks to create a favorable environment for agricultural land transfer while improving the construction of regional agricultural land transfer markets and promoting the development of market transformation.

1. Introduction

Agricultural land transfer has an essential role in promoting moderate-scale agricultural operations [1], improving agricultural production efficiency [2], increasing farmers’ income [3], and promoting the transfer of rural labor [4,5]. From the policy level, the Chinese government has recently implemented several supportive measures to speed up the process of agricultural land transfer, dramatically increasing farmers’ enthusiasm for agricultural land transfer. In 2021, China’s agricultural land transfer area reached 37.1 million hectares according to pertinent information from the Ministry of Agriculture and Rural Affairs, an increase of 16.3% over the proportion of transfer at the end of 2016, and the agricultural land transfer scale increased. However, according to the available research, the personification of agricultural land transfer transactions holds a prominent position in the agricultural land transfer market. The fair and standardized agricultural land transfer market system is imperfect [6]. For example, Zhu et al. analyzed the current situation of farmland transfer in 1015 households in Chongqing. They found that zero rent and no agreement for farmland transfer were common in rural areas, and farmland transfer generally occurred among farmers [7]. Although the incidence of farmland transfer in China has increased, farmland transfer is characterized as small scale, short term, and informal. It is at the primary development stage, and further exploration of the constraints of the development bottleneck of farmland transfer is needed to improve the efficiency of farmland transfer.

As an essential practical activity in rural China, how to effectively promote agricultural land transfer has been the focus of scholars’ attention and a hot issue. Most scholars have conducted rich research on the drivers of agricultural land transfer, and studies have shown that factors such as property rights incentives [8], government intervention [9], transaction costs [10], individual characteristics [11], and household endowment [12] all influence farm households’ agricultural land transfer decisions. In addition, agricultural land transfer occurs in rural societies and is influenced by social factors. Especially in rural China, social capital based on blood–geographical kinship constitutes an intangible resource that can influence farmers’ behavioral decisions [13]. When the formal system construction is imperfect, the informal system formed by social capital is crucial for farmers to realize resource allocation [14]. Research practices based on Chinese scenarios also confirm the social capital resource allocation function. For example, Zhang and Li [15] found that social capital can generate nepotism, help relatives and friends obtain jobs, and provide employment opportunities. Subsequently, in the farmland transfer decision, the social capital formed within rural areas may influence farmers’ land resource allocation decisions and prompt farmers to prefer relatives and friends as the transfer objects, so the current socioeconomic behavior of automatic internal transfer of farmland can be well explained from the perspective of social capital. However, existing relevant studies generally focus on objective influencing factors such as economy, individual, and family. The influence effect between social capital and farmland transfer decisions needs to be sufficiently studied. Putnam divided social capital into three dimensions—social networks, social trust, and social norms—and this division is shared and followed by mainstream research. Although a few scholars have focused on the influence of social capital on agricultural land transfer, the research content has emphasized the influence mechanism of social networks. As the carrier of information transmission, the social network can effectively reduce transaction costs in farmland transfer by mitigating farmers’ risk perceptions to promote their long-term participation in farmland transfer [16,17]. However, social capital is multidimensional and heterogeneous, and considering only a single dimension of social capital may not provide a comprehensive understanding of farmers’ farmland transfer decisions. In addition to the social network, social norms can regulate farmers’ behavior through supervision and restraint mechanisms [18], while social trust can reduce transaction risk through a trust guarantee mechanism [19]. Therefore, by improving the explanatory perspective and research content, this study comprehensively explores the influence of social capital on agricultural land transfer decisions from three aspects—social networks, social trust, and social norms—in order to provide a reference for promoting agricultural land transfer practices.

At the present stage, China’s agricultural land transfer market has gradually improved. However, different agricultural land transfer market levels in various regions are due to variations in policy execution and regional development. The effect of social capital on agricultural land transfer may be heterogeneous under different levels of regional market development because the market can also facilitate land transfer [20]. The impact of the agricultural land transfer market on social capital at different stages is different. Therefore, based on the analysis of the influence of social capital on agricultural land transfer decisions, it is meaningful to analyze how the factor mobility affects social capital change during the marketization of agricultural land transfer. Regarding the research on the impact of marketization on social capital, many scholars have explored the correlation between the two from a macro perspective, using the Chinese provincial marketization composite index to measure the degree of marketization. There are two main views about the impact of marketization on social capital. Some scholars believe that marketization has a substitution effect on social capital [21]. Other scholars believe that marketization has a complementary effect on social capital [22]. The difference between the two views may be that varying aspects of social capital have varying effects at various phases of marketization. Social capital is embedded in rural society to influence farmers’ farmland transfer decisions, and the resource allocation function it carries may change with the marketization of farmland transfer. It can, thus, be seen that the relationship between social capital, marketization, and agricultural land transfer still needs to be studied. In addition, in terms of research perspective, this paper chooses to measure the marketization level of agricultural land transfer by selecting evaluation indicators from a micro perspective, which is different from the previous provincial marketization indices and can better solve the problem of the unsynchronized regional market and internal rural market. In other words, based on 1017 questionnaires from farmers in Hubei Province, we constructed evaluation indicators to measure the marketability index of agricultural land transfer, and measured social networks, social trust, and social norms to answer the following two research questions: First, does social capital significantly contribute to farmers’ agricultural land transfer decisions? Second, what is the impact of the marketization of agricultural land transfer on rural social capital, or has the resource allocation function carried by social capital changed during the marketization of agricultural land transfer? Answering these two questions has important implications for improving the social capital theory with Chinese characteristics and institutional innovation of agricultural land transfer. Also, it provides new policy references for promoting agricultural land transfer.

To sum up, the possible contributions of this paper are reflected in the following. First, the research perspective integrates the influence of social capital in multiple dimensions on agricultural land transfer from a sociological perspective. Although some studies have recognized the role of social capital, scholars generally consider the influence of social capital on agricultural land transfer decisions from a single dimension. However, social capital is a multidimensional concept, and the influence of social capital on agricultural land transfer decisions from various aspects still needs to be improved. Second, the influence of social capital varies in the research content depending on the situation of the farmland transfer market. However, the existing studies have less emphasis on and analysis of the differences in farmers’ farmland transfer behavior in different farmland transfer market environments. This paper incorporates agricultural land transfer markets into the study. Comparatively, it analyzes the influence of social capital on agricultural land transfer under different degrees of agricultural land transfer marketization to provide policy recommendations for establishing a tangible market for agricultural land transfer.

2. Theoretical Framework

With the problem of rural governance becoming prominent, the social capital theory has been introduced into various fields of rural social research. Existing studies have shown that social capital significantly impacts the mental status of the elderly in rural areas, farmers’ life satisfaction, rural industry development, rural revitalization, farmers’ increased income, and so on [23,24,25,26,27]. The essence of agricultural land transfer is a comprehensive decision made by both sides of the circulation in considering the multiple economic and social returns. Its transaction behavior is also affected by social factors [28]. Therefore, based on the social capital theory, this study embeds farmers’ farmland transfer decision making into the social environment for analysis.

2.1. Impact of Social Capital on Agricultural Land Transfer

Social network refers to the relationship network formed by long-term interaction and connection between people. Network size, density, and heterogeneity are three critical attributes of social networks [29]. Researchers have hypothesized that social networks with larger network sizes, higher network height, and higher network density may provide more affluent social capital and have tested this hypothesis in empirical studies [30,31]. In this paper, we define three attributes of social networks in terms of network breadth, network strength, and network height. Among them, the breadth reflects the scope and scale of social network coverage. The intensity reflects the closeness of interaction among network members, and highly reflects the top resources of the network that individuals can access. Network breadth is conducive to broadening the channels of information acquisition, promoting farmers to obtain trading individuals that meet the target conditions in the network nodes, realizing effective matching between supply and demand, and promoting the conclusion of transactions; the network intensity is advantageous to strengthen the trust degree between the central bodies. The network node with a close connection can better mobilize the network resources and promote the realization of agricultural land transfer. Network height is conducive to providing farmers with heterogeneous information, alleviating the dilemma of information asymmetry within the village, improving farmers’ cognitive level, and then affecting farmers’ behavior and decision making. Therefore, the social network is helpful for farmers to obtain and transmit the transaction information of farmland transfer with low cost and high efficiency, and promotes farmers to participate in farmland transfer.

Social trust refers to people’s communicative attitudes formed in the process of social interaction. The social trust of farmers mainly includes two types; one is the special trust of farmers in informal subjects such as relatives, friends, and fellow villagers, and the other is the general trust of farmers in formal subjects such as government and the collective [32,33]. In vast rural areas, rural members build trust through a long-term game. When the subject of the mutual trust transfers farmland, both sides do not need to pay extra costs to understand the land information and the information of the transaction subject, which effectively reduces the transaction cost [34]. At the same time, social trust can reduce the expectation of opportunistic behavior between the two sides, reduce the perceived risk that farmers experience while participating in farmland transfers, and encourage transaction subjects’ collaboration [35]. In addition, the local community and the government are the two main action forces that encourage farmland transfer [36]. The greater the farmers’ confidence in the village collective and the government, the higher the positive evaluation of the government in the development of the village, and farmers’ motivation to respond to government calls increases as their trust in the government grows [37,38].

Social norms are a set of norms and standards formed over a long period within a specific social group to restrict the behavior of social members [39,40]. As an invisible restraint mechanism and guarantee mechanism, social norms make the members embedded in the rural social network supervised by other network members. Individuals avoid social opinion and conscience condemnation by consciously following the rules of the village group [41,42]. Social norms include both imperative and descriptive categories [43,44]. The former refers to what others think an individual should do in a certain case, manifested in the individual’s perception of the pressure and expectations of social groups [45]. Therefore, the perception of individual farmers on the expectations and pressures of the government and village collectives will affect their decision-making behavior. Farmers’ likelihood of taking part in agricultural land transfer will rise if the government and local collectives encourage farmers to do so. The latter refers to the impact of the actions taken by others in a specific context on the individual, manifested in the individual’s perception of the majority’s behavior in the social situation. When most rural households are involved in a particular social activity, they are influenced by the herd mentality to actively participate in activities [46].

Hypothesis H1.

Social networks, social trust, and social norms have a positive impact on agricultural land transfer.

2.2. Regulation Effect of Marketization of Agricultural Land Transfer

Because the agricultural land transfer mechanism has yet to be established and perfected, social capital, as an informal system, makes up for the lack of a market mechanism. Farmers realize the effective allocation of land resources using social capital. With China’s agricultural land transfer market gradually maturing, the transaction scope of agricultural land circulation has gradually expanded. The original agricultural land circulation mode relying on an acquaintance relationship network and a human trust mechanism is limited. At the same time, the transaction in a broader range depends on the market operation mechanism [47]. Relevant studies show that the marketization of agricultural land transfer has a moderating effect between social capital and agricultural land transfer. The relevant mechanisms can be summarized as the following three aspects: (1) With the improvement of the marketization level of agricultural land transfer, the access channels and transmission mechanism of circulation information are also improved, and the adequate mobility of information is increased [48]. Furthermore, the circulation object’s marketization and the circulation mode’s liberalization significantly expand the scale of agricultural land transfer transactions, which favors removing the geographic limitations of the traditional agricultural land transfer market and expanding the agricultural land transfer market’s reach. (2) With the improvement of the agricultural land transfer market and the gradual establishment of farmers’ contract consciousness, the traditional customary trust based on emotional ties among acquaintances is gradually replaced by the contractual trust based on laws and regulations, and the standardized transaction procedures and contracts provide a guarantee for the establishment of contractual trust. Good contractual arrangements can serve to reduce transaction costs and guard against opportunism. (3) Social norms play the role of informal institutions in facilitating agricultural land transfer through moral constraints and herd mentality. Then, the market—as a new system of rules—mainly restricts the rights and obligations of both parties through contracts, which replace oral agreements to prevent market transaction risks.

Hypothesis H2.

The marketization of agricultural land transfer negatively moderates the relationship between social networks, social trust, social norms, and agricultural land transfer.

Figure 1 shows the analytical framework developed in this study. Social capital is incorporated into the analytical framework of agricultural land transfer decision, the influence of social capital on agricultural land transfer decision is analyzed, and marketization of agricultural land transfer is introduced as a moderating variable to test the moderating effect.

3. Materials and Methods

3.1. Data Sources

Considering the regional differences in social and economic development and topography in Hubei Province, seven cities (prefectures) in Hubei Province, including Wuhan, Yichang, Jingzhou, Xiaogan, Tianmen, Jingmen, and Enshi, are selected as empirical regions. Two phases of the survey were completed. The first step was to finish the research work of Jingzhou City, Xiaogan City, Tianmen City, and Jingmen City from 15 December to 24 December 2020, and the second stage was to complete the research work of Wuhan City, Yichang City, and Enshi Prefecture from 8 March to 15 March 2021. In this survey, 54 administrative villages of 1021 farmers were chosen from 18 townships throughout eight counties as the subject of investigation. A total of 1017 valid questionnaires were obtained after questionnaire screening. Among the interviewed households, 499 households had farmland transferred out, 137 households had farmland transferred in, and 381 households had not transferred farmland. The survey’s primary components include the following: the village characteristics, the basic situation of the interviewed peasant households, the status of agricultural land transfer, and social capital.

Table 1. Essential characteristics of sample farmers in the study area.

Features	Category	Frequency/Person	Proportion/%
Age	<35 years old	7	0.69
	35~45 years	43	4.23
	46~55 years	234	23.01
	56~65 years	326	32.06
	Age 66 and over	407	40.02
Education level	Never attended school	299	29.40
	Primary school	425	41.80
	Junior middle school	203	20.00
	High school	82	8.10
	College and above	8	0.80
Gender	Male	631	62.00
	Female	386	38.00
Annual household income	Less than 10,000 yuan	31	3.05
	10,000~30,000 yuan	131	12.88
	30,000~50,000 yuan	136	13.37
	50,000~80,000 yuan	198	19.47
	80,000~140,000 yuan	272	26.75
	140,000~200,000 yuan	127	12.49
	More than 200,000 yuan	122	12.00

displays the descriptive statistics’ findings on the questioned farmers’ fundamental traits. Regarding gender, most of the farmers interviewed were male, accounting for 62%; in terms of age, most respondents were older than 46, accounting for 95.09%. In terms of educational level, the educational level of most respondents was lower than that of high school, accounting for 91.2%. From the perspective of annual household income, the annual income of peasant households is less than 140,000 yuan, accounting for 75.52%.

3.2. Variable Selection

3.2.1. Explained Variables

Farmers’ farmland transfer decision. This is measured by whether the sample farmers participate in the transfer of farmland, including “1 = transfer-in or transfer-out of farmland” and “0 = no transfer of farmland” (

Table 2. Variable names and descriptive statistics.

Variable Name	Variable Description	Minimum	Maximum	Mean	S.D.
Transfer-out or not	Yes = 1; No = 0	0.000	1.000	0.571	0.494
Transfer-in or not	Yes = 1; No = 0	0.000	1.000	0.263	0.440
Social networks	Calculated value by entropy method	0.005	0.653	0.143	0.109
Social trust	Calculated value by entropy method	0.000	1.000	0.604	0.201
Social norms	Calculated value by entropy method	0.000	1.000	0.626	0.220
Gender	Male = 1; Female = 0	0.000	1.000	0.620	0.486
Age	The actual age of the interviewee/year	30.000	87.000	61.540	9.660
Level of education	No schooling = 1; Primary = 2; Junior high school = 3; High school (technical secondary school) = 4; College and above = 5	1.000	5.000	2.090	0.938
Area of household contracted land	Household-contracted cultivated land area	0.000	30.000	7.1514	4.398
Overall quality of cultivated land	Very poor~very good = 1~5	1.000	5.000	3.620	0.799
Family size	Family population/unit	1.000	11.000	4.100	1.608
Number of farmers participating in cooperatives	Number of participating farmers’ cooperatives/unit	0.000	5.000	0.260	0.557
Landform of the village	Plain = 1; Hill = 2; Mountain = 3	1.000	3.000	1.660	0.764
Distance between the village and the nearest town	Distance from rural household to urban government/km	5.000	66.000	26.030	13.936
Marketization degree of agricultural land transfer	Calculated value by entropy method	0.326	0.980	0.453	0.240

).

3.2.2. Explanatory Variables

(1) Social networks. The number of close relatives, friends, and neighbors can reflect the frequency and intensity of social interaction in the interpersonal relationship of farmers. The expenditure of human relations in one year can be regarded as the maintenance cost of farmers to the social network, and the cost of maintenance increases with network growth. Relatives and friends working in government agencies, as the critical resources of interpersonal communication, reflect the high degree of the social network. This study selects the above three proxy variables to measure the comprehensive index of the social network. (2) Social trust. Social trust is the basis of all activities. The questionnaire set up four indicators to measure social trust: farmers’ confidence levels in friends and family, their fellow villages, village collectives, and the government [49,50]. The trust degree scores were assigned from low to high as 1–5. The comprehensive index of social trust was measured. (3) Social norms. Regarding farmland transfer, social norms mainly refer to farmers’ subjective perception of relatives and friends, fellow villagers, village collectives, and the government supporting or opposing farmland transfer. Referring to a related study [51], the four questions in the questionnaire are as follows: “the enthusiasm of relatives and friends to take part in farmland transfer is high”; “the enthusiasm of villagers to take park in farmland transfer is high”; “the village collective encourages villagers to participate in agricultural land transfer”; and “the local government encourages villagers to take part in agricultural land transfer”. The comprehensive index of social norms is calculated based on the degree of farmers’ agreement with the problem, which is assigned from low to high as 1–5. Table 2 provides a detailed definition of each variable.

3.2.3. Control Variables

The basic situation of farmers (age, gender, education level); the basic situation of families (the number of farmers joining the cooperatives, the number of family members); land endowment (the area of household-contracted land, the overall quality of cultivated land); and village characteristics (village topography, and the separation between the settlement and the closest town) are selected as the control variables [52,53,54] (Table 2).

3.2.4. Adjustment Variable

The marketization of farmland transfer. The overall agricultural land transfer market is evaluated from three aspects: the balance of market supply and demand, the level of market standardization, and the perfection of supporting mechanisms. A well-developed market can achieve the state of supply and demand balance through the price function, so the agricultural land transfer rate is used to measure farmland’s equilibrium state of supply and demand. The higher the circulation rate, the more active the market is, indicating a higher degree of marketization of agricultural land transfer; a standardized agricultural land transfer market is the guarantee to promote the orderly circulation of farmland. This study selects the paid circulation rate of farmland; the open circulation rate of farmland; the stability of the circulation period; the marketization of circulation objects; and the liberalization of circulation mode to measure the degree of market standardization from the micro level. The agricultural land transfer market needs the government to publicize the relevant policies, so the promotion rate of the relevant policies is selected to quantify the perfection of the market-supporting mechanism. On this basis, the entropy method calculates the degree of agricultural land transfer marketization (Table 2).

3.3. Model Construction

3.3.1. Probit Model

Whether farmers participate in farmland transfer is a binary classification variable. Therefore, the Probit model is chosen to examine how social capital affects the choice to transfer lands. The following is how the Probit model is built:

Y = α + β_i x_i + γ θ + ε

(1)

In Equation (1): Y is the farmland transfer decision; ${\mathrm{x}}_{\mathrm{i}}$ is each variable characterizing social capital; $\mathsf{\theta }$ is the control variable; α is the constant term; β_i and γ are the coefficients to be estimated; and ε is the random error term.

3.3.2. SUEST Inspection

In order to investigate the impact of social capital on agricultural land transfer decision making under various marketization scenarios, this study makes a grouping regression of farmers’ samples according to the degree of agricultural land transfer marketization. It compares the differences in regression coefficients among the sub-samples. Because there may be a bias in interpreting the results by comparing the significance level of regression coefficients of two samples alone, the moderating impact of marketization is examined using the SUEST test technique by comparing the differences of regression coefficients between groups under different degrees of marketization. The model of the subgroup regression is as follows:

Y₁ = α₁ + β_1ix_i + γ₁θ₁ + ε₁

(2)

Y₂ = α₂ + β_2i x_i + γ₂θ₂ + ε₂

(3)

First, on the basis of the degree of marketization of agricultural land circulation, the samples are split into two groups. Then the Probit model regression is carried out based on the two sub-samples according to Formulas (2) and (3). Finally, the difference in regression coefficients between the two groups was compared using the SUEST test. If there is a significant difference between ${\mathsf{\beta }}_{1\mathrm{i}}$ and ${\mathsf{\beta }}_{2\mathrm{i}}$, research Hypothesis 2 is verified.

4. Results

4.1. Analysis of the IMPACT of Social Capital on Agricultural Land Transfer

4.1.1. Analysis of the Impact of Social Capital on the Farmland Transfer-Out

The column (1) results in

Table 3. Impact of social capital on farmland transfer decision.

	Transfer-Out or Not				Transfer-In or Not
Variable Name	(1) Probit	(2) Marginal Effect	(3) Probit	(4) Marginal Effect	(5) Probit	(6) Marginal Effect	(7) Probit	(8) Marginal Effect
Social networks	1.490 ** (2.17)	0.439 ** (2.19)			2.861 *** (2.82)	0.528 *** (2.90)
Social network breadth			0.788 * (1.83)	0.230 * (1.84)			0.995 * (1.68)	0.181 * (1.67)
Social network strength			1.301 ** (2.23)	0.379 ** (2.25)			2.248 ** (2.51)	0.409 ** (2.51)
Social network height			0.033 (0.10)	0.010 (0.10)			0.296 (1.36)	0.054 (1.37)
Social trust	1.563 *** (3.70)	0.461 *** (3.80)			2.604 *** (3.69)	0.480 *** (3.78)
Special trust			1.266 *** (3.21)	0.369 *** (3.27)			1.271 ** (2.22)	0.231 ** (2.25)
General trust			0.391 (1.16)	0.114 (1.16)			1.171 ** (2.05)	0.213 ** (2.09)
Social norms	1.740 *** (5.40)	0.460 *** (4.65)			1.688 *** (2.76)	0.311 *** (2.82)
Descriptive specification			0.652 ** (2.39)	0.190 ** (2.42)			0.558 (1.13)	0.102 (1.12)
Imperative specification			0.987 *** (3.25)	0.288 *** (3.32)			1.229 *** (2.93)	0.224 *** (2.97)
Gender	0.074 (0.73)	0.022 (0.73)	0.087 (0.84)	0.025 (0.84)	0.109 (0.67)	0.020 (0.67)	0.096 (0.57)	0.017 (0.58)
Age	−0.002 (−0.24)	−0.000 (−0.24)	−0.002 (−0.24)	−0.000 (−0.24)	−0.008 (−0.76)	−0.001 (−0.76)	−0.005 (−0.47)	−0.001 (−0.47)
Level of education	−0.223 *** (−3.29)	−0.066 *** (−3.35)	−0.250 *** (−3.52)	−0.073 *** (−3.60)	−0.017 (−0.16)	−0.003 (−0.16)	−0.013 (−0.13)	−0.002 (−0.13)
Area of household-contracted land	0.043 *** (3.59)	0.013 *** (3.68)	0.046 *** (3.71)	0.013 *** (3.82)	−0.032 (−1.48)	−0.006 (−1.49)	−0.032 (−1.47)	−0.006 (−1.48)
The overall quality of cultivated land	−0.059 (−0.96)	−0.017 (−0.96)	−0.057 (−0.89)	−0.017 (−0.89)	−0.174 * (−1.78)	−0.032 * (−1.79)	−0.148 (−1.20)	−0.027 (−1.22)
Family size	−0.082 * (−1.90)	−0.024 * (−1.92)	−0.080 * (−1.84)	−0.023 * (−1.85)	0.327 *** (4.61)	0.060 *** (4.84)	0.332 *** (4.88)	0.060 *** (5.16)
Number of farmers participating in cooperatives	2.185 *** (7.60)	0.644 *** (8.22)	2.199 *** (7.62)	0.641 *** (8.24)	2.141 *** (6.14)	0.395 *** (6.66)	2.048 *** (4.35)	0.372 *** (4.99)
Number of non-farm employed labor force	0.165 *** (2.95)	0.049 *** (3.00)	0.161 *** (2.86)	0.047 *** (2.90)	−0.341 *** (−4.00)	−0.063 *** (−4.14)	−0.358 *** (−4.21)	−0.065 *** (−4.43)
Landform of the village	−0.034 (−0.46)	−0.010 (−0.46)	−0.043 (−0.58)	−0.013 (−0.59)	0.169 (1.45)	0.031 (1.45)	0.172 (1.35)	0.031 (1.38)
Distance between the village and the nearest town	−0.007 * (−1.71)	−0.002 * (−1.72)	−0.005 (−1.40)	−0.002 (−1.40)	−0.006 (−0.94)	−0.001 (−0.94)	−0.005 (−0.80)	−0.001 (−0.81)
The marketization of agricultural land circulation	0.599 *** (2.73)	0.175 *** (2.77)	0.602 *** (2.60)	0.174 *** (2.63)	1.05 *** (2.72)	0.181 *** (2.77)	1.027 *** (2.61)	0.175 *** (2.65)
Wald chi-square value	139.810		155.170		1117.17		114.65
p value	0.0000		0.0000		0.0000		0.0000
Observation value	1017		1017		1017		1017

Notes: ***, **, * represent significance at the 1%, 5%, and 10% levels, respectively. Bracketed numbers indicate t-values.

show that social networks, social trust, and social norms all promote farmers’ farmland transfer-out decisions. The social network significantly positively impacts farmers’ farmland transfer-out at 5%. Among them, the strength and breadth of the social network significantly promote farmers’ farmland transfer-out decisions, and the more robust the social network, the wider the channels for farmers to learn about farmland transfer are, which is more favorable to lowering transaction costs and moral hazard in farmland transfer and promoting farmers’ farmland transfer. Social trust significantly promotes farmers to transfer-out farmland at 1%. From the perspective of each variable’s action path, special trust significantly promotes farmers’ decision making of farmland transfer-out, and farmers’ choice of transferring land to mutually trusted farmers helps to reduce the uncertainty in the process of farmland transfer, thus improving farmers’ willingness to transfer-out farmland. The transfer-out of farmlands is not significantly impacted by general trusts, which may be because the role of special trust in agricultural land transfer inhibits the role of general trust. Farmers’ choice of acquaintances to transfer-out farmland still has an advantage in rural areas. Social norms have a strongly favorable effect on farmers’ transfer-out of farmland at the level of 1%. Among them, descriptive and imperative standards support farmers’ decisions to transfer-out farmland. When the householder feels that most farmers in the village participate in the farmland transfer-out, they will follow the majority’s decision to participate in the transfer of farmland out. The government considers the herd mentality of farmers and guides farmers to actively participate in the transfer-out of agricultural land through policy incentives. In addition, the marketization of agricultural land transfer significantly promotes the farmers’ farmland transfer-out at the 1% level. A perfect agricultural land transfer market encourages the transfer-out of farmland.

4.1.2. Analysis of the Impact of Social Capital on the Farmland Transfer-In

The column (5) results in Table 3 show that social networks, social trust, and social norms all promote farmers’ farmland transfer-in decisions. The social network significantly positively impacts farmers’ farmland transfer-in decisions at 1%. The strength and breadth of social networks positively impact farmers’ farmland transfer-in decisions. Farmers with a greater “number of close contacts” and more “annual human exchange expenditure” have a stable network relationship. Farmers with sufficient social networks can obtain and publish complete land transfer information through their relationship network to encourage farmers’ farmland transfer-in. Social trust significantly promotes farmers’ transfer in farmland, and the special trust relationship between relatives and friends in rural society can promote farmers’ reciprocal cooperation. Meanwhile, the general trust in the village collective and government can effectively reduce the risk loss expectation of transferring to farmland and promote farmers’ farmland transfer-in. Social norms promote the farmers’ farmland transfer-in. From the perspective of the practical path of each variable, the imperative social norms significantly promote farmers’ decision making of farmland transfer-in, the government supports farmers to participate in farmland transfer, and farmers are influenced by policy incentives, which helps to improve farmers’ willingness and ability to participate. Descriptive norms do not impact farmland transfer-in choices made by farmers. Farmers’ choice of farmland transfer-in is constrained by many conditions, such as agricultural production capacity, agricultural labor force, input costs, and other aspects of a higher threshold, so the active participation of relatives, friends, and fellow villagers in farmland transfer has no significant effect on farmers’ transfer-in decision. The marketization of agricultural land transfer significantly promotes farmers to transfer-in farmland at 1%, and the improvement of marketization helps promote farmland transfer-in.

4.1.3. A Comparative Analysis of Social Capital on Farmland Transfer-In and Farmland Transfer-Out Decisions

The results in columns (2) and (6) of Table 3 show that the marginal effects of the social network, social trust, and social norms on the farmland transfer-out decision were 0.439, 0.461, and 0.460, respectively. The marginal effects on farmland transfer in the decision were 0.528, 0.480, and 0.311, respectively. The comparison results show that the marginal effects of the social network and social trust on farmland transfer-in are greater than those on farmland transfer-out. The marginal effects of social norms on farmland transfer-out are greater than those on farmland transfer-in. One possible explanation is that the transfer-in agents generally need to negotiate with multiple transfer-out agents, and the diversity of transaction agents brings higher transaction costs and uncertainty. The marginal effects of both descriptive and imperative norms on farmland transfer-out are higher than their marginal effects on farmland transfer-in. In rural areas, if most farmers participate in farmland transfer-out, other farmers may also decide to transfer their land. Farmers’ willingness to participate in the activity will increase with government encouragement; however, farmland transfer-in is often constrained by farmers’ ability, capital, and technology, and they will have less chance of farmland transfer-in. Therefore, the effect of social norms affecting farmers’ farmland transfer-out is far greater than farmland transfer-in.

4.2. The Moderating Effect of Marketization of Agricultural Land Transfer between Social Capital and Agricultural Land Transfer Decisions

4.2.1. Measurement of the Degree of Marketization of Agricultural Land Transfer and Analysis of the Results

Seven indicators—the transfer rate of agricultural land, the paid transfer rate of agricultural land, the open transfer rate of agricultural land, the stability of the transfer period, the marketization of transfer objects, the free transfer mode, and the promotion rate of agricultural land transfer policy—are chosen in this study to quantitatively characterize the development of the agricultural land transfer market in each region. First, using the entropy value approach, the weights of the evaluation indices of the growth of the agricultural land transfer markets were established (

Table 4. Evaluation index weights of agricultural land transfer market.

	Rate of Agricultural Land Transfer	Rate of Paid Transfer of Agricultural Land	Rate of Open Transfer of Agricultural Land	Stability of Agricultural Land Transfer Period	Marketization of the Target of Agricultural Land Transfer	Liberalization of Agricultural Land Transfer Method	Promotion Rate of Agricultural Land Transfer policy
Weights	0.086	0.130	0.093	0.115	0.116	0.169	0.290

). The degree of complete development of the agricultural land transfer market was then determined in each region of Hubei Province using the comprehensive evaluation method (

Table 5. Evaluation results of the degree of marketization of agricultural land transfer.

Region	Rate of Agricultural Land Transfer	Rate of Paid Transfer of Agricultural Land	Rate of Open Transfer of Agricultural Land	Stability of Agricultural Land Transfer Period	Marketization of the Target of Agricultural Land Transfer	Liberalization of Agricultural Land Transfer Method	Promotion Rate of Agricultural Land Transfer Policy	Degree of Marketization of Agricultural Land Transfer
Wuhan	0.043	0.064	0.093	0.115	0.031	0.169	0.290	0.805
Yichang	0.076	0.112	0.072	0.113	0.098	0.147	0.270	0.889
Jingzhou	0.000	0.130	0.066	0.105	0.116	0.016	0.080	0.514
Xiaogan	0.086	0.000	0.000	0.000	0.000	0.125	0.087	0.298
Jingmen	0.086	0.088	0.061	0.076	0.052	0.000	0.012	0.376
Enshi	0.079	0.030	0.058	0.088	0.076	0.035	0.012	0.379
Tianmen	0.071	0.042	0.037	0.024	0.058	0.124	0.000	0.357

). This study examines the elements with the lowest scores as the barriers to the growth of the regional agricultural land transfer market to analyze the current issues in developing each region’s agricultural land transfer market. The one having the clearest restrictions on the rate of agricultural land transfer is Jingzhou City. According to the interviews with farmers in Jingzhou City on the reasons for not participating in agricultural land transfer, farmers generally indicate that they are fully capable of farming and do not wish to take part in the transfer of agricultural land. Farmers’ willingness to transfer agricultural land restricts the development of the regional agricultural land transfer market. The most apparent constraint by the standardization of the market for agricultural land transfers is Xiaogan City, which has an active invisible market for agricultural land transfer. The central bodies of transfer are mostly small farmers; the form of transfer is mainly in the form of oral contract; the duration of the transfer is short term; and the rent of transfer is low primarily or even no rent. The most obvious restriction by the liberalization of circulation mode is Jingmen City. Despite the high rate of agricultural land transfers in this area, the government-led approach is generally used as a solid force for promoting agricultural land transfer. The lack of respect for farmers’ wishes during the transfer process, the lack of post-transfer supervision, and farmers’ satisfaction with the transfer of agricultural land have decreased, and this has had an impact on the steady growth of the regional agricultural land transfer market. In Tianmen City, where farmers frequently lack knowledge of agricultural land transfer policies, and where lack of policy awareness among farmers limits the growth of the regional agricultural land transfer market, the low promotion rate of agricultural land transfer policy is the limiting factor in the region.

4.2.2. Regression Analysis of the Regulating Effect of Market-based Agricultural Land Transfer

In this study, the average value of the marketization of agricultural land circulation is used as the criterion for grouping, and those whose values fall below the average are split into low-value groups. In contrast, those above the average value are divided into high-value groups. Based on the SUEST test method after grouping regression, the significant changes of the two groups of coefficients were compared to verify the regulatory effect of the marketization of agricultural land circulation on social capital and agricultural land circulation.

Table 6. Analysis of the effect of marketization on social capital influencing agricultural land transfer.

Explain Variables	Transfer-Out or Not		Transfer-In or Not
	Low-Value Group	High-Value Group	Low-Value Group	High-Value Group
Social networks	3.511 *** (3.77)	1.442 * (1.64)	7.469 *** (5.75)	3.075 * (1.92)
Social trust	2.584 *** (3.83)	0.976 (1.58)	6.047 *** (3.97)	2.047 * (1.75)
Social norms	1.740 *** (5.40)	2.391 *** (4.89)	2.211 * (1.92)	2.135 ** (1.98)
Gender	−0.058 (−0.34)	0.106 (0.79)	−0.407 (−1.41)	0.752 ** (2.54)
Age	0.009 (0.92)	−0.005 (−0.65)	−0.014 (−0.77)	−0.002 (−0.14)
Level of education	−0.198 * (−1.84)	−0.220 ** (−2.38)	−0.066 (−0.34)	−0.068 (−0.38)
Area of household-contracted land	0.020 (0.99)	0.062 *** (3.63)	−0.025 (−0.59)	−0.003 (−0.10)
The overall quality of cultivated land	−0.062 (−0.59)	−0.126 (−1.55)	−0.266 (−1.57)	−0.142 (−0.86)
Family size	−0.098 (−1.30)	−0.106 * (−1.92)	0.284 ** (2.25)	0.325 *** (2.73)
Number of farmers participating in cooperatives	2.022 *** (4.64)	2.492 *** (5.47)	1.751 *** (3.26)	2.662 *** (4.32)
Number of non-farm employed labor force	0.098 (1.11)	0.235 *** (2.91)	−0.430 *** (−2.84)	−0.341 ** (−2.19)
Landform of the village	0.223 ** (2.16)	−0.288 ** (−2.38)	0.282 (1.56)	−0.081 (−0.36)
Distance between the village and the nearest town	−0.010 (−1.57)	−0.003 (−0.44)	−0.002 (−0.18)	−0.005 (−0.43)
Wald chi-square value	105.620	230.050	170.740	150.560
p value	0.000	0.000	0.000	0.000
Observation value	342	538	252	266

Notes: ***, **, * represent significance at the 1%, 5%, and 10% levels, respectively. Bracketed numbers indicate t-values.

and

Table 7. Tests for differences in coefficients between groups based on the seemingly unrelated model.

Explain Variables	Transfer-Out or Not		SUR	Transfer-In or Not		SUR
	(1)	(2)	(3)	(4)	(5)	(6)
	Low-Value Group	High-Value Group	Chi2 and p	Low-Value Group	High-Value Group	Chi2 and p
Society Network	3.511 *** (3.77)	1.442 * (1.64)	Chi2 = 2.72 p = 0.099	7.469 *** (5.75)	3.075 * (1.92)	Chi2 = 3.85 p = 0.050
Society Trust	2.584 *** (3.83)	0.976 (1.58)	Chi2 = 3.21 p = 0.073	6.047 *** (3.97)	2.047 * (1.75)	Chi2 = 6.53 p = 0.011
Society Standard	1.740 *** (5.40)	2.391 *** (4.89)	Chi2 = 10 p = 0.002	2.211 * (1.92)	2.135 ** (1.98)	Chi2 = 0 p = 0.955
Observation value	342	538	880	252	266	518

Notes: ***, **, * represent significance at the 1%, 5%, and 10% levels, respectively. Bracketed numbers indicate t-values.

present the outcomes.

The test results of column (1) low-value group and column (2) high-value group in Table 7 show that the coefficient of the social network between the two groups is positive, the size has decreased, and a significant test of difference in coefficients between groups—that is, the improvement of marketization of agricultural land circulation—reduces the effect of social networks on the land transfer-out, which proves that marketization has a negative regulatory influence on social network and land transfer-out. The influence of social trust changes from positive significant to insignificant, and the influence coefficient decreases, which proves that the marketization of agricultural land transfer has a negative moderating influence between social trust and land transfer-out. The effect of social norms was significant and positive in both groups, with increased impact coefficients and significant between-group coefficient difference tests, demonstrating the positive moderating effect of marketization on social norms and land transfer-out.

The test results of column (4) low-value group and column (5) high-value group in Table 7 show that social network and social trust have significant positive effects between the two groups. However, the influence coefficients are reduced to a certain extent. The test of coefficient differences between groups is significant, which proves that the marketization of agricultural land transfer negatively regulates the role of social networks and trust in the land transfer-in. The coefficient of social norms in the two groups is positive, and the magnitude of change is small. Moreover, according to the test, there is no statistically significant variation between the two groups’ regression values, demonstrating that the marketization of agricultural land transfer does not significantly regulate social norms and agricultural land transfer-in.

The following are some potential causes: the farmland transfer market broadens the channels for farmers to obtain resource information, and the market gradually becomes the main body of resource allocation, weakening the role of social networks. Whether for farmland transfer-out or farmland transfer-in, the contractual trust formed by marketization will weaken the impact of human trust on farmland transfer, and the standard of transaction realization is not entirely the trust mechanism constructed by consanguinity and kinship between farmers. With the development of marketization, a more significant role for social norms in the decision of agricultural land transfer-out may be due to the function of government in providing policy guidance and legal protection in the construction of the farmland transfer market system. The survey found that the government’s support is more vital in areas with a higher degree of marketization of farmland transfer. Government departments and village committees can enhance farmers’ policy perception and promote farmland transfer-out decision making by publicizing and releasing relevant policies and information on farmland transfer to farmers. With the development of marketization, there has been little change in the influence of social norms on farmland transfer-in decisions. This may be explained by the fact that farmers face higher capital investment and risk losses when transferring into farmland than when transferring out of farmland, and subjects participating in agricultural land transfer-in tend to have a higher awareness of risk prevention and clear awareness of farmland policy, so the promotion of marketization of farmland transfer will not significantly affect the impact of social norms on farmland transfer-in decision making.

4.3. Robustness Test

4.3.1. Subsample Test Method

To further verify the reliability of the research results, this paper adopts a split-sample regression method to divide the sample data into the Jianghan Plain area and the Ecological and Cultural Tourism Circle of Western Hubei, and estimates the correlation between farmers’ social capital and farmland transfer decision in the two regions separately to verify the influence of regional differences on the overall sample regression results. The results in

Table 8. Estimation results of the sample model in the Jianghan plain area.

	Transfer-Out or Not				Transfer-In or Not
Variable Name	(1) Probit	(2) Marginal Effect	(3) Probit	(4) Marginal Effect	(5) Probit	(6) Marginal Effect	(7) Probit	(8) Marginal Effect
Social networks	2.369 ** (2.81)	1.033 ** (4.37)			4.518 *** (4.09)	0.791 *** (4.54)
Social network breadth			1.654 *** (2.76)	0.536 *** (2.82)			1.332 * (1.47)	0.265 * (1.50)
Social network strength			2.227 *** (2.91)	0.721 *** (3.01)			2.589 * (1.82)	0.515 * (1.86)
Social network height			0.430 (0.95)	0.139 (0.95)			0.527 (0.99)	0.105 (1.00)
Social trust	2.739 *** (4.11)	0.692 *** (3.65)			5.957 *** (4.99)	1.044 *** (5.75)
Special trust			1.157 ** (2.04)	0.375 ** (2.07)			1.580 * (1.75)	0.314 * (1.78)
General trust			0.118 (0.24)	0.383 (0.24)			0.537 *** (3.33)	0.505 *** (3.52)
Social norms	1.997 *** (3.73)	0.616 *** (4.02)			2.332 *** (2.59)	0.409 *** (2.68)
Descriptive specification			1.160 *** (3.04)	0.376 *** (3.14)			0.466 (0.62)	0.093 (0.62)
Imperative specification			1.555 *** (3.58)	0.504 *** (3.75)			2.284 *** (3.08)	0.455 *** (3.26)
Control variables	Control	Control	Control	Control	Control	Control	Control	Control
Wald chi-square value	93.19		101.72		119.40		95.05
p value	0.0000		0.0000		0.0000		0.0000
Observation value	420		420		261		261

Notes: ***, **, * represent significance at the 1%, 5%, and 10% levels, respectively. Bracketed numbers indicate t-values.

and

Table 9. Estimation results of the sample model in the Ecological and Cultural Tourism Circle of Western Hubei.

	Transfer-Out or Not				Transfer-In or Not
Variable Name	(1) Probit	(2) Marginal Effect	(3) Probit	(4) Marginal Effect	(5) Probit	(6) Marginal Effect	(7) Probit	(8) Marginal Effect
Social networks	1.602 * (1.84)	0.454 * (1.86)			3.219 *** (2.60)	0.616 *** (2.74)
Social network breadth			0.567 (0.97)	0.191 (0.97)			0.216 (0.22)	0.043 (0.22)
Social network strength			1.337 * (1.43)	0.450 * (1.44)			2.866 ** (2.15)	0.568 ** (2.20)
Social network height			0.151 (0.38)	0.507 (0.38)			0.189 (0.31)	0.374 (0.31)
Social trust	2.389 *** (3.86)	0.678 *** (4.09)			4.531 *** (4.30)	0.867 *** (4.86)
Special trust			1.091 ** (2.15)	0.367 ** (2.19)			1.796 * (1.76)	0.356 * (1.79)
General trust			0.666 * (1.58)	0.224 * (1.59)			2.372 *** (2.66)	0.470 *** (2.77)
Social norms	1.208 *** (2.72)	0.343 *** (2.79)			2.584 *** (3.36)	0.495 *** (3.62)
Descriptive specification			1.436 *** (3.49)	0.483 *** (3.64)			0.892 (0.93)	0.177 (0.94)
Imperative specification			0.355 (0.95)	0.120 (0.95)			0.363 * (1.74)	0.270 * (1.77)
Control variables	Control	Control	Control	Control	Control	Control	Control	Control
Wald chi-square value	156.03		148.41		130.26		127.54
p value	0.0000		0.0000		0.0000		0.0000
Observation value	462		462		252		252

Notes: ***, **, * represent significance at the 1%, 5%, and 10% levels, respectively. Bracketed numbers indicate t-values.

show that the direction and significance of the influence of social capital on farmland transfer decisions are similar to those in Table 3, indicating that the estimation results of the model in this paper are robust.

4.3.2. Model Replacement Method

The impact of different model selections is taken into account, and the interaction test was used to replace the group regression model to demonstrate the robustness of the regulation effect. By constructing the interaction terms of the social network, social trust, social norms and marketization, the impact of social capital on the decision making of agricultural land transfer in the marketization process is studied from the symbol and significance of the interaction terms.

Table 10. Robustness test of regulation effect: interaction test.

Variable Name	Transfer-Out or Not	Transfer-In or Not
Social networks	1.937 *** (3.25)	8.961 *** (5.44)
Social trust	0.513 * (1.65)	5.755 *** (3.48)
Social norms	1.753 *** (5.25)	2.918 ** (2.01)
Marketization	1.661 *** (4.95)	7.640 ** (2.21)
Social network * Marketization	−3.549 * (−1.70)	−12.830 *** (−2.95)
Social trust * Marketization	−2.700 * (−1.76)	−6.409 * (−1.67)
Social norms * Marketization	2.382 * (1.67)	−1.655 (−0.51)
Other variables	Control	Control
Wald chi-square value	334.080	311.450
p value	0.000	0.000
Pseudo R2	0.277	0.527
Observation value	1017	1017

Notes: ***, **, * represent significance at the 1%, 5%, and 10% levels, respectively. Bracketed numbers indicate t-values.

displays the outcomes of the robustness test. The interaction test results are consistent with the conclusions in Table 7, which verifies the robustness of the conclusions.

5. Discussion

In China’s particular interpersonal context, social capital can influence farmers’ behavioral decisions through information transfer, trust construction, and behavioral norms. Li et al. affirm social capital’s impact on cultivated land availability [55]. This paper draws on Putnam’s definition of social capital, makes a new division of social capital dimensions, and integrally examines the impact of different dimensions of social capital on agricultural land transfer decisions. We reaffirm the positive role of social capital in facilitating farmers’ agricultural land transfer decisions based on the different research dimensions. Although various factors may influence farmers’ agricultural land transfer decisions, the presence of social capital largely influences farmers’ behavior. Again, this result is consistent with Zhu et al.’s conclusion that social capital positively influences farmers’ environmental behavior [56] and Zhang et al.’s conclusion that social capital promotes farmers’ participation in domestic wastewater treatment [57], indicating that social capital still has an essential role in rural social development in various areas of rural social governance. Therefore, the comprehensive value of social capital needs to be addressed. It is worth noting that, in this study, in measuring the impact paths of each variable of the social network, social trust, and social norms, the impact of individual variables differs from the findings of existing related studies. In terms of social networks, existing studies classify network types according to the strength of social networks and analyze their effects on agricultural land transfer [58], based on which this paper expands the studies by measuring social capital in terms of network strength, network breadth, and network height. The findings show that social network strength and network breadth significantly contribute to agricultural land transfer decisions. In the area of social trust, social trust can facilitate farmers’ farmland transfer-in decisions, and this finding is similar to that of the Li et al. study [59]. Most studies have shown that social trust can also promote farmland transfer-out, but this paper’s results show that general trust’s effect on farmland transfer-out is insignificant. The reason for this is that, in the field research, the research shows that some regional governments pursued the speed and scale of farmland transfer in the form of administrative orders in order to pursue political performance, which frustrated farmers’ enthusiasm to participate in farmland transfer and also led to farmers’ distrust in the government. Therefore, in promoting the transfer of agricultural land, the government should respect the primary will of farmers and guide the standardized and orderly transfer of agricultural land. In terms of social norms, the effect of descriptive norms in social norms on the decision to transfer-out farmland is significant. However, the effect of the decision to transfer-in farmland is insignificant. The conclusion confirms that farmland transfer-out behavior in China shows a significant social interaction effect, consistent with the findings of the Fang et al. study [60], which concluded that the transfer-out behavior of other farmers positively influences individual farmers’ participation in farmland transfer. However, this interaction effect is not evident in the decision of farmland transfer-in, and it is a future research direction to work on how to let village cadres, party members, and competent rural people, such as large planters, play a leading role in guiding small farmers to participate in farmland transfer-in and realize the organic connection with modern agriculture.

Based on considering market-oriented factors, we further analyze the complex relationship between social capital, farmland transfer decisions, and farmland transfer market and incorporate the study of the proposition of the interaction between market mechanism and social capital into the analytical framework of this paper. Compared with official statistics, the survey data of the rural agricultural land transfer market used in this paper—with village area as the basic research unit—can reflect the current situation and effects of the agricultural land transfer market in rural areas more accurately and intuitively. The results of this paper show that the marketization of agricultural land transfer has a significant substitution effect on social capital, specifically the substitution of social networks and social trust by the marketization of agricultural land transfer. The increase in the marketization of agricultural land transfer weakens the influence of social networks and social trust on agricultural land transfer decisions. It is foreseeable that the agricultural land transfer market will play a more critical role in promoting agricultural land transfer. Therefore, it is necessary to emphasize that the agricultural land transfer policy should focus on constructing the agricultural land transfer market and play the role of the market mechanism in resource allocation. At the same time, considering the complexity of China’s rural problems and the vital role of social capital, it is also necessary to play the role of traditional rural society in governance and to realize the efficient integration of market mechanisms and social capital. In contrast to the existing theoretical assumptions, social norms are embedded in the agricultural land transfer market to exert a more significant influence. The agricultural land transfer market and social norms complement and jointly promote each other. The reputation mechanism assumed by social norms and the normative system formulated by the farmland transfer market both have the fundamental purpose of preventing opportunistic behaviors from occurring and maintaining the stability of farmland transfer transaction behaviors. When social norms and market mechanisms guide the behavior of agricultural land transfer in the same way, market mechanisms can operate dependent on the self-implementation mechanism of social norms, can provide institutional support for social norms, and can guide agricultural land transfer to become a universal behavior in rural society.

This study found that control variables such as education level, household-contracted land area, household size, and the number of non-farm employed laborers significantly affect farming households’ farmland transfer decisions. China’s rural areas have problems of hollowing out and aging, where middle-aged and older people have become the main force of agricultural production, limited by education level and farming ability, which will prompt middle-aged and older people with a lack of agricultural production capacity to transfer-out of farmland. In addition, as China’s urbanization continues, rural China provides a large portion of the labor force for China’s economic development, and increasing numbers of young and middle-aged laborers in rural areas are choosing to work outside the countryside. The allocation structure of the rural labor force between agricultural production and non-farm labor has changed significantly which, in turn, has an important impact on the agricultural production decisions of farm households. The more laborers in the household population are engaged in non-farm industries, the more it helps to positively promote farmland transfer-out and negatively inhibit farmland transfer-in, which is consistent with the findings of Xu et al. [61].

Our study also has some areas that could be improved. First, this paper only collects cross-sectional social capital and farmland transfer data. However, rural areas’ markets and social environment will change over time. In the future, we can establish panel data with long-term tracking to analyze the evolutionary process of rural social capital on farmland transfer decisions from a dynamic perspective. Second, the multi-group analysis needs to be explored in greater detail. This paper only divides the marketization stage of agricultural land transfer into two categories: low-value groups and high-value groups. The criteria for dividing agricultural land transfer marketization stages need to be improved and precise. A more in-depth study of the impact mechanisms of different marketization stages is also needed to provide a firm reference for guiding local targeted policies. Third, the microscopic research data used in this paper were limited by the research area and time, and data were only collected from a few typical counties in Hubei Province. In the future, we plan to expand the study area to other typical provinces, expand the sample selection, and conduct further demonstration and in-depth studies on a larger scale.

6. Conclusions and Suggestions

6.1. Conclusions

Based on 1017 farmland transfer research data, this paper analyzes the influence mechanism of social capital on farmland transfer decisions using the Probit model. Further, it investigates the moderating role of the marketization of farmland transfer in both influences. The following are the key research findings.

First, social networks, social trust, and social norms have dual positive effects on farmers’ farmland transfer decision making. Improving the social network, social trust, and social norms will not only promote the likelihood of farmers’ farmland transfer-in but also promote the likelihood of farmers’ farmland transfer-out.

Second, in the decision of farmland transfer-out, as the degree of marketization increases, the impact of social networks and social trust on the transfer-out of farms declines, and marketization plays a negative regulatory role. In contrast, the impact of social norms on farmland transfer-out increases with the increase of marketization degree, and marketization plays a positive regulatory role.

Third, in the decision making of farmland transfer-in, the influence of social networks and social trust on farmland transfer-in decreases with the increase of marketization degree, and marketization plays a negative moderating role. In contrast, the influence of social norms on farmland transfer-in has no significant change with the increase of marketization degree, and marketization has no moderating role.

6.2. Suggestions

First, improve the construction of the regional agricultural land transfer market. A standardized and orderly farmland transfer market can not only directly encourage farmers to take part in farmland transfer but can also enhance the impact of social norms. Therefore, it is required to improve the trading market’s administration, actively introduce outside money, technology and talents in rural areas, encourage all kinds of scale operators to centralize land transfer, and promote the marketization of farmland transfer objects. As the guide and supervisor of agricultural land transfer, the government should strengthen the propaganda and guidance of agricultural land transfer policy, help farmers correctly interpret the policy documents, and set up a supervisory group to oversee and arbitrate the process and disputes of agricultural land transfer to provide farmers with the necessary institutional environment and correct guidance.

Second, strengthen the construction of social norms and create a suitable environment for the transfer of farmland. Social norms have a positive effect on agricultural land transfer, especially on agricultural land transfer-out. The empirical findings indicate that this positive effect has been enhanced in the agricultural land transfer market development. On the one hand, there is a need to improve the agricultural skills of farmers, cultivate the main body of the agricultural land transfer market, and use the advantages of local people’s role to play a leading part in the demonstration. On the other hand, we should improve the leadership mechanism of village party organizations, motivate farmers to take an active role in the transfer of agricultural land, give the governing role of local collectives its due, and promote the orderly construction of the agricultural land transfer market.

Third, realize the reconstruction of trust and build the trust basis for operating the farmland transfer market. Social trust is effective in facilitating the transfer of agricultural land in the early stage of market development. However, its role is weakened with the improvement of the circulation market of agricultural land. As a result, it is necessary to strengthen the construction of market mechanisms and build the institutional trust of farmers. On the one hand, we should cultivate farmers’ contract consciousness, sign contracts relying on the normative system, and farmers should consciously abide by the contract regulations. On the other hand, we should improve the village-level management system, give due consideration to the intermediary function of village collectives in the circulation of agricultural land, and shape indirect trust.