Impact of Farmland Rental Contract Disputes on Farmland Rental Market Participation

Abstract

In the context of an ambiguous structure of farmland property rights, farmland rental contract disputes exhibited an upward tendency over the past decade, in practice, which damaged the expectations of land tenure security and, consequently, influenced the willingness of people to participate in farmland rental markets. Thus, the main aim of this paper was to analyze the impact of farmland rental contract disputes on farmland rental market participation, which was empirically tested by using a logit model, IV-probit model, and survey data for 762 apple growers. The results indicated that (1) farmland rental contract disputes significantly boosted the transfer of farmland to relatives and friends, (2) obtaining reasonable compensation in the process of solving farmland rental contract disputes significantly boosted farmland rental market participation for all types of objects of farmland transfers, and (3) farmland rental contract disputes significantly promoted farmers’ choice of relatives as the objects of farmland transfers. Participating in farmland rental markets was a freely pursued decision-making process, based on the actual situation, and farmers improved their expectations of the risks to farmland property rights by resorting to the reputation mechanism. However, obtaining reasonable compensation was conducive to reducing market segmentation and discrimination. Based on our analysis, the key factor in enhancing farmland rental market participation and reducing market segmentation was improving the stability of farmland property rights, and, in practice, farmland rental contract disputes should be reasonably reduced and regulated. This paper not only emphasizes the importance of solving farmland rental contract disputes reasonably, but also provides a reference for improving relevant laws and regulations on farmland transfers, which is conducive to developing farmland rental markets in the context of an ambiguous structure of farmland property rights in some developing countries.

1. Introduction

With the acceleration of urbanization and the development of farmland rental markets, the incidence of farmland rental contract disputes has increased every year and become the main cause of current social conflicts in rural areas [1,2]. Farmland disputes are disputes or contradictions that occur in the process of farmland management, farmland transfers, or farmland expropriation and, in rural China, mainly include farmland contract management disputes, farmland rental contract disputes, and farmland expropriation disputes [1,3]. In 2017, according to a statistical annual report on rural operation and management, China had a total of 104,000 farmland rental contract disputes, of which 78,000 occurred between practicing farmers (China Rural Magazine, Ministry of Agriculture and Rural Affairs, 2018). Notably, non-standard farmland rental systems or contracts, the substantial growth in agricultural income, and imperfect dispute settlement mechanisms were the main reasons for farmland rental contract disputes [4]. In the context of China’s Household Responsibility System (HRS), restricting the capacity of agricultural operators and the discriminatory design of laws and regulations resulted in a fuzzy structure of farmland property rights [5], which increased the incidence of farmland rental contract disputes and weakened the effectiveness of dispute settlement mechanisms.

Some agricultural operators obtained residual claimant rights and on-site control rights by participating in farmland rental markets and using fixed-rent contracts, but the ultimate control rights still belonged to village collectives—an arrangement that was not conducive to stabilizing the expectations of the risks to farmland property rights [5]. When farmers who participated in farmland rental markets had the ability to foresee all possible states and to stipulate their respective rights and responsibilities in advance, the farmland rental contracts were in the Pareto optimal condition [6,7]. Simultaneously, rational farmers who participated in farmland rental markets were encouraged to abide by farmland rental contracts and reduce the possibility of farmland rental contract disputes, which improved the efficiency of farmland rental markets.

However, in the process of economic transformation, the choice of farmland rental contracts was affected by imperfect systems of farmland property rights and by relevant laws and regulations. Moreover, the bounded rationality and opportunistic behaviors for farmers who participated in farmland rental markets resulted in incomplete farmland rental contracts [6,7]. The risks to farmland property rights were high and, in practice, the development of farmland rental markets was inhibited. A possible reason was that when the risk of non-agricultural unemployment was unpredictable or when agricultural income increased, farmers who rented out farmland might have been motivated to take back their rented land. However, when the prices of agricultural products were expected to fall, farmers who rented land might be motivated to abandon the rented land. Such situations reduced the stability of farmland rental contracts, which was not conducive to encouraging participation in the farmland rental markets. At the same time, if the relevant legal systems failed, it was usually difficult to obtain reasonable compensation that met necessary standards, which further restricted the development of farmland rental markets. In this paper, we attempt to analyze the influencing mechanisms of farmland rental contract disputes on farmland rental market participation.

Existing literature focused on the status quo and the characteristics, and settlement mechanisms of farmland rental contract disputes. However, the consideration of the influencing mechanisms of farmland rental contract disputes on farmland rental market participation needed to be improved and expanded. Compared with existing studies, the main contributions of this paper are as follows: (1) From a research perspective, the impact of farmland rental contract disputes on farmland rental market participation was analyzed—an analysis that could supplement and improve the existing literature on the relationship between the stability of farmland property rights and farmland transfers; (2) In terms of research content, the influencing mechanisms of farmland rental contract disputes on farmland rental market participation were studied, enriching the research conclusions of the previous literature. Specifically, when the practical risks to farmland property rights were high, farmers participated in farmland rental markets by resorting to the reputation mechanism. In addition, relatives were usually selected as the objects of farmland transfers, which meant that participating in farmland rental markets was a freely pursued decision-making process, based on the actual situation. Meanwhile, obtaining reasonable compensation in the process of solving farmland rental contract disputes significantly enhanced farmland rental market participation and was conducive to reducing market segmentation and market discrimination, but that impact was not significant for the objects of farmland transfers. Moreover, the impact on farmland transfers of the interaction between farmland rental contract disputes and obtaining reasonable compensation was also analyzed.

2. Literature Review

The risks to farmland property rights consist mainly of the degree of clarity in the systems of farmland property rights and the stability of farmland rental contracts [6]. In theory, a clear structure of farmland property rights helps to reduce the enforcement costs of farmland property rights, and stable farmland rental contracts lower the high risks to farmland property rights in the context of an ambiguous structure of farmland property rights [6,8]. The main reason is that a clear structure of farmland property rights is conducive to improving the effectiveness of relevant laws and policy documents and reducing the possibility of opportunistic behaviors by farmers who participate in farmland rental markets, thereby decreasing the transaction costs in signing stable farmland rental contracts. Tian et al. found that frequent farmland adjustments increased the expectations of the risks to farmland property rights and were not conducive to signing written contracts [9]. However, in the context of an ambiguous structure of farmland property rights, if farmland rental contracts were complete, both parties who participated in farmland rental markets presented their respective rights and responsibilities beforehand, which reduced opportunistic behaviors afterward and improved the expectations of the risks to farmland property rights. Kousar and Abdulai indicated that stable farmland rental contracts were conducive to improving the expectations of the risks to farmland property rights, which could motivate long-term agricultural investment [10].

In practice, the complexity and uncertainty of the external environment, the bounded rationality, and opportunistic behaviors had caused incomplete farmland rental contracts [6,7], and incomplete farmland rental contracts were not conducive to motivating both parties who participated in farmland rental markets to jointly abide by farmland rental contracts, which resulted in high risks to farmland property rights and restricted the development of farmland rental markets. Theoretically, the optimal farmland rental contracts relied on a trade-off between the expected protection of rigid rights and the improvement of ex-post efficiency [7], which meant that the stability of farmland rental contracts depended on the negotiation ability and the commitment ability for farmers who participated in farmland rental markets and the degree of difficulty in performance assessment [5]. Hence, the clauses for compensation for violating farmland rental contracts could reduce opportunistic behaviors and the risks of default. Zou et al. demonstrated that high forecast costs and high supervising costs led to asymmetric information for both parties who participated in farmland rental markets, and the penalty clauses for breaching farmland rental contracts were conducive to decreasing the possibility of default and stabilizing the expectations of the risks to farmland property rights [11].

And the influencing mechanisms of the risks to farmland property rights on farmland transfers were as follows: Firstly, low risks to farmland property rights motivated farmers to engage in agricultural investment and increased the intrinsic value of farmland, which promoted the willingness to participate in farmland rental markets; secondly, low risks to farmland property rights reduced the transaction costs of participating in farmland rental markets, which promoted the development of farmland rental markets [12,13]. Yet the two-tier agricultural management systems resulted in an ambiguous structure of farmland property rights and damaged the expectations of the stability of farmland property rights, which hindered the development of farmland rental markets [14,15]. Cheng et al. analyzed the impact of the stability of farmland property rights on farmland transfers in rural China and found that stable farmland property rights were conducive to not only increasing the intrinsic value of farmland but also decreasing the transaction costs of participating in farmland rental markets, which enhanced farmland rental market participation [13].

Notably, in the context of an ambiguous structure of farmland property rights and incomplete farmland rental contracts, reducing the risks to farmland property rights could resort to informal rules such as customs and habits [16]. A plausible explanation was that, due to associated games and repeated games, the widespread existence of the reputation mechanism improved the expectations of the stability of farmland property rights. Existing studies demonstrated that high risks to farmland property rights caused the scope of farmland transfers to be restricted within relatives and friends, which led to the segmentation of farmland rental markets [17]. At this time, farmers used the reputation mechanism to select the suitable objects of farmland transfers within the acquaintance society. Due to the fact that, in most cases, a formal rental contract with high transaction costs was not necessary to ensure the smooth operation of farmland rental markets [18].

In addition, thin farmland rental markets led most farmers to lack experience in directly participating in farmland rental markets, but within the acquaintance society, farmers who intended to participate in farmland rental markets could refer to the behaviors of other farmers who participated in farmland rental markets. Chen et al. analyzed the relationship between social capital and farmland transfers and indicated that social capital enhanced significantly farmland transfers by reducing the transaction costs of farmland transfers [19]. In other words, when the transaction costs of participating in farmland rental markets were high, the willingness to participate in farmland rental markets could be embedded in rural social networks, and whether farmers participated in farmland rental markets or not depended on the behaviors of participating in farmland rental markets for relatives, friends or other farmers. The empirical results showed that the behaviors of participating in farmland rental markets for relatives and friends, which represented a “strong relationship”, and the behaviors of participating in farmland rental markets for other farmers, which represented a “weak relationship”, significantly promoted the willingness to participate in farmland rental markets. More importantly, when the costs of acquiring and using the policies of farmland transfers were relatively high, the behaviors of participating in farmland rental markets for relatives, friends, or other farmers could reduce the transaction costs of market participation, which resulted in the herd effect in the process of farmland transfers [20].

In sum, previous studies mainly discussed the influencing mechanisms of the risks to farmland property rights on farmland transfers from the perspectives of the systems of farmland property rights and the stability of farmland rental contracts, which provided the theoretical basis and empirical reference for our study. More importantly, in the context of an ambiguous structure of farmland property rights, the incidence of farmland rental contract disputes was continually increased, but the impact of farmland rental contract disputes on farmland rental market participation was commonly ignored by previous literature.

3. Theoretical Framework

3.1. Impact of the Risks to Farmland Property Rights on Farmland Transfers

Following the theoretical model of Deininger et al. [21], a representative farmer (D) used their own labor endowment (L) and rented land (q) to engage in agricultural production, and his household income came from agricultural production and non-agricultural employment. The production function for the representative farmer (D) conformed to the standard form of the production function, that is, $\mathrm{f}\left({\mathrm{l}}^a,q\left(1-\mathrm{v}\right)\right)$, and satisfied the following conditions: ${\mathrm{f}}_{{\mathrm{l}}^a}>0, {\mathrm{f}}_q>0, {\mathrm{f}}_{{\mathrm{l}}^a{\mathrm{l}}^a}<0$ and ${\mathrm{f}}_{qq}<0$. Thus, the utility function for the representative farmer (D) could be written as follows:

$$\mathrm{u}\left({\mathrm{l}}^a,q\right)=\mathrm{Pf}\left({\mathrm{l}}^a,q\left(1-\mathrm{v}\right)\right)+{\mathrm{l}}^o\mathrm{w}-\mathrm{rq}$$

(1)

where P denoted the price of agricultural products, which was standardized as 1 for simplicity. Additionally, v denoted the probability of losing rented land for farmers who rented in farmland, which satisfied $\mathrm{v}\in \left[0,1\right]$. In the context of the Household Responsibility System (HRS), farmland rental contract disputes referred to these behaviors that when farmers who rented out farmland lost their off-farm jobs and returned home or when revenues on agricultural production became high, rented land might be taken back in advance, and land rents were increased by farmers who rented out farmland. Moreover, r denoted land rents, and w denoted the wage of off-farm employment, while ${\mathrm{l}}^a$ denoted the number of agricultural laborers, and ${\mathrm{l}}^o$ denoted the number of non-agricultural laborers.

The main aim of participating in farmland rental markets for the representative farmer (D) was to achieve the maximum utility. Thus, the first-order condition could be written as follows:

$$\left(1-\mathrm{v}\right){\mathrm{f}}_q\left({\mathrm{l}}^a,q\left(1-\mathrm{v}\right)\right)=\mathrm{r}$$

(2)

where Equation (2) was that when the probability of losing rented land was considered, the marginal production value of land was equal to land rents, which reflected the equilibrium condition of farmland rental markets. According to the Equation (2) (Assuming that $\mathrm{F}=\left(1-\mathrm{v}\right){\mathrm{f}}_q\left({\mathrm{l}}^a,q\left(1-\mathrm{v}\right)\right)-\mathrm{r},$ then $\frac{\partial F}{\partial \mathrm{v}}=-{\mathrm{f}}_q-q\left(1-\mathrm{v}\right){\mathrm{f}}_{\mathit{qq}}, \frac{\partial F}{\partial q}={\left(1-\mathrm{v}\right)}^2{\mathrm{f}}_{\mathit{qq}},$$\frac{\partial F}{\partial r}=-1$. If $1+\frac{{\mathrm{f}}_q}{q•{\mathrm{f}}_{\mathit{qq}}}<\mathrm{v}<$1, then $\frac{\partial q}{\partial \mathrm{v}}<0;$ If $1+\frac{f_q}{q•f_{qq}}>\mathrm{v}$, then $\frac{\partial q}{\partial \mathrm{v}}>0$), the main conclusion was as follows: when the risks to farmland property rights were relatively high, a lower probability of farmland rental contract disputes motivated farmers to participate in farmland rental markets. However, when the risks to farmland property rights were relatively low, a higher probability of farmland rental contract disputes motivated farmers to participate in farmland rental markets.

Theoretically, the risks to farmland property rights consisted of the degree of clarity in the systems of farmland property rights and the stability of farmland rental contracts [6]. Specifically, perfect systems of farmland property rights reduced the execution costs of farmland rental contracts, and stable farmland rental contracts weakened the negative impact of an ambiguous structure of farmland property rights. From the perspective of the stability of farmland rental contracts, the execution efficiency of farmland rental contracts determined the risks to farmland property rights, and the execution efficiency of farmland rental contracts depended on the effectiveness of the governance mechanism of farmland rental contracts [22,23]. Hence, the effectiveness of the governance mechanism of farmland rental contracts determined the risks to farmland property rights. In the process of economic transformation, the governance mechanism of farmland rental contracts mainly included private execution mechanisms and public execution mechanisms. Specifically, the essence of the private execution mechanism lay in the reputation mechanism that resulted from repeated games and related games, while the essence of the public execution mechanism lay in the effectiveness of legal systems [22,24]. Within the society of acquaintances, the transaction order of farmland rental markets was usually established on the basis of blood, kinship, and geographic relationships, which led to repeated games and associated games among the objects of farmland transfers. Notably, in the context of repeated games and associated games, the reputation mechanism was conducive to increasing the revenues on executing farmland rental contracts and improving the implementation efficiency of farmland rental contracts, which reduced the expectations of the risks to farmland property rights [22,23]. Meanwhile, in the context of valid legal systems, public enforcement mechanisms such as courts effectively increased the costs of violating farmland rental contracts, which improved the expectations of the risks to farmland property rights.

3.2. Impacts of Farmland Rental Contract Disputes on Farmland Rental Market Participation

Farmland rental contract disputes were defined as these disputes resulting from violating farmland rental contracts in the process of farmland transfers. Theoretically, according to the differences in the objects of violating farmland rental contracts, farmland rental contract disputes could be divided into two types, that is, these disputes that occurred when these farmers themselves participated in farmland rental markets and these disputes that occurred when other farmers participated in farmland rental markets. This paper focused on analyzing the impact of farmland rental contract disputes that occurred when other farmers participated in farmland rental markets on farmland transfers. A plausible reason was that due to the high transaction costs of participating in farmland rental markets, thin farmland rental markets led most farmers to lack experience in directly participating in farmland transfers in rural China. Thus, some farmers could have beliefs about farmland transfers from information regarding participation in farmland rental markets by their relatives, friends, or other farmers, which were used as references for their own participation in farmland rental markets. When the costs of acquiring and using the information on farmland transfers were high, learning from the behaviors of farmland transfers for relatives, friends or other farmers was conducive to reducing the transaction costs of participating in farmland rental markets, which resulted in the herd effect of farmland transfers [20].

Meanwhile, if legal systems were effective in practice, information about obtaining reasonable compensation was conducive to increasing their confidences in the effectiveness of legal systems and improving the expectations of jointly complying with farmland rental contracts, which reduced the risks to farmland property rights. On the contrary, in the context of noneffective legal systems, farmland rental contract disputes strengthened the expectations of the risks to farmland property rights. Compared with farmland rental contract disputes that occurred when other farmers participated in farmland rental markets, there might be a lag impact of farmland rental contract disputes that occurred when these farmers themselves participated in farmland rental markets on farmland transfers. A plausible explanation was that when these farmers who participated in farmland rental markets rented in or out farmland again, the impact of farmland rental contract disputes on farmland transfers might be significant, but, in the context of thin farmland rental markets, the impact of farmland rental contract disputes on farmland transfers was difficult to be captured for these farmers. Thus, when the herding effect of farmland transfers was considered, the impact of farmland rental contract disputes that occurred when other farmers participated in farmland rental markets on farmland transfers was analyzed in this paper, which was considered to be reasonable and appropriate.

Further, the mathematical logic of the relationship between farmland rental contract disputes and the risks to farmland property rights was given. Assuming that the relationship between farmland rental contract disputes and the risks to farmland property rights was as follows: $\mathrm{v}\left(\mathrm{x},\mathrm{s}\right)=\prod \left(\varnothing -\Psi \left(\mathrm{s}\right)-\mathrm{x}\left(\mathrm{s}\right)\right)$. $\prod \left(•\right)$ denoted the conversion function of risk probability, which was used to ensure the following relationship, that is, $\mathrm{v}\in \left[0,1\right]$. $\varnothing$ denoted the coefficient of the risks to farmland property rights, which reflected the initial status of the risks to farmland property rights. $\Psi \left(•\right)$ denoted the network value of the risks to farmland property rights, which reflected the impact of farmland transfers for other farmers on the expectations of the risks to farmland property rights for the representative farmer (D). Specifically, $\Psi \left(•\right)>0$ denoted that other farmers’ behaviors of farmland transfers reduced the expectations of the risks to farmland property rights, and $\Psi \left(•\right)<0$ denoted that other farmers’ behaviors of farmland transfers strengthened the expectations of the risks to farmland property rights. Additionally, s denoted that when other farmers participated in farmland rental markets, farmland rental contract disputes occurred within the administrative village, and at this time, the network value was negative, that is, $\Psi \left(\mathrm{s}\right)<0$. Thus, farmland rental contract disputes that occurred when other farmers participated in farmland rental markets denoted the high risks to farmland property rights, that is, $\frac{\partial \mathrm{v}}{\partial \Psi }•\frac{\partial \Psi }{\partial \mathrm{s}}>0$.

Notably, x denoted formal rules such as laws and regulations, or informal rules such as customs and habits. Existing studies indicated that the risks to farmland property rights depended on the effectiveness of formal rules and informal rules [6,16,25]. Specifically, farmland rental contract disputes increased the expectations of the risks to farmland property rights, but participation in farmland rental markets reduced the expectations of the risks to farmland property rights by using the reputation mechanism that resulted from repeated games and associated games within the acquaintance society or using the laws and regulations to obtain effective compensation for rental contract disputes.

In sum, in the context of an ambiguous structure of farmland property rights, farmland rental contract disputes denoted the high risks to farmland property rights, but participating in farmland rental markets was a freely pursued decision-making process, based on the actual situation: farmers improved the expectations of the risks to farmland property rights by resorting to the reputation mechanism within the acquaintance society, which boosted farmland rental market participation. Meanwhile, in the context of effective legal systems, obtaining reasonable compensation for farmland rental contract disputes increased the costs of violating farmland rental contracts and improved the expectations of the risks to farmland property rights, which enhanced farmland rental market participation. Thus, these two hypotheses were proposed as follows:

H1: 

In the context of associated games and repeated games, farmland rental contract disputes boosted farmland rental market participation between relatives and friends.

H2: 

Obtaining reasonable compensation for farmland rental contract disputes boosted farmland rental market participation for all types of the objectives of farmland transfers.

4. Data Source and Statistical Evidence

4.1. Data Source

The data set used in this paper was derived from an empirical investigation in apple production areas. To ensure the representativeness of sample farmers, according to the multistage stratified random sampling method, the number of sample counties was determined by using the following equation [26]:

$$\mathrm{n}={\left(\frac{u_{\alpha }m}{1-p_c}\right)}^2$$

(3)

where the reliability was set to be as follows: $\alpha =0.05$, so ${\mu }_{\alpha }=1.96$. $m$ denoted the coefficient of variation, and 0.25 was used here. $p_c$ denoted the accuracy of sample estimation, and 0.80 was used here. Finally, according to the Probability Proportional to Scale (PPS) method, 6 high-apple-producing counties (Fengxiang, Fuxian, and Baota in Shaanxi Province and Yiyuan, Yishui, and Muping in Shandong Province) were randomly selected in the surrounding Bohai Bay areas and the Loess Plateau areas.

Then, the equation for determining the number of sample farmers was written as follows:

$$N=\frac{z_{\alpha }p(1-p)}{e^2}$$

(4)

where the confidence was set to be as follows: $\alpha =0.01$, so $z_{\alpha }=2.58$. $p$ denoted the probability of occurrence, and $p$ was set to be 0.5. $e$ denoted sampling error, which was controlled within 3%. Finally, the household-level sample size was at least 717, that is, N ≥ 717. The specific sampling method of sample farmers was as follows: 6~9 sample villages were randomly selected from each sample county, and almost 20 sample farmers were randomly selected from each sample village. This investigation was conducted by the means of face-to-face interviews, and 762 effective sample farmers were obtained in this survey.

4.2. Variable Selection and Descriptive Statistics

Table 1. Descriptive statistics.

Variable Names	Indicators	Obs.	Mean	S.D.
Dependent variables
Farmland rental market participation	Whether or not farmers participated in farmland transfers (yes = 1, no = 0)	762	0.610	0.488
The objects of farmland transfers	Relatives = 1, friends = 2, ordinary farmers = 3, village collectives = 4	465	2.994	1.155
Key independent variables
Farmland rental contract disputes	Whether or not they knew about farmland rental contract disputes for other farmers within the administrative villages in the past five years (yes = 1, no = 0)	762	0.161	0.368
	Whether or not they knew about obtaining reasonable compensation for farmland rental contract disputes (yes = 1, no = 0)	762	0.033	0.178
Control variables
Household head characteristics	Age (year)	762	52.202	10.020
	Edu (year)	762	7.280	3.323
	Dummy variable for village cadres (yes = 1, no = 0)	762	0.127	0.334
Household characteristics	The ratio of agricultural laborers to household size (%)	762	0.570	0.234
	The distance from home to the nearest national highway (km)	762	6.863	9.674
Social network	The ratio of the expenditures for favors and gifts to total expenditures (%)	762	0.159	0.112
	The number of village cadres who frequently communicated	762	3.136	6.348
Location characteristics	Dummy variable for Shandong Province (yes = 1, no = 0)	762	0.467	0.499
	Dummy variable for Shaanxi Province (yes = 1, no = 0)	762	0.532	0.499

gave descriptive statistics of dependent variables, key independent variables, and control variables. The results indicated that the number of sample farmers who knew about farmland rental contract disputes that occurred when other farmers participated in farmland rental markets was 123, which accounted for 16.1%. Furthermore, the average number of farmland rental contract disputes within the administrative village was 0.5. Additionally, the main reason for farmland rental contract disputes was that farmers who rented out farmland violated farmland rental contracts. Specifically, owing to the improvement in agricultural income, farmers who rented out farmland received rented land back in advance, which accounted for 33.3%. Then, when farmers who rented out farmland and engaged in non-agricultural employment lost their jobs, they received rented land back in advance, which accounted for 22.8%. Additionally, the number of farmers who rented out farmland and worried about losing rented land in the future accounted for 8.1%. Negotiation between both parties who participated in farmland rental markets was the main approach to solving farmland rental contract disputes, which accounted for 65.0%, and only a small number of farmers resorted to laws in the process of solving farmland rental contract disputes, which accounted for 5.69%. Additionally, when farmland rental contract disputes occurred, only 19.5% of farmers who suffered financial losses obtained reasonable compensation. Moreover, the number of farmland transfers in which the objects of land transaction were relatives and friends amounted to 127, and the number of farmland transfers in which the objects of land transaction were ordinary farmers and village collectives amounted to 338.

This study used age, years of education, and the identity of village cadres to denote the quality of human capital of the household head, and the household head was the main decision-maker of agricultural production for a farm household. For household characteristics, the variable for the ratio of agricultural laborers to household size was used to denote the number of human capital, which reflected the willingness of farmers to participate in farmland rental markets. The variable for the distance from home to the nearest national highway was used to denote the convenience degree of transportation. For social networks, the two variables for the ratio of the expenditures for favors and gifts to total expenditures and the number of village cadres who frequently communicated were used to denote the size and quality of social capital.

Table 2. The differences in farmland transfers and household characteristics.

Variable Names	With Farmland Rental Contract Disputes	Without Farmland Rental Contract Disputes	Diff	With Reasonable Compensation	Without Reasonable Compensation	Diff
	(1)	(2)	(3) = (1) − (2)	(4)	(5)	(6) = (4) − (5)
Market participation	0.707 (0.041)	0.592 (0.019)	0.115 **	0.960 (0.040)	0.598 (0.018)	0.362 ***
The objects of farmland transfers	2.655 (0.133)	3.071 (0.058)	−0.416 ***	2.708 (0.237)	3.009 (0.055)	−0.301
Age	49.943 (0.987)	52.715 (0.379)	−2.772 ***	46.520 (1.702)	52.463 (0.363)	−5.943 ***
Edu	7.699 (0.287)	7.199 (0.132)	0.500	9.240 (0.521)	7.213 (0.123)	2.027 **
Dummy variable for village cadres	0.114 (0.029)	0.130 (0.013)	−0.016	0.120 (0.066)	0.128 (0.012)	−0.008
The ratio of agricultural laborers to household size	0.542 (0.018)	0.575 (0.009)	−0.033	0.506 (0.033)	0.572 (0.009)	−0.066
The distance from home to the nearest national highway	6.157 (0.748)	6.999 (0.392)	−0.842	6.624 (2.153)	6.871 (0.355)	−0.247
The ratio of the expenditures for favors and gifts to total expenditures	0.156 (0.009)	0.160 (0.005)	−0.004	0.165 (0.022)	0.159 (0.004)	0.006
The number of village cadres who frequently communicated	3.106 (0.849)	3.142 (0.221)	−0.036	6.240 (3.925)	3.031 (0.198)	3.209 **

Notes: Columns (1), (2), (4), and (5) gave the mean values, and numbers in parentheses denoted standard deviation. Column (3) gave the differences in farmland transfers and household characteristics between farmers who knew about farmland rental contract disputes and farmers who did not know about farmland rental contract disputes, and column (6) gave the differences in farmland transfers and household characteristics between farmers who knew about obtaining reasonable compensation and farmers who did not know about obtaining reasonable compensation. ***, ** denoted statistical significance at the 1% and 5% levels, respectively.

gave the differences in farmland transfers and household characteristics between farmers who knew about farmland rental contract disputes and farmers who did not know about farmland rental contract disputes or between farmers who knew about obtaining reasonable compensation and farmers who did not know about obtaining reasonable compensation. The results showed that the differences in farmland rental market participation and the objects of farmland transfers between farmers who knew about farmland rental contract disputes and farmers who did not know about farmland rental contract disputes were significant. Similarly, the differences in farmland rental market participation between farmers who knew about obtaining reasonable compensation and farmers who did not know about obtaining reasonable compensation were significant. However, the above-mentioned results did not control the impacts of other variables. Meanwhile, apart from age of the household head, the differences in household head characteristics, household characteristics, and social networks between farmers who knew about farmland rental contract disputes and farmers who did not know about farmland rental contract disputes were not significant. Additionally, apart from age, the year of education, and the number of village cadres who frequently communicated, the differences in other control variables between farmers who knew about obtaining reasonable compensation and farmers who did not know about obtaining reasonable compensation were not significant. Hence, the selection issue might be trivial in this paper.

5. Empirical Results

5.1. Impact of Farmland Rental Contract Disputes on Farmland Transfers

In the context of the different risk levels of farmland property rights, the impact of farmland rental contract disputes on farmland rental market participation was heterogeneous. Notably, the reputation mechanism that resulted from associated games and repeated games within the acquaintance society improved the execution efficiency of farmland rental contracts and weakened the expectations of the risks to farmland property rights. Thus, according to whether or not the objects of farmland transfers were relatives and friends, the impact of farmland rental contract disputes on farmland rental market participation was analyzed here.

Table 3. Estimation results of the impacts of farmland rental contract disputes on farmland transfers.

Variable Names	Dummy Variable for Participating in Farmland Transfers (Logit Model)
	Model 1 (Relatives and Friends)	Model 2 (Other Farmers)
Farmland rental contract disputes	0.805 *** (0.276)	0.262 (0.236)
Age	−0.053 *** (0.013)	−0.051 *** (0.010)
Edu	0.083 ** (0.035)	0.027 (0.026)
Dummy variable for village cadres	0.075 (0.358)	0.047 (0.251)
The ratio of agricultural laborers to household size	0.529 (0.504)	0.216 (0.370)
The distance from home to the nearest highway	−0.001 * (0.011)	−0.023 ** (0.009)
The ratio of the expenditures for favors and gifts to total expenditures	−0.415 (1.246)	1.542 * (0.800)
The number of village cadres who frequently communicated	0.055 (0.037)	0.023 (0.018)
Dummy variable for Shandong Province	−0.850 *** (0.257)	−0.176 (0.183)
Constant	1.162 (0.759)	2.414 *** (0.602)
Pseudo R-squared	0.117	0.058
Observations	424	635

Notes: Numbers in parentheses denoted robust standard deviation. ***, **, * denoted statistical significance at the 1%, 5%, and 10% levels, respectively.

gave estimation results of the impact of farmland rental contract disputes on farmland rental market participation. Specifically, model 1 and model 2 showed the impact of farmland rental contract disputes on farmland rental market participation for farmland transfers occurred between relatives and friends and farmland transfers occurred between non-relatives and non-friends, respectively.

The result of model 1 indicated that farmland rental contract disputes significantly enhanced the transfer of farmland to relatives and friends at the 1% level, which was consistent with H1. However, the result of model 2 showed that farmland rental contract disputes insignificantly boosted the transfer of farmland to non-relatives and non-friends. A plausible explanation was that participating in farmland rental markets was a freely pursued decision-making process according to the actual situation, and farmers who participated in farmland rental markets improved the expectations of the risks to farmland property rights by resorting to the reputation mechanism that resulted from associated games and repeated games within the acquaintance society. Existing studies indicated that when farmers participated in farmland rental markets, farmers could select suitable trading partners by using the reputation mechanism, which meant that informal rules were used to ensure the smooth operation of farmland rental markets [18].

5.2. Impact of Obtaining Reasonable Compensation on Farmland Transfers

In the context of effective legal systems, obtaining reasonable compensation for farmland rental contract disputes weakened the expectations of the risks to farmland property rights, but if the legal systems were noneffective, low compensation for farmland rental contract disputes was not conducive to improving the expectations of the risks to farmland property rights.

Table 4. Estimation results of the impact of obtaining reasonable compensation on farmland transfers.

Variable Names	Dummy Variable for Participating in Farmland Transfers (Logit Model)
	Model 3	Model 4 (Relatives and Friends)	Model 5 (Other Farmers)
Obtaining reasonable compensation	2.362 ** (1.032)	2.066 ** (1.026)	2.452 ** (1.046)
Age	−0.051 *** (0.010)	−0.051 *** (0.013)	−0.050 *** (0.010)
Edu	0.033 * (0.025)	0.080 ** (0.035)	0.021 (0.026)
Dummy variable for village cadres	0.061 (0.242)	0.014 (0.355)	0.052 (0.255)
The ratio of agricultural laborers to household size	0.253 (0.357)	0.404 (0.498)	0.222 (0.371)
The distance from home to the nearest highway	−0.019 ** (0.008)	−0.005 (0.011)	−0.024 ** (0.010)
The ratio of the expenditures for favors and gifts to total expenditures	1.007 (0.737)	−0.392 (1.212)	1.507 * (0.806)
The number of village cadres who frequently communicated	0.026 (0.019)	0.040 (0.038)	0.025 (0.019)
Dummy variable for Shandong Province	−0.316 * (0.170)	−0.849 *** (0.257)	−0.153 (0.183)
Constant	2.787 *** (0.573)	1.337 * (0.761)	2.368 *** (0.600)
Pseudo R-squared	0.074	0.113	0.069
Observations	762	424	635

Notes: Numbers in parentheses denoted robust standard deviation. ***, **, * denoted statistical significance at the 1%, 5%, and 10% levels, respectively.

gave estimation results of the impact of obtaining reasonable compensation on farmland rental market participation. Specifically, model 3 gave the impact of obtaining reasonable compensation on farmland rental market participation. Model 4 and model 5 showed the impacts of obtaining reasonable compensation on farmland rental market participation for the transfer of farmland to relatives and friends and the transfer of farmland to non-relatives and non-friends, respectively.

The result of model 3 indicated that obtaining reasonable compensation for farmland rental contract disputes boosted significantly farmland rental market participation at the 5% level, which was consistent with H2. In other words, obtaining reasonable compensation increased the costs of violating farmland rental contracts and consequently, reduced the expectations of the risks to farmland property rights. Similarly, the results of model 4 and model 5 displayed that the impacts of obtaining reasonable compensation on farmland rental market participation were significant at the 5% level for the transfer of farmland to relatives and friends and the transfer of farmland to non-relatives and non-friends, which means that the differences in the impacts of obtaining reasonable compensation on farmland rental market participation for farmland transfers occurred between relatives and friends and farmland transfers occurred between non-relatives and non-friends were trivial. Thus, obtaining reasonable compensation was conducive to reducing market segmentation and discrimination.

5.3. Robustness Test

The impact of farmland rental contract disputes on farmland transfers might omit some variables such as personal ability, and farmland transfers might affect the incidence of farmland rental contract disputes, in turn, which resulted in endogenous issues. Thus, the two variables for whether or not laws and regulations related to farmland transfers were often publicized within this village, and the average distance from the village to the police station or township government were used as instrumental variables in this paper because the variable for whether or not laws and regulations related to farmland transfers were often publicized within this village was conducive to popularizing the relevant contents of the reforms of farmland property rights systems and raising legal awareness for farmers, which reduced the incidence of farmland rental contract disputes. Similarly, the closer average distance from the village to the police station or township government was conducive to decreasing the costs of obtaining relevant information about the reforms of farmland property rights systems and employing third-party enforcement mechanisms to solve farmland rental contract disputes, which improved the expectations of the risks to farmland property rights. However, the above-mentioned variables mainly involved the characteristics of the villages, and it was usually difficult for a single small farmer to change these characteristics of the villages, which could ensure the exogeneity of instrumental variables. Additionally, the IV-Probit model used here did not have the issue of weak instrumental variables.

Table 5. Estimation results of instrumental variables.

Variable Names	Dummy Variable for Participating in Farmland Transfers (IV-Probit model)		Dummy Variable for Participating in Farmland Transfers (IV-Probit Model)
	Relatives and Friends	Other Farmers
Disputes	2.790 *** (0.102)	1.140 (1.237)
Obtaining compensation			5.194 *** (0.963)
Control variables	Yes	Yes	Yes
Observations	424	635	762

Notes: Numbers in parentheses denoted robust standard deviation. *** denoted statistical significance at the 1%, 5%, and 10% levels, respectively. “Yes” denoted that these variables had been controlled here.

presented the estimation results of instrumental variables. The results displayed that farmland rental contract disputes were conducive to significantly enhancing the transfer of farmland to relatives and friends. Additionally, obtaining reasonable compensation significantly enhanced farmland rental market participation. Thus, the above-mentioned results indicated that research conclusions were robust in this paper.

6. Extension

6.1. Impact of Farmland Rental Contract Disputes on the Objects of Farmland Transfers

To further analyze the impacts of farmland rental contract disputes and obtaining reasonable compensation on farmland transfers, the mlogit model was used to study the impacts of farmland rental contract disputes and obtaining reasonable compensation on the choice of the objects of farmland transfers.

Table 6. Estimation results of mlogit model.

Variable Names	Relatives	Friends	Ordinary Farmers	Relatives	Friends	Ordinary Farmers
Disputes	0.695 ** (0.322)	0.238 (0.453)	0.013 (0.338)
Obtaining compensation				0.331 (0.571)	0.179 (0.850)	0.488 (0.578)
Age	−0.009 (0.015)	−0.051 *** (0.021)	−0.059 *** (0.015)	−0.013 (0.015)	−0.053 ** (0.022)	−0.059 *** (0.015)
Edu	−0.002 (0.042)	0.014 (0.054)	−0.086 ** (0.044)	−0.001 (0.042)	0.014 (0.055)	−0.092 ** (0.044)
Dummy variable for village cadres	−0.249 (0.414)	−0.024 (0.626)	−0.158 (0.399)	−0.223 (0.412)	−0.005 (0.627)	−0.134 (0.395)
The ratio of agricultural laborers to household size	−0.111 (0.617)	−0.380 (0.914)	−0.477 (0.668)	−0.112 (0.610)	−0.376 (0.922)	−0.469 (0.669)
The distance from home to the nearest highway	0.026 * (0.015)	0.014 (0.017)	0.006 (0.016)	0.027 * (0.015)	0.014 (0.017)	0.006 (0.016)
The ratio of the expenditures for favors and gifts to total expenditures	−2.994 * (1.672)	−1.018 (2.072)	−0.739 (1.116)	−2.959 * (1.664)	−0.994 (2.069)	−0.758 (1.119)
The number of village cadres who frequently communicated	0.029 * (0.018)	−0.013 (0.047)	0.031 * (0.019)	0.028 (0.018)	−0.015 (0.048)	0.029 (0.019)
Dummy variable for Shandong Province	−1.101 *** (0.295)	−0.794 * (0.453)	−1.161 (0.281)	−1.159 *** (0.294)	−0.807 * (0.444)	−1.135 *** (0.281)
Constant	0.251 (0.971)	1.279 (1.266)	3.736 *** (0.919)	0.622 (0.935)	1.379 (1.269)	3.747 *** (0.908)
Pseudo R-squared	0.073			0.069
Observations	465			465

Notes: Numbers in parentheses denoted robust standard deviation. ***, **, * denoted statistical significance at the 1%, 5%, and 10% levels, respectively. Village collectives were regarded as a base category of the objects of farmland transfers.

displayed the estimation results of mlogit model. The results displayed that farmland rental contract disputes significantly motivated farmers to choose relatives as the objects of farmland transfers because participating in farmland rental markets was a freely pursued decision-making process according to the actual situation. Additionally, in the context of the high risks to farmland property rights, the expectations of the risks to farmland property rights were weakened by resorting to the reputation mechanism, which means that farmland rental contract disputes affected the choice of the objects of farmland transfers. However, obtaining reasonable compensation could not facilitate significantly the transfer of farmland to relatives and friends. A plausible explanation was that in the context of effective legal systems, farmers needed not to rely on informal rules to weaken the expectations of the risks to farmland property rights, which reduced market segmentation of farmland transfers.

Considering that the mlogit model must be satisfied with IIA assumption,

Table 7. Estimation results of Small-Hsiao test.

Variable Names	lnL (Full)	lnL (Omit)	Chi-Square	df	p > Chi-Square	Evidence
Relatives	−164.691	−153.870	21.642	20	0.360	Not reject H0
Friends	−209.291	−195.713	27.156	20	0.131	Not reject H0
Ordinary farmers	−141.125	−130.768	20.713	20	0.414	Not reject H0
Village collectives	−126.430	−114.251	24.357	20	0.227	Not reject H0

presented the estimation results of the Small-Hsiao test. The results indicated that the mlogit model did not reject the null hypothesis that the alternatives were independent of each other. Thus, the mlogit model used in this paper was appropriate.

6.2. Impacts of Obtaining Reasonable Compensation and Farmland Rental Contract Disputes on Farmland Transfers

To further analyze the impact of the interactive item between farmland rental contract disputes and obtaining reasonable compensation on farmland transfers, sample farmers were divided into three types, that is, farmers without farmland rental contract disputes, farmers with farmland rental contract disputes who were not obtaining reasonable compensation, and farmers with obtaining reasonable compensation for farmland rental contract disputes.

Table 8. Estimation results of the impact of the interactive item on farmland transfers.

Variable Names	Dummy Variable for Participating in Farmland Transfers (Logit Model)	The Objects of Farmland Transfers (Mlogit Model)
		Relatives	Friends	Ordinary Farmers
Disputes	0.127 (0.228)	0.769 ** (0.362)	0.244 (0.508)	−0.192 (0.390)
Obtaining compensation	2.381 ** (1.033)	0.478 (0.576)	0.220 (0.852)	0.454 (0.583)
Household head characteristics	Yes	Yes	Yes	Yes
Household characteristics	Yes	Yes	Yes	Yes
Social networks	Yes	Yes	Yes	Yes
Location characteristics	Yes	Yes	Yes	Yes
Pseudo R-squared	0.074	0.075
Observations	762	465

Notes: Numbers in parentheses denoted robust standard deviation. ** denoted statistical significance at the 1%, 5%, and 10% levels, respectively. “Yes” denoted that these variables had been controlled here.

gave the estimation results of the impact of the interactive item between farmland rental contract disputes and obtaining reasonable compensation on farmland transfers. The results indicated that compared with these farmers without farmland rental contract disputes, these farmers with farmland rental contract disputes who were not obtaining reasonable compensation were not significantly motivated to participate in farmland rental markets, but the variable for obtaining reasonable compensation for farmland rental contract disputes enhanced significantly farmland rental market participation.

And compared with these farmers without farmland rental contract disputes, farmers with farmland rental contract disputes who were not obtaining reasonable compensation were significantly encouraged to transfer farmland to relatives, but the positive impact of obtaining reasonable compensation on the transfer of farmland to relatives was not significant, which meant that obtaining reasonable compensation for farmland rental contract disputes was conducive to recovering the economic losses that resulted from farmland rental contract disputes, which improved the expectations of the risks to farmland property rights and reduced the market segmentation of farmland transfers. Thus, obtaining reasonable compensation was conducive to improving the expectations of the risks to farmland property rights, which promoted farmland rental market participation and reduced the market segmentation of farmland transfers.

7. Conclusions and Policy Implication

This paper analyzed theoretically the impacts of farmland rental contract disputes and obtaining reasonable compensation on farmland rental market participation. Additionally, the survey data of 762 apple growers was used to examine empirically the above-mentioned impacts. The results indicated that: firstly, farmland rental contract disputes enhanced significantly the transfer of farmland to relatives and friends, but farmland rental contract disputes influenced insignificantly the transfer of farmland to non-relatives and non-friends. Secondly, obtaining reasonable compensation for farmland rental contract disputes significantly enhanced farmland rental market participation for all types of the objects of farmland transfers, which meant that obtaining reasonable compensation could increase the costs of violating farmland rental contracts and stabilize the expectations of the risks to farmland property rights. Thirdly, farmland rental contract disputes significantly motivated farmers’ choice of relatives as the objects of farmland transfers. Meanwhile, compared with the variable for no farmland rental contract disputes, the variable for obtaining reasonable compensation stabilized the expectations of the risks to farmland property rights, which enhanced farmland rental market participation and reduced the market segmentation of farmland transfers. Notably, previous studies mostly emphasized the relationship between farmland property rights systems and farmland rental market participation [6,27], but this paper not only analyzed the impact of land tenure insecurity on farmland transfers from the perspective of farmland rental contract disputes, but also examined the effectiveness of relevant laws and regulations of farmland transfers, which explained the relationship between land tenure security and farmland transfers from the perspective of the rule execution of farmland transfers.

In sum, in the context of an ambiguous structure of farmland property rights, participating in farmland rental markets was a freely pursued decision-making process, based on the actual situation. Farmland rental contract disputes resulted in high risks to farmland property rights, but farmers could choose relatives as the objects of farmland transfers, which meant that farmers participated in farmland rental markets by resorting to the reputation mechanism that resulted from repeated games and related games within the acquaintance society. Additionally, obtaining reasonable compensation for farmland rental contract disputes could reduce the market segmentation and discrimination of farmland transfers, which boosted farmland rental market participation. Thus, when the systems of farmland property rights were imperfect in the actual world, the key factor in enhancing farmland transfers and decreasing the market segmentation and discrimination was to stabilize the expectations of the risks to farmland property rights and reduce and solve farmland rental contract disputes. Notably, the above conclusions not only presented the importance of solving reasonably farmland rental contract disputes, but also provided a reference for improving relevant laws and regulations of farmland transfers, which was conducive to strengthening the quality of farmland rental markets for some developing countries.

Hence, several suggestions were proposed as follows: (1) farmland rental contract disputes should be reasonably reduced and solved to weaken the expectations of land tenure insecurity. The stability of farmland property rights depended on perfect farmland property rights systems and stable farmland rental contracts. In the context of an ambiguous structure of farmland property rights, stable farmland rental contracts could improve the expectations of land tenure security. More importantly, solving reasonably farmland rental contract disputes strengthened the stability of farmland rental contracts; (2) farmland rental contracts should be more standardized, and the effectiveness of relevant laws and regulations should be improved. Standardized farmland rental contracts were conducive to stipulating the responsibility of all parties who participated in farmland transfers and obtaining reasonable compensation for farmland rental contract disputes. Additionally, the effectiveness of relevant laws and regulations decreased the enforcement costs of farmland rental contracts; (3) The role of informal rules within the acquaintance society should be fully used to reasonably solve farmland rental contract disputes. With the gradual withdrawal of state power from rural society, the maintenance of village order relied on local public power and authority in rural areas [27,28]. Village cadres not only acted as the agents of the higher-level governments, which possessed the authority of formal rules, but also acted as the agents of farmers, which possessed the authority of informal rules. Hence, the roles of village cadres and informal rules should be fully employed to reduce the costs of farmland transfers. Moreover, in the context of high transaction costs of farmland rental markets, most farmers might use other farmers’ behaviors of participating in farmland rental markets as a reference, and a higher incidence of farmland rental contract disputes could increase the expectations of the risks to farmland property rights. Thus, solving reasonably farmland rental contract disputes was conducive to stabilizing the expectations of land tenure security, which developed farmland rental markets and improved allocative efficiency.