How Does Farmland Tenure Security Affect Rural Household Income? Empirical Evidence of China’s Survey Data

Abstract

A stable farmland tenure can optimize farmers’ allocation of production factors and then determine the income level and structure of farmers. In recent years, the reform of China’s farmland tenure system had been making efforts to strengthen the stability of farmland tenures. Will the farmland tenure security restrict agricultural development? Based on the data of the 2018 China Labor Dynamics Survey (CLDS), this paper empirically analyzes the impact of farmland tenure stability on rural household income and its mechanism. The results showed that the stability of land rights increased the per capita total income, wage income and agricultural productive income of farmers significantly. The stability of the farmland tenure significantly increased the income of rural households in plain areas and the income of rural households in villages with highly mechanied; additionally, employment opportunities increased significantly. The mechanism analysis showed that the stability of the farmland tenure had a significant positive effect on the total income of farmers by promoting the division of labor and agricultural investment. On this basis, this paper puts forward measures such as strictly controlling the readjustment of agricultural land, improving the rural labor market and financial market, and guiding nonplain areas to develop tertiary industry according to local conditions.

1. Introduction

As rural areas and farmers are the core of agriculture, increasing farmers’ income is a common concern of all sectors of society, and it is also the “Top priority” of agricultural and rural development. Increasing farmers’ income has always been a difficult problem to be solved urgently, and determining ways to ensure an increase in farmers’ income and reduce the income gap between urban and rural residents while in the process of eliminating absolute poverty and alleviating relative poverty in developing countries is always affecting the governments of various countries as well as the hearts of scholars from all walks of life. Farmers’ income is the key factor that affects the development of modern agriculture to achieve common prosperity [1]. As a country with a large population, China’s research on farmers’ income is of great significance to the transformation of agricultural development. In recent years, in order to increase farmers’ income, the Chinese government has introduced and implemented a series of reform measures to actively adjust the agricultural development mode, promote rural economic development, and increase farmers’ income. From 2004 to 2020, for 17 consecutive years, the No. 1 Document of the Central Committee has dealt with issues related to agriculture, rural areas and farmers in every year, stressing the need to increase farmers' incomes. According to the National Bureau of Statistics, the per capita income of rural residents grew by 3.8 percent in 2020, which was 2.6 percent faster than that of urban residents. However, in recent years, the economy has faced increasing downward pressure, and farmers lack the internal power and endurance to sustain a steady growth of income. The average annual growth rate of rural household income during the 13th Five-Year Plan period was 6.74 percent, which was 4.21 percent lower than that during the 12th Five-Year Plan period. Obviously, only by solving the problem of a continuous increase in rural household income can we achieve Chinese-style modernization development and achieve common prosperity [2].

The concurrent employment of farmers and the diversity of the employment structure determine the diversification of economic activities and income structure of farmers. Therefore, rural household income is not only affected by its own factors, but it is also closely related to rural institutional innovation, policy reform, market development, and other factors [3,4,5,6]. With the expansion of the scale of the nonagricultural transfer of the rural labor force, the impact of the implementation of the rural internal system on household income has become a new focus of academic circles. As an important land system reform in rural areas, rural land confirmation is an important measure for the CPC Central Committee to promote rural development and increase rural household income. The strengthening of the stability of farmland tenures has changed the allocation of the agricultural production factors of farmers [7,8] and is the incentive to affect the change in rural household income level and income structure [9]. As a major development in China’s rural land system, land-use confirmation was introduced on a pilot basis in 2009 and was implemented nationwide in 2013. In 2018, the land-use confirmation, certification, and registration of farmland before contracting were basically completed. The reform of China’s farmland property rights system were making efforts to strengthen the stability of farmland tenures. Now, the problem of whether the stability of farmland tenures can effectively promote the increase in rural household income still remains. Under the background of central rural revitalization, it is of great significance to explore the influence of the stability of farmland tenures on rural household income to promote the development of agricultural and the rural economy and promote the sustainable growth of rural household income.

The academic circle has carried out extensive discussions on rural household income, but research mainly focuses on the impact of nonagricultural labor transfer and farmland allocation on rural household income [10,11]. With the deepening of the reform of China’s farmland tenure system, more and more scholars have paid attention to the impact of the stability of farmland tenures on rural household income [12,13]; however, there are still differences in research conclusions. Some scholars believe that rural land property rights change the land rights relationship between the government, village collectives, and farmers; realize the ownership of collective assets and the right of income distribution for farmers; and help to expand new channels of income increases for farmers [14,15]. The influence of farmland tenures on rural household income needs a certain path. Scholars have analyzed the influence of farmland tenures on rural household income from different influence paths. Yao et al. (2000) believe that the enhancement of the stability of farmland tenures increases the wage income and transfer income of farmers by promoting the transfer of farmland [16]. Zhang et al. (2020), based on the survey data of farmers in 2010 and 2015, believe that the enhancement of the security of farmland tenures in China promotes the operational income of farmers by improving the possibility that farmers obtain formal credit [17]. Other scholars believe that the stability of farmland tenures has no significant impact on rural household income [18,19,20,21]. Do and Iyer (2008) pointed out that the land property rights system reform in Vietnam in 1993 promoted farmers’ nonfarm labor time input but had no significant impact on their income [22]. Hombrados et al. (2015), based on the survey data of Tanzanian farmers, used the propensity score matching method to demonstrate that the confirmation of farmland rights had no impact on the income growth of farmers [23].

In conclusion, research on the impact of farmland tenures on rural household income has not received enough attention and lacks more rigorous proof. Most studies on the impact of farmland tenures on rural household income are derived from the research conclusions on the impact of farmland tenures on farmers’ production behavior and performance without systematic theoretical and empirical analysis. In addition, most of the existing studies focus on the behavioral effects caused by the policy change from the instability of land rights to the confirmation of farmland rights, but few studies focus on whether there are differences in rural household incomes after completion of the confirmation of farmland rights and causes of the differences. In view of this, this paper takes 2018, the last year of farmland ownership registration and certification, as the starting point of the study, and adopts the data of the China Labor Dynamics Survey (CLDS) in 2018 by using a linear regression model, Tobit model, propensity score matching model, and intermediary effect model. This paper analyzes the influence of the enhanced stability of farmers’ farmland tenure on rural household income and its internal transmission mechanism under the background of land ownership confirmation and further examines the heterogeneity of the influence of farmers’ farmland tenure stability on rural household income from the perspective of economic location and business endowment, so as to further enrich the research in this field.

The research arrangement of this paper is as follows: Section 2 is the theoretical analysis and research hypotheses; Section 3 is the materials and methods; Section 4 is the empirical analysis; and Section 5 is the conclusions and recommendations.

2. Theoretical Analysis and Research Hypotheses

2.1. Analysis of the Change in Land System and Current Situation of Peasant Household Income in China

After 40 years of reform and opening, the reform of our country’s agricultural land system always advances in the direction of strengthening the protection of land property rights, specifically from two aspects of agricultural land power empowerment. On the one hand, it strengthens the dominant position of farmers’ farmland property rights. In 1978, 18 farmers in Xiaogang Village, Anhui Province, implemented the “big contract work”, and the main body of agricultural management was changed from “team” to farmer, which marked the beginning of the transformation of the main body of agricultural management. In 1982, the No. 1 Document of the Central Committee clearly affirmed the dominant position of agricultural operation of “contracting production to households and contracting drying to households” at the policy level. In 2002, the central government promulgated the Land Contract Law, which legally established the dominant position of farmers in agricultural operations and protected farmers’ land rights and interests. On the other hand, it also strengthened the stability of farmers’ farmland property rights. In 1984, the No. 1 Central Document proposed that the land contract management relationship should remain unchanged for 15 years. In 1993, it was stipulated that the second round of the household contract for land should be extended for another 30 years from 15 years. In 2008, for the first time, the central government explicitly proposed to do a good job in determining the right to agricultural land and carry out pilot work. In 2013, we comprehensively carried out the work of determining, registering, and certificating rural land rights; clarified the four-to-four ranges of rural households’ land; allowed farmers to allocate rural land resources within the limits of property rights; clearly defined rural land property rights; and further strengthened rural households’ land rights.

According to the data of the China Statistical Yearbook, since the reform and opening up, China’s agriculture has been developing rapidly, and the income level of rural households has been growing continuously; however, the growth rate of rural households’ income has shown fluctuating changes. See Figure 1 for details. Generally speaking, from 1999 to 2007, the growth rate of farmers’ income was in a process of continuous fluctuation. In the early 20th century, the government continued to improve relevant laws and gradually promoted the process of determining agricultural land rights, which further consolidated farmers’ land rights and interests, promoted agricultural production, and increased farmers’ income to a certain extent. From 2008 to 2011, the growth of farmers’ income first decreased slightly and then continued to rise. In 2008, the confirmation and certification of agricultural land rights began. In 2009, the first pilot program, the Work Plan for Pilot Registration of Contracted Rural Land Management Rights, was released, and for the first time, the pilot certification and confirmation of rural land rights began at the village level, which further clarified rural households’ ownership of farmland. From 2012 to 2019, the growth of the per capita disposable income of rural residents first decreased slowly and slightly, and then rose slowly and slightly. Overall, the growth rate of farmers’ income in this period tended to be stable. This period was also the stage of the comprehensive implementation of the determination of farmland rights. The “separation of three rights” of farmland defined the main body of farmland property rights, enhanced the land rights and interests of rural residents, promoted the development of agriculture to a certain extent, and improved the income level of farmers.

Generally speaking, with the implementation of China’s agricultural land system reform, farmers’ income level presented a relatively high growth rate. Thus, we cannot help but propose that the implementation of land reform policies to promote the stability of farmland property rights has a positive impact on peasant household income to some extent.

2.2. Theoretical Framework and Research Hypothesis

Clear land tenures can give farmers more rights to freely allocate production factors and influence the total income and income structure of farmers by influencing their decision making and allocation behavior of agricultural production factors [9]. The enhanced stability of land tenures, on the one hand, makes farmers feel the solidification and unadjustability of the contracted land management rights [24], improves the security perception of farmers’ land tenures, and promotes farmers to increase agricultural investment. On the other hand, it improves farmers’ control over farmland tenures, promotes the transfer of farmland and free flow of rural labor in agricultural and nonagricultural sectors [25], and expands the income channels of peasant families. By combing the theory of property rights and related literature [2], this study believes that from the theoretical level, the enhancement of the stability of farmland property rights affects the income of farmers from three aspects: agricultural land circulation, household labor division, and agricultural investment. See Figure 2 for details.

2.2.1. Farmland Tenure Security, Agricultural Land Circulation, and Rural Household Income

With the development of the economy, the land system arrangement of “equity-oriented and per-capita distribution” does not consider the heterogeneity of farmers’ agricultural production capacity, which results in the loss of the allocative efficiency of labor and agricultural land resources. The enhancement of the stability of farmland tenures improves the efficiency of resource allocation by influencing the transfer behavior of farmers [26]. Specifically, only by clearly defining the subject of property rights, allowing the free transfer of property rights, and effectively protecting the transfer income rights of subjects of property rights [27,28] can the subject of property rights allocate resources to the maximum extent within the scope of property rights constraints to obtain maximum income [29,30]. For transferred households of agricultural land, the transfer of agricultural land can achieve the large-scale and intensive management of agricultural land, reduce agricultural production costs, improve land productivity, and increase the agricultural operation income of transferred households [31,32,33,34]. As for the transfer of agricultural land to households, with the development of the economy, the income gap between urban and rural areas increased. Based on the principle of the maximization of household income, farmers with comparative advantages of nonagricultural employment transfer to agricultural land [35,36,37,38]. On the one hand, farmers can obtain land rent by transferring land, which improves the property income of farmers. On the other hand, the transfer of farmers from land can release a part of the labor force of the family to engage in nonagricultural work with higher income, which thus increases the wage income of farmers.

2.2.2. Farmland Tenure Security, Household Labor Division, and Rural Household Income

Farmland tenures not only affect the utilization mode and yield rate of farmland resources, but they also affect farmers’ employment choices [39]. With the acceleration of urbanization, the realization of a demographic dividend has been promoted. A large number of surplus rural labor forces has flooded into cities, which not only contributes to the accumulation of social wealth but also contributes to an increase in family income [40]. A large number of rural surplus labor forces transfer to the second and third industries, which requires the redistribution of rural labor resources. Clear farmland tenures can enable farmers with different resource endowments to play their comparative advantages, can make farmers with agricultural production advantages engage in agricultural production, and can make those with nonagricultural employment advantages transfer from the agricultural sector to the nonagricultural sector, which thus reduces the relative probability of farmers falling into poverty [41,42,43]. On the one hand, the enhanced stability of the tenure reduces the risk of land loss in the off-farm transfer process of farmers, solves their worries about off-farm transfer, and further promotes the off-farm transfer of rural labor [44,45]. On the other hand, a clear tenure of agricultural land is conducive to the accumulation of land by farmers with high productivity from those with a lower professional level [46], which is conducive to the realization of the large-scale management of agricultural land. Furthermore, agricultural production factors can be optimized to improve the agricultural production efficiency and rural household income by adding short-term inputs, increasing agricultural employees, and purchasing large-scale agricultural machinery and tools.

2.2.3. Farmland Tenure Security, Agricultural Investment, and Rural Household Income

In the final analysis, a property right is a property subject related to material interests, and the property subject exercises the property right based on the maximization of interests [47]. According to the economic incentive function of property rights, the instability of property rights causes farmers to lack long-term expectations for land, and they have no confidence to invest a large number of fixed assets with high specificity in land. The enhanced stability of farmland tenures reduces the risk of farmland being expropriated by village collectives and occupied by tenants [48], protects farmers’ land rights and interests from being infringed, improves the expected behavior of property owners, stimulates farmers to make medium- and long-term investments in land [49], improves their enthusiasm in agricultural production and management, and increases their agricultural operational income.

In summary, this study puts forward the following hypotheses:

Hypothesis 1.

The farmland tenure security improves rural household income by promoting the transfer of farmland.

Hypothesis 2.

The farmland tenure security improves rural household income by optimizing the division of labor.

Hypothesis 3.

The farmland tenure security improves rural household income by promoting agricultural investment.

3. Materials and Methods

3.1. Data Sources

The data used in this paper were from the 2018 China Labor Dynamics Survey (CLDS) conducted by the Social Science Survey Center of Sun Yat-sen University. The survey samples covered 29 provinces in China and involved relevant information of family farmland property rights; family structure and income; land use and agricultural production, which had a certain representativeness and stability; and provided effective data support for the empirical research of this paper. This paper took the family as the research object and mainly adopted the dataset at the family level, which was supplemented by the relevant characteristic information at the individual and village level. The data included a sample of 16,537 labor force individuals, 13,501 households, and 368 communities. After excluding the nonrural residents and the samples missing key variables, the household questionnaire, individual questionnaire, and village questionnaire were matched and combined to obtain 2106 household samples.

3.2. Variable Selection

3.2.1. Explained Variable

The dependent variable of this paper is rural household income. In order to further analyze the influence of the stability of farmland tenures on different income categories of peasant households, this paper also takes the per capita household gross income, per capita wage income, per capita agricultural productive income, and per capita property income as the explained variables. In the process of the empirical analysis, this paper takes the logarithm of each income index and puts it into a regression model.

3.2.2. Explanatory Variables

The explanatory variable of this paper is the stability of farmland tenures, which is measured by whether the farmland has been adjusted since 2003. If the farmland has not been adjusted, the value is 1, indicating that the farmland tenure is stable. If the farmland has been adjusted, the value is 0, indicating that the farmland tenure is unstable.

3.2.3. Intermediary Variable

According to the above theoretical analysis, clear property rights can affect farmers’ total income and income structure by influencing their decision making and allocation behaviors regarding agricultural production factors such as agricultural land, labor, and capital. Therefore, three variables of the household division of labor, agricultural investment, and farmland transfer are used as intermediary variables in this paper. In the empirical analysis, this paper made dimensionless processing to the intermediary variables and took the logarithmic treatment of the agricultural land transfer and agricultural investment and included them into a regression model.

3.2.4. Control Variable

The control variables in this paper include other variables that affect the income of farmers. Referring to the practice of the existing literature, this paper introduces control variables from three levels: household head characteristics, family characteristics, and village characteristics. The characteristics of the household head include two variables: the age of the household head and the political status of the household head [50,51]. In terms of family characteristics, five variables such as the number of family members, the proportion of females, the per capita cultivated land area, tractors in the family, and other large agricultural machinery in the family were selected for measurement [52,53,54]. In terms of the village characteristics, three variables were selected: nonagricultural economy of villages, unified irrigation and drainage service of villages, and training of agricultural production technology [25,55,56].

In addition, with different levels of regional economic development, farmers’ income levels will also be different. Therefore, this study selects the Eastern region, Western region, and Central region as the regional variables to control the impact of regional variables on income. See

Table 1. Definition of variables and descriptive statistical analysis.

Variable Name	Variable Definitions	Standard Deviation	Mean	Minimum Value	Maximum Value
Gross household income per capita	Total income/family size (yuan)	9088	9186	80	330,000
Per capita wage income	Wage income/family size (yuan)	14,865	3376	0	280,000
Per capita agricultural productive income	Income from agriculture, forestry, animal husbandry, and fishery/household number (yuan)	2476	481.7	0	31,568
Per capita property income	Property income/household size (yuan)	696	58.4	0	11,030
Farmland tenure security	Whether farmland has been adjusted since 2003; no = 1, yes = 0	0.477	0.629	0	1
Division of labor at home	Percentage of households engaged in agricultural production (%)	0.20	0.451	0.1	1
Investment in agriculture	Total cost of agricultural production (yuan)	25,561	16,781	100	200,000
Rural land circulation	Transfer area of cultivated land (mu)	47.16	17.00	0.4	566
Household age	Age of head of household (age)	10.75	52.34	26	79
Political status of household	head party member = 1, democratic party = 2, masses = 3	0.259	1.898	1	2
Size of household	Size of household	1.691	4.658	1	10
Female ratio	Number of women in a household/family	0.167	0.512	0	1
Per capita area of cultivated land	Land/number of households	9.56	5.112	0	100
Have a tractor at home	Do you have a tractor? Yes = 1, no = 0	0.582	0.517	0	1
There are large farm machines and implements at home	Is there any large farm machinery? Yes = 1, no = 0	0.307	0.0735	0	1
Village nonagricultural economy	Is there an off-farm economy? Yes = 1, no = 0	0.45	0.153	0	1
Village mechanical farming service	Does it provide mechanical tillage service? Yes = 1, no = 0	0.416	0.401	0	1
Village agricultural production technology training	Does it provide unified agricultural production technology training? Yes = 1, no = 0	0.413	0.699	0	1
Eastern region	Yes = 1, no = 0	0.501	0.398	0	1
Central region	Yes = 1, no = 0	0.389	0.198	0	1
Western region	Yes = 1, no = 0	0.511	0.389	0	1

for details.

3.3. Model Specification

The following three models were constructed in this study.

First, the stability of the farmland tenure on the impact of the household per capita total income.

This paper constructed the following empirical model of the stability of the farmland tenure affecting the per capita household gross income:

$$Y_i=c_0+c_{1}Title+{\displaystyle \sum _{n=1}{c}_{2n}}{X}_{ni}+{\epsilon }_{\mathrm{i}}$$

(1)

where $Y_i$ represents the per capita household income of the I-th farmer; $Title$ indicates the stability of the farmland property rights of the I-th farmer; $X_{ni}$ represents the other control variables affecting the per capita household gross income; and ${\epsilon }_{\mathrm{i}}$ is random interference. Among them, the estimation method used is the least square method (OLS).

Second, the stability of the farmland tenure on the structure of the household per capita total income.

In this paper, the following Tobit model was constructed to analyze the influence of the stability of farmland property rights on the income structure of farmers:

$$y_i=\{\begin{array}{l}{y}_i^{*} \mathrm{if} y_i^{*}>0\\ 0 \mathrm{if} y_i^{*}\le 0\end{array}$$

(2)

$${\mathrm{y}}_{\mathrm{i}}^{*}={\lambda }_0+v_{1}Title+{\displaystyle \sum _{n=1}{\lambda }_{2n}}{X}_{ni}+{\delta }_{\mathrm{i}}$$

(3)

where $y_i$ denotes the per capita wage income, per capita agricultural productive income, and per capita property income of the I-th farmer and is not negative; $y_i^{*}$ is a latent variable; $Title$ indicates the stability of the farmland property rights of the I-th farmer; $X_{ni}$ represents the other control variables that affect the dependent variable; and ${\delta }_{\mathrm{i}}$ is random interference.

Third, the stability of the farmland tenure affects the mechanism of the farmers’ income.

The mediating effect model established is as follows:

$$Y_i=v_0+v_{1}Title+{\displaystyle \sum _{n=1}{v}_{2n}}{X}_{ni}+{\gamma }_1$$

(4)

$$ME{D}_i=k_0+k_{1}Title+{\displaystyle \sum _{n=1}{k}_{2n}}{X}_{ni}+{\gamma }_2$$

(5)

$$Y_i=d_0+d^{,}Title+d_{1}ME{D}_{\mathrm{i}}+{\displaystyle \sum _{n=1}{d}_{3n}}{X}_{ni}+{\gamma }_3$$

(6)

Among them, $Title$ represents the stability of farmland property rights; $ME{D}_i$ represents the transfer of agricultural land, agricultural investment, and household labor allocation; $Y_i$ represents household income; $X_{ni}$ is the control variable; $v,k$, and $d$ are the parameters to be estimated, respectively; and $\gamma$ (i = 1, 2, and 3) are the random error terms.

4. Empirical Results and Analysis

4.1. Benchmark Regression of Impact of Farmland Tenure on Rural Household Income

According to Formula (1), regression result 1 was obtained, and according to Formula (2), regression results 2, 3, and 4 were obtained. See

Table 2. Regression results of influence of farmland tenure on rural household income.

	Regression 1	Regression 2	Regression 3	Regression 4
Variable	Gross Household Income per Capita	Per Capita Wage Income	Per Capita Agricultural Productive Income	Per Capita Property Income
Farmland tenure security	0.341 **	0.278 *	0.398 *	0.698
	(2.75)	(1.21)	(1.78)	(0.31)
Household age	−0.012 *	−0.029	−0.255 *	0.036
	(−1.89)	(−0.48)	(−1.68)	(0.31)
Political status of household	−0.239	0.489	−0.001	2.999
	(−1.70)	(0.28)	(−0.00)	(0.40)
Size of household	−0.168 ***	0.369	0.118	0.654
	(−3.88)	(1.17)	(0.14)	(0.61)
Female ratio	−0.068	−6.588 **	11.572	0.601
	(−0.14)	(−1.89)	(1.34)	(0.05)
Per capita area of cultivated land	0.021 **	−0.369 ***	1.766 **	−1.519
	(2.39)	(−3.41)	(2.27)	(−1.09)
Have a tractor at home	0.219*	−0.148	3.702	−1.35
	(1.6)	(−0.12)	(1.14)	(−0.32)
There are large farm machines and implements at home	−0.048	1.560	6.598	−6.412
	(−0.17)	(0.81)	(1.51)	(−0.79)
Village nonagricultural economy	0.328 **	−1.689	−1.984	10.112 **
	(2.38)	(−1.12)	(−0.43)	(2.13)
Village mechanical farming service	0.11	−1.139	3.879 *	2.712
	(0.92)	(−1.11)	(1.33)	(0.65)
Village agricultural production technology training	−0.006	3.178 **	1.536 **	0.131
	(−0.04)	(2.40)	(2.44)	(0.03)
Regional variables	Controlled	Controlled	Controlled	Controlled
cons	9.941 ***	0.746	−14.794	−32.678 *
	(15.1)	(0.13)	(−1.05)	(−1.68)
N	2106	2106	2106	2106
F	9.285 ***
LR		35.62 ***	31.83 ***	27.44 ***
r2	0.1345

Note: *, **, and *** denote significance at the 1%, 5%, and 10% levels, respectively.

for the specific regression results. Regression result 1 showed that the stability of the farmland tenure had a positive impact on the per capita total income of households. In addition, from the perspective of the farmers’ income structure, regression results 2, 3, and 4 showed that the stability of the farmland tenure had a significant positive impact on the farmers’ per capita wage income and per capita agricultural productive income but had no significant impact on the per capita property income. It can be seen that the improvement effect of the stability of the farmland tenure on the per capita total income of rural households mainly came from the increase in the per capita agricultural productive income and per capita wage income of rural households. As for the control variables, the age of the head of household significantly negatively affected the per capita total income of the household and the per capita agricultural productive income of the household. The number of family members had a positive impact on the per capita total income of the family. The female share ratio had a negative effect on the per capita wage income of the household. The per capita cultivated land area positively affected the per capita household total income and per capita agricultural productive income and negatively affected the per capita wage income. Having a tractor at home positively affected the total income per capita. The rural nonagricultural economy had a positive impact on the per capita household income and per capita property level. The unified provision of irrigation water services in villages had a significant positive impact on the per capita agricultural productive income. In addition, rural agricultural production technology training had a significant positive impact on the household per capita wage income and per capita agricultural productive income.

4.2. Heterogeneity Analysis of Impact of Farmland Tenure on Rural Household Income

In order to further analyze the relationship between the stability of farmland property rights and the income of farmers, this paper carried out a heterogeneity analysis on the different mechanization levels, employment opportunities, and topography of households. The results are shown in

Table 3. Group regression results of the impact of farmland right confirmation on per capita household gross income.

Variable	Whether the Village Provides Machine Tillage Service		Whether the Village Organizes and Arranges the Labor Force to Go Out for Service		Village Is Plain Area
	No	Yes	No	Yes	No	Yes
Farmland tenure security	0.239 *	0.428 **	0.274 *	0.357 **	0.351	0.404 **
	(1.97)	(2.11)	(1.78)	(2.05)	(1.51)	(2.08)
Control variables	Controlled	Controlled	Controlled	Controlled	Controlled	Controlled
cons	9.890 ***	11.182 ***	10.153 ***	9.653 ***	9.721 ***	10.186 ***
	(9.8)	(16.55)	(12.82)	(9.81)	(16.01)	(8.62)
N	1278	828	1586	520	1212	894
r2	0.1812	0.1507	0.1412	0.1671	0.1418	0.1378

Note: *, **, and *** denote significance at the 1%, 5% and 10% levels, respectively.

.

First of all, for villages with good mechanization conditions and villages with poor mechanization conditions, the improvement in the land right stability can significantly promote farmers’ income, but villages with better mechanization conditions have a greater increase in farmers’ income as the stability of farmland property rights is enhanced. It shows that the improvement in the stability of land rights and the matching of agricultural external production conditions will be more conducive to the effect of the agricultural land property rights policy. Secondly, for villages with or without organizing labor going out to serve, the stability of farmland property rights can significantly improve the income of rural households. However, with the enhancement in the stability of property rights, the income of rural households in villages with organizing labor going out to serve increases more. Finally, the stability of farmland property rights significantly increased the income of farmers in plain areas but had no significant effect on the income of farmers in nonplain areas.

4.3. Robustness Test of the Influence of the Stability of Farmland Tenure Right on Rural Household Income

In this paper, propensity score matching (PSM) was used to test the robustness of the original regression model. In order to ensure the validity of matching, the cosupporting hypothesis should be tested first. See Figure 3 for details. It can be seen that the probability distributions of the propensity scores of the last two samples were closer, which indicates that the observable characteristics of the two samples were more similar. At the same time, the observed values of the vast majority of the samples fell within the common value range, which fully indicates that the matching effect was good and the common supporting hypothesis was satisfied.

Further, a balance test was conducted on the data matching propensity scores, and the results are shown in

Table 4. Balance test results.

Variable		Mean		Standard Deviation %	Standard Deviation Reduction	t Test
		Treatment Group	Control Group			t Value	p Value
Household age	Before match	50.89	51.85	−6.5		−0.68	0.478
	After match	50.89	51.70	−5.7	11	−0.73	0.456
Size of household	Before match	4.453	4.692	−12.0		−1.38	0.158
	After match	4.453	4.361	4.1	66.2	0.56	0.569
Female ratio	Before match	0.482	0.480	1.3		0.14	0.875
	After match	0.482	0.489	−2.8	−101	−0.33	0.729
Have a tractor at home	Before match	0.457	0.332	25.4		2.78	0.005
	After match	0.457	0.461	0.0	100	0.00	1.000
There are large farm machines and implements at home	Before match	0.0828	0.0841	−0.3		−0.04	0.960
	After match	0.0828	0.0693	4.5	−962.8	0.58	0.538
Village nonagricultural economy	Before match	0.141	0.151	−4.0		−0.43	0.661
	After match	0.141	0.158	−5.3	−33	−0.67	0.491
Village mechanical farming service	Before match	0.359	0.421	−10.9		−1.21	0.232
	After match	0.359	0.401	−6.5	40.6	−0.81	0.408
Village agricultural production technology training	Before match	0.810	0.656	36.8		4.01	0.000
	After match	0.810	0.780	0.6	97.1	0.10	0.911

. It can be seen from the test results that the standard deviation of each covariate after matching was less than 10%, indicating that matching effectively reduced the difference between households in the nonadjusted group and the adjusted group of agricultural land. In addition, a t test showed that after the propensity score matching, the p values of each variable were all greater than 0.05, which indicated that there was no coefficient difference between the treatment group and the control group. Therefore, the matching results balanced the data well, and the matching was effective.

Then, the average treatment effect (ATT) of the stability of the farmland property rights on household income can be further calculated. In this paper, three methods of nearest neighbor matching, caliper matching, and kernel matching were comprehensively adopted to estimate the average processing effect of the samples, and the results are shown in

Table 5. PSM test of farmland right confirmation on rural household income.

ATT (Average Treatment Effect)	Gross Household Income per Capita	Per Capita Wage Income	Per Capita Agricultural Productive Income	Per Capita Property Income
Nearest neighbor matching	0.5801 ** (2.29)	0.1489 * (1.86)	0.2981 * (1.91)	0.0088 (0.04)
Caliper matching	0.4321 *** (2.64)	0.1731 * (2.12)	0.3144 * (2.16)	0.0762 (0.55)
Kernel matching	0.4313 *** (2.74)	0.1717 * (2.13)	0.3186 * (2.25)	0.0759 (0.56)

Note: *, **, and *** denote significance at the 1%, 5%, and 10% levels, respectively.

. Except for the significant increase in the household per capita gross income in the caliper matching and kernel matching results, the other coefficient directions and significance sizes were consistent with the basic regression model, which confirms that the baseline regression results were robust.

4.4. Mechanism of Effect of Stability of Farmland Tenure on Rural Household Income

4.4.1. Analysis of the Mediating Effect of Farmland Transfer

Table 6. Mediating effect of farmland transfer.

Variable	(1)	(2)	(3)
	Gross Household Income per Capita	Rural land Circulation	Gross Household Income per Capita
Farmland tenure security	0.341 **	−0.051	0.329 **
	(2.75)	(−0.42)	(2.17)
Rural land circulation			0.066
			(0.85)
Control variables	Controlled	Controlled	Controlled
cons	9.941 ***	2.656 ***	9.672 ***
	(15.10)	(6.02)	(14.60)
N	2106	2106	2106
r2	0.1345	0.3689	0.1356
Sobel p value		0.6359

Note: ** and *** denote significance at the 5% and 10% levels, respectively.

shows the mediating effect of the farmland transfer. Regression result 2 was obtained by substituting the intermediary variable of the farmland transfer into Formula (5). The empirical results showed that the stability of the farmland property rights did not have a significant impact on the farmland transfer. The next step was the Sobel test. The p value of the Sobel test was 0.6359, which did not pass the significance test, so the test was stopped. It showed that the intermediary effect of the farmland transfer did not exist. Therefore, hypothesis 1 in Section 3 is not valid. The possible explanations are as follows: First of all, land is a kind of personified property for farmers. The enhancement of the stability of the property rights of the agricultural land strengthens the identity of farmers, enhances the holding type of the agricultural land held by farmers, further intensifies the “endowment effect” of farmers, increases the expected price of farmers transferring out of agricultural land, and inhibits the transfer of agricultural land. Secondly, for farmland transfer households, only land transfer can bring a high enough land yield rate to promote the increase in household income. But at present, it is difficult to realize the large-scale management of the average cultivated land area of the peasant household, and the income function of the peasant household is limited.

4.4.2. Analysis of the Mediating Effect of Family Division of Labor

Table 7. The mediating effect of household division of labor.

Variable	(1)	(2)	(3)
	Gross Household Income per Capita	Family Division of Labor	Gross Household Income per Capita
Farmland tenure security	0.341 **	0.278 ***	0.297 **
	(2.75)	(4.90)	(2.26)
Family division of labor			0.169 ***
			(4.61)
Control variables	Controlled	Controlled	Controlled
cons	9.941 ***	0.780 ***	19.895 ***
	(15.10)	(12.26)	(15.01)
N	2106	2106	2106
r2	0.1345	0.4962	0.1287

Note: ** and *** denote significance at the 5% and 10% levels, respectively.

shows the mediating effect of the family division of labor, in which the mediating variable family division of labor is substituted into the intermediary model Formula (5) to obtain regression result 2, which shows that the stability of farmland property rights has a positive impact on the family division of labor, which indicates that the enhancement of the stability of farmland property rights prompts farmers to increase the proportion of household farming. The stability of farmland property rights and the division of labor in the family were put into Formula (6) to obtain regression result 3, which showed that the division of labor in the family had a significant positive impact on the per capita household income. Further, after adding the household labor input as an intermediary variable, the regression coefficient of the farmland property stability on the household per capita gross income decreased from 0.341 in regression result 1 to 0.297 in regression result 3 with the significance remaining the same, which indicated that the family division of labor played a partial intermediary role. Therefore, hypothesis 2 in Section 3 is valid.

4.4.3. Analysis of the Mediating Effect of Agricultural Investment

Table 8. The mediating effect of agricultural investment.

Variable	(1)	(2)	(3)
	Gross Household Income per Capita	Investment in Agriculture	Gross Household Income per Capita
Farmland tenure security	0.341 **	0.208 *	0.305 *
	(2.75)	(1.68)	(2.13)
Investment in agriculture			0.186 ***
			(3.23)
Control variables	Controlled	Controlled	Controlled
cons	9.941 ***	9.287 ***	7.955 ***
	(15.1)	(11.82)	(12.15)
N	2106	2106	2106
r2	0.1345	0.2208	0.16

Note: *, **, and *** denote significance at the 1%, 5%, and 10% levels, respectively.

shows the mediating effect of agricultural investments. Regression result 2 showed that the stability of farmland property rights had a positive impact on farmers’ agricultural investment. To further test the mediating effect of agricultural investment, regression result 3 could be obtained by substituting the stability of farmland property rights and agricultural investment into Formula (6). The results showed that agricultural investments had a significant impact on the per capita total income of rural households, and the coefficient was positive. In addition, after the addition of the intermediary variable agricultural investment, the influence of the stability of farmland property rights on the per capita gross income of rural households was still significant, the coefficient was positive, and the coefficient decreased from 0.341 in regression result 1 to 0.305 in regression result 3. It showed that agricultural investment had a partial mediating effect. Therefore, hypothesis 3 is valid: farmland confirmation can improve the per capita gross income of peasant households by promoting agricultural investments.

5. Conclusions and Suggestions

This paper took 2018, the last year of farmland ownership registration and certification, as the starting point to analyze the theoretical mechanism of the influence of farmland property stability on farmers’ income. Using the data of the China Labor Dynamics Survey (CLDS) in 2018, this paper used a linear regression model, Tobit model, propensity score matching model, and intermediary effect model. This paper analyzed the influence of the stability of farmers’ farmland property rights on farmers’ income and the influence of the internal transmission mechanism. The main conclusions are as follows:

First, on the whole, the increased stability of the farmland property rights increased the per capita income of the rural households. From the perspective of farmers’ income structure, the stability of farmland property rights had a positive impact on farmers’ per capita wage income and per capita agricultural productive income but had no significant impact on per capita property income.

Second, the influence mechanism of farmland property stability on household income was as follows: farmland property stability not only had a direct impact on the per capita household income, but also indirectly increased the per capita household income by promoting the division of labor and agricultural investment in the family. In addition, it had not been proved that farmland transfer had an impact on farmers' income through the stability of farmland property rights

Thirdly, the subsample regression results showed the following: First of all, for villages with good mechanization conditions and villages with poor mechanization conditions, the improvement in the land right stability can significantly promote farmers’ income, but villages with better mechanization conditions have a greater increase in farmers’ income as the stability of farmland property rights is enhanced. This result shows that the stability of land rights and the external agricultural production conditions match each other, which will be more conducive to the effect of the farmland property rights policy. Secondly, for villages with or without the labor force going out to serve, the stability of the farmland property rights can significantly improve the income of rural households, and with the enhancement of the property rights stability, the income of rural households in villages with the labor force going out to serve increases more. Finally, there is regional heterogeneity in the influence of farmland property stability on household income. The stability of the farmland property rights significantly increased the income of the farmers in the plain areas but had no significant effect on the income of the farmers in the nonplain areas.

Based on the above research conclusions, the following policy suggestions are put forward:

First, in the actual game of rural land property rights, the readjustment of rural land should be strictly controlled, and the synergistic efficiency of rural land property rights policies should be brought into play so as to reduce and eliminate the disconnection and disharmony between rural land policies, laws, and regulations, so as to realize the stability of rural land property rights.

Second, the rural labor market should be improved and the rural labor flow of the institutional barriers should be removed. The influence of the stability of farmland property rights on farmers’ income requires the conduction effect of the household division of labor. The perfection and cultivation of the rural labor market plays an important role in the effect of the farmland property rights policy. Therefore, the current focus should be on promoting equal access to public services, promoting the integration of rural migrants into enterprises, putting children into schools and families into communities, and realizing that rural migrants “bring their capital into cities”.

Third, diversified rural financial institutions should be developed and rural financial markets should be perfected. Agricultural investment plays an intermediary role in the influence of farmland property stability on peasant household income. It shows that the effect of the farmland property rights policy needs a perfect rural financial market to maximize the income of farmers. But at present, our rural financial market is mainly controlled by the government through administrative means, and the market mechanism does not play a decisive role. Going forward, we should further open up the rural financial market, accelerate the development of new rural financial organizations, encourage private capital to participate in the reform of local financial institutions, form diversified rural financial institutions, and expand farmers’ financing channels.

Fourthly, we should guide the development of tertiary industry in nonplain areas and reduce the effect of farmers’ land endowment. There is regional heterogeneity in the influence of land property stability on household income. Farmland property rights in the plain areas can play a better role in the allocation of land, labor, and capital and promote farmers to increase production, which will in turn increase income of farmers in the plain area. Therefore, on the one hand, the government should increase policy publicity in mountainous and hilly areas to enhance farmers’ awareness of farmland property rights. On the other hand, mountainous areas should give full play to their ecological advantages to drive the development of tertiary industry. Based on the advantages and limitations of mountainous areas, the development of rural tourism is encouraged, the integration of primary and tertiary industries is promoted, the income channels of rural households are expanded, and the survival dependence of rural households on land is reduced.