Does Land Transfer Enhance the Sustainable Livelihood of Rural Households? Evidence from China

Abstract

Land transfer and its socio-economic impact are key areas of research interest. Such an examination can help to enhance the sustainability of farming livelihoods, maximise livelihood strategies, and achieve sustainable development. This paper establishes a sustainable livelihood evaluation index for rural households based on sustainable livelihood theory. It measures the degree of sustainability in the livelihoods of farmers based on field research data from 650 rural households in Hubei Province, China, and analyses the impact of land transfer using a multiple linear regression model. A number of control variables were identified and introduced into the analysis. It also uses the regression decomposition approach to investigate the impact of each factor on the sustainable livelihoods of rural households. The findings revealed that (1) land transfer can significantly increase the sustainability of rural households’ livelihoods; (2) livelihood sustainability increases with the size of the land transfer area; and (3) the primary elements determining the ability of rural households to maintain sustainable living are land transfers, the amount of land transferred, and the size of the family. Based on these findings, this study argues for the need to improve land transfer management, accelerate agricultural industrialisation and promote the transfer of land contract management rights to improve the livelihoods of rural households.

1. Introduction

Agriculture, rural areas, and farmers have long been the focus of research by agricultural economists [1,2,3,4]. One of the key issues is the dynamics of agricultural land [5,6,7,8]. For farmers to survive and grow, land serves two purposes: first, as a farmer’s primary means of production [9,10,11] and livelihood, and second, as a capital resource that can be (even partially) transferred (sold or rented) and used to earn money to live on [12,13,14]. During China’s planned economy and the early stages of economic reform and “opening up”, land had the sole function of providing basic living conditions for farmers. With the recent acceleration of urbanization in China and the development of secondary and tertiary industries, the social environment of rural households is changing rapidly. As a result, income opportunities are diversifying [15,16,17,18]. In addition, some rural households are even abandoning farmland due to the current inadequate land supply and low land use efficiency [19,20,21]. As members of rural households gradually abandon agricultural production (either partially or completely) to work in towns and cities [22,23,24], the amount of idle farmland is increasing [25,26], and, unfortunately, this is a growing trend [5,8,27]. Economical, sustainable and well-planned land use is therefore essential [28,29].

Scholars from many nations and regions have shown increasing interest in the problems associated with farming livelihoods [30,31,32,33,34]. Livelihood capital can be classified as natural, physical, financial, social, and human [35]. Early research on livelihoods concentrated only on poverty [36,37,38]; in particular, attention was paid to income levels, consumption potential, and other aspects of life’s necessities. As research has progressed, income and consumption have fallen out of favour as the sole indicators of poverty, due in part to the large number of successful poverty alleviation practices and theoretical advancements [39,40,41].

The World Commission on Environment and Development initially presented the idea of sustainable livelihoods in the late 1980s. The term “sustainable livelihoods” primarily relates to means of subsistence, sources of income, and sustainable resources and capacities. For instance, Chambers and Conway claim that sustainable livelihoods largely pertain to people’s capacity to support themselves [35]. According to Scoones [38], a full livelihood maintenance system consists of the skills, resources (physical and social), and activities required to maintain life.

Numerous methods of analysing sustainable lifestyles have been developed as a result of the widespread usage of the term in studies of rural development and rural poverty [35,36,38]. Among them, the framework of the sustainable livelihoods approach established by the UK Department for International Development (DFID) is well known [35]. This framework comprises five main aspects: vulnerability context, livelihood capital, structures and processes, livelihood strategies, and livelihood outcomes. The vulnerability context, which is often employed as the basis for research, encompasses shocks, trends, and seasonal aspects that are unquestionably compelling. Human capital, natural capital, physical capital, financial capital, and social capital are all considered forms of livelihood capital. Structures and processes relate to modifications to organisational structures and policies that have an impact on livelihood. The steps individuals take to convert livelihood capital into livelihood outcomes are known as livelihood strategies, and the accomplishment of livelihood plans or objectives constitutes a livelihood result.

Research on sustainable farming family livelihoods now focuses on how to alter farming household livelihood models and improve farming household livelihood tactics [42,43,44]. Transforming farmers’ livelihood patterns requires a multifaceted approach that includes diversifying agricultural production, enhancing farmers’ skills and knowledge, bolstering farmers’ organisations and social networks, and improving infrastructure and market environments, all of which can help farmers improve their livelihoods, enhance their well-being [45], increase economic efficiency, and promote environmental sustainability [46,47]. Optimising livelihood strategies is another important means of promoting sustainable livelihoods among rural households, including enhancing their capital and resources, promoting social security and welfare, and establishing rural chains and rural cooperatives, all of which can increase farming incomes [48,49], improve the productive capacity and market competitiveness of rural households [50], increase labour productivity [51], and promote sustainable rural development [52].

In China, rural land transfer refers to the transfer of the right to use the land by the farmer, who has the contractual right to operate the land while keeping his contractual rights unchanged [53]. China first proposed land transfer in 1984 in Document No. 1 of the Central Government. The 2002 Law of the People’s Republic of China on Rural Land Contracting stipulates that “the land contracted by farmers can be transferred without changing its use”. In 2007, the Law of the People’s Republic of China on Property Rights stipulated that “the owner of the right to contract for land management has the authority to transfer the right to contract for land management by way of subcontracting, swap and transfer in accordance with the provisions of the rural land contract law”. The Opinions on Guiding the Orderly Transfer of Rural Land Management Rights to Develop Moderate-Scale Agricultural Operations, issued in 2014, proposed to set up three separate rights of “ownership, contracting, and management”, i.e., to insist on collective ownership of rural land, stabilise farmers’ contracting rights, and liberalise land management rights. On 1 March 2021, the “rural land management rights transfer management measures” were implemented. As stated in Article 6, the contractor has the right to decide independently within the contract period whether the land management rights are transferred, as well as the object, mode, and duration of the transfer, in accordance with the law.

By the end of 2020, China had basically completed the registration and certification of contracted land in rural areas, involving more than 200 million farming households and about 1.5 billion acres of contracted land. The area of family-contracted farmland in China has been transferred to more than 555 million acres, more than 30% of the confirmed contracted land [54], and land transfer has become an important means to realise moderate-scale agricultural operation and modernisation of agriculture and rural areas. The trend of land transfer is irreversible. Researchers are currently examining the problem of sustaining the livelihoods of farming households from the standpoint of land transfer. Some researchers are examining the current livelihoods of land-transfer farmers [55,56], investigating the causes of livelihood changes, and examining the issues and causes related to land transfer. Other researchers are examining particular populations within the land transfer situation, such as forced migrants [57,58,59,60], landless farmers [61,62], and the elderly [63]. Still, other researchers are examining the impact of land transfer on the livelihood capital of farmers’ livelihood assets in specific areas [64], livelihood diversification [65,66,67], and livelihood strategies [68,69,70].

The existing literature is of great relevance to this paper, but there are also the following shortcomings: First, the current literature has concentrated more on certain categories related to land transfer, such as the elderly, landless farmers, and forced migrants. Therefore, the findings of these studies are not generalisable. Second, from a research standpoint, the current literature has concentrated on changes in the status, capital, and vulnerability of land transfer in relation to farmers’ livelihoods, but there are currently no quantitative standards that can be used to determine whether these farmers’ livelihoods are sustainable. Third, most current quantitative studies focus on the marginal impact of land transfers and are unable to determine the influence they have on the ability of rural household farmers to support themselves.

Therefore, the current study collected data from a rural field study conducted in Hubei Province, China, in 2021 to assess rural households’ sustainable means of subsistence. The influence of land transfer on the sustainable lives of farmers in these households was examined using a multiple linear regression model, and the contribution of each factor to the sustainable livelihoods of these households was examined using the regression decomposition approach. The following questions are addressed in this paper: What relationship exists between the transfer of land and the ability of rural households to sustain their way of life? How much does the transfer of land affect rural households’ ability to maintain a sustainable standard of living? To what extent does land transfer contribute to the degree of livelihood sustainability among rural households?

The current study contributes in the following ways: In terms of research methodology, this study combined qualitative and quantitative research, created a livelihood sustainability indicator for land-transfer farmers, and used field research data to quantitatively measure the level of livelihood sustainability among land-transfer farmers. In terms of research design, the study used the regression decomposition method to investigate the impact from each factor on the sustainable livelihoods of rural households. In terms of research content, this study characterises the livelihood sustainability traits of land-transfer farmers and creates a sustainable livelihood framework for farmers based on changes in farmers’ livelihoods before and after land transfer.

The paper is organised as follows: Based on the sustainable livelihood framework, livelihood capital is classified into five categories: human, natural, financial, physical, and social. Land-transfer farmers in the province of Hubei are taken as an example. This takes into account the land transferred out and the surface area of the transfer, explores the change in farmers’ livelihoods after the transfer, systematically analyses the impact of the transfer on farmers’ sustainable livelihoods, and calculates the contribution of the transfer and other factors to the difference in farmers’ sustainable livelihood levels on this basis. Even though Hubei Province in China is the subject of this study, other developing or developed countries can benefit from the indicator system, the design of the theoretical analysis framework, and the research ideas presented here to enhance their land-transfer policies and ensure land-transfer farmers’ long-term viability.

2. Theoretical Framework and Research Hypothesis

According to the economic theory of property rights, farmers who take part in land transfers are basically giving up some of their ownership rights to the land they now possess, at least temporarily [71]. Along with the transfer of property rights, the amount, location, and environment of the land and the carrier of rights also change. This has a direct impact on how farmers live and the environment in which they live. It may also cause changes in how farmers work, the makeup of their household incomes, and how they interact with their neighbours and local relatives. This has an impact on farmers’ livelihood capital and results in new livelihood methods and tactics, which can change the results around livelihood and the type and status of livelihood capital [72,73].

Specifically, as shown in Figure 1, after land transfer, farmers suffer some loss of natural capital. This stems from the reduction in the size of the land under cultivation, and the decline in natural capital becomes more pronounced the larger the size of the transfer [74]. Physical capital related to farming operations, such as productive tools like farm machinery, is also indirectly implicated. Although the transfer out of farmland reduces farmers’ operating income, the value of farmland assets is revealed to bring a straightforward increase in property income-transfer fees, and the larger the scale of the transfer out, the higher the property income is obtained [75]. The decrease in natural capital increases the possibility of farmers working outside the farm and engaging in off-farm production [76]. Farmers are freed from heavy farming activities and have more time and energy to engage in off-farm production activities. Farmers’ income sources then become diversified and their income increases, which in turn leads to an escalation of consumption demand, and farmers will invest more in housing, durable goods, health, and education, leading to the accumulation of physical and human capital.

The sustainable livelihood levels of rural households are a manifestation of livelihood capital capacity. The accumulation of livelihood capital can enhance the sustainable livelihood levels of rural households as well as strengthen the sustainable development capacity of rural households. Therefore, farmers’ land-transfer behaviour enhances sustainable livelihood levels by altering working patterns, maximising revenue structures, and fostering social ties.

Based on the analysis outlined above, research hypotheses for this study were proposed as follows:

H1: 

Land transfer has a significant positive impact on the livelihoods of rural households that is stronger than without land transfer.

H2: 

The size of the land transferred has a significant positive impact on the livelihoods of rural households, and the larger the land transfer, the greater the impact on the livelihoods of rural households.

3. Materials and Methods

3.1. Data Sources

The data for this study were obtained from field research in rural Hubei Province, China, in 2021. In this study, a multi-stage sampling survey was applied according to a probability proportional to the size (PPS) of the population for each successive sampling unit [77]. After preparing the sampling frame with the municipalities with more land-transfer farmers in Hubei Province as the target population, the sampling unit in this case was defined as four levels of counties, townships, villages, and farmers according to the administrative planning. In the province, a total of two counties were sampled; each county sampled four townships, each township sampled six villages, and each village sampled 16 farm households. A total of 768 questionnaires were distributed, and 650 valid questionnaires were returned, showing an efficiency rate of 85%. The contents of the questionnaires included basic household information, land transfers, household production, household consumption, household livelihood capital, and so on. The questionnaire process was in the form of face-to-face interviews between the researcher and the farmers, and the answers were recorded while the interviews were conducted. The questionnaire was semi-structured, and the data were analysed later by setting dummy variables and assigning weights.

3.2. Variable Measures

3.2.1. Measure of the Sustainable Livelihoods of Rural Households

A sustainable livelihood index of the farming families involved in the land transfer was used as the explained variable. Firstly, the evaluation index system of sustainable livelihood capital for rural households is constructed. Natural, human, physical, financial, and social capital represent five categories of livelihood capital. According to the relevant studies of Sun et al. [78,79,80] and Zhang [81], this study used 14 representative variables to measure the livelihood level of farming families involved in land transfer.

Human capital is the combination of knowledge, skills, talents, and health status. Indicators of human capital include four in particular: the size of the labour force, the proportion of trained workers, the average educational attainment of the labour force, and the average level of health of the labour force. When we talk about physical capital, we are considering the kind of durable commodities that are always present in the household. The quantity of livestock owned and the dwelling area per capita represent the two indicators used to measure it. The flow of resources used to support livelihoods and associated services is referred to as natural capital. To measure it, the amount of cultivated land per person was used. Financial capital refers to the accumulation and flow required by rural households in the production and consumption processes. Four indicators, such as indebtedness, property income, wage income, and transfer income, were selected for measurement. The term “social capital” describes the value of a farmer’s position within the organisational structure of society, which includes three main aspects: whether they have friends or relatives who are civil servants; neighbourhood relations; and family relations.

Table 1. Sustainable livelihood indicator system, considered variables and corresponding weights.

Category 1: Human Capital (Workforce)
Variables	Variable modes or levels	Weights
Size	Number of household workers	0.0019138
Trained level	Skilled workers (% of total workforce)	0.0019337
Average education level	1 = not enrolled in school 2 = not completed primary school 3 = primary school 4 = junior high school 5 = high school or junior college 6 = college and above	0.0017825
Average health status	1 = very poor 2 = poor 3 = fair 4 = better 5 = very good	0.0018763
Category 2: Physical capital
Dwelling space	Square meter per capita	0.0015493
Number of livestock	Total number of pigs, cattle and sheep	0.0027027
Category 3: Natural capital
Cultivated land area	Acres per capita	0.004167
Category 4: Financial capital
Liabilities	Yuan (log)	0.0067573
Property income	Yuan (log)	0.0086177
Wage income	Yuan (log)	0.0031548
Transferable income	Yuan (log)	0.0029146
Category 5: Social capital
Neighbourhood relations	1 = very good, 2 = better, 3 = fair, 4 = worse, 5 = very bad	0.0033129
Whether you have friends or relatives of civil servants	1 = yes, 0 = no	0.0019646
Family and friend relationship	Number of households visiting each other during the Chinese New Year	0.0018819

presents the specific indicators.

Then, a weight was assigned to each selected variable explaining the sustainable livelihood index of rural households; these weights were calculated using the entropy weighting technique [82,83,84] as compared to the subjective empirical assignment technique (detailed steps of the entropy weighting technique can be found in Appendix A.1 in Appendix A); this method can avoid the interference of human factors in the subjective assignment method and solve the problem of redundant information among multiple indicators. More importantly, the method is transparent and reproducible [85].

3.2.2. Measurement of Land Transfer

Land transfer includes the direction and area of the transfer. Referring to the studies of Yang et al. [55], this study selected two core explanatory variables, whether land is transferred out and the size of the land transferred, to measure land transfer.

3.2.3. Measurements of Control Variables

Azumah et al. [86] showed that the gender of farmers is related to sustainable livelihoods. Amjed et al. [87] studied the sustainability of non-farming incomes and the livelihoods of rural households. Additionally, Zhao et al. [74] found that farming household livelihoods are influenced by additional factors such as the education level of the household head and the size of the household. Therefore, this study drew on existing studies and selected variables that may affect the sustainable livelihoods of rural households in terms of their personal characteristics and household characteristics as the control variables. Among them, personal characteristics were characterised by the gender, age, education level, and health status of the household head. Household characteristics were described by the size of the respondent’s household and the proportion of household workers. The definitions of the specific variables are detailed in

Table 2. Explained, explanatory and control variables.

Category	Variable Name	Definition
Explained variables	Sustainable Livelihood Index	Sustainable Livelihood Score Index
Explanatory variables	Land transfer out or not	1 = yes, 0 = no
	The size of the land transferred	Acres
Control variables	Gender of the head of household	1 = Male, 0 = Female
	Age of the head of household	Actual age of head of household
	Education level of household head	1 = not enrolled in school, 2 = not graduated from elementary school, 3 = elementary school, 4 = junior high school, 5 = high school or junior college, 6 = college and above
	Health status of the head of household	1 = very poor, 2 = poor, 3 = fair, 4 = better, 5 = very good
	Family size	Total number of family members
	Proportion of family workers	The proportion of outworkers to the total number of workers

.

3.3. Model Setting

The sustainable livelihood index is used as the explained variable; the land transfer out or not and the size of the land transferred selected in this paper are used as explanatory variables; and the gender, age, education level, and health status of the household head, the size of the household, and the proportion of household workers are selected as control variables to construct the model of this paper:

$$\mathrm{Y}=\mathsf{\alpha }+{\mathsf{\beta }}_1{\mathrm{X}}_1+{\mathsf{\beta }}_3{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i}}+\mathsf{\epsilon }$$

(1)

$$\mathrm{Y}=\mathsf{\alpha }+{\mathsf{\beta }}_2{\mathrm{X}}_2+{\mathsf{\beta }}_3{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i}}+\mathsf{\epsilon }$$

(2)

where $\mathrm{Y}$ is the sustainable livelihood index, $\mathsf{\alpha }$ is a constant, ${\mathrm{X}}_1$ indicates whether the land is transferred out or not, ${\mathrm{X}}_2$ means the amount of land transferred, and ${\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i}}$ stands for a series of control variables according to Table 2. Finally, $\mathsf{\epsilon }$ denotes the random error term.

3.4. Robustness Tests

3.4.1. Robustness Test of Land Transfer out on the Sustainable Livelihoods of Rural Households

This paper uses the propensity score matching method to test the robustness of the impact of land transfer on the sustainable livelihoods of rural households. Regarding the problem studied in this paper, the reason for the higher level of sustainable livelihoods of rural households may not be the participation of rural households in land transfer out, but may be due to the fact that the head of the household is male, their age is lower, they have a higher education and better health, the size of the farm household is smaller, and the proportion of family workers is higher, which is more likely to cause farmers to make the decision to land transfer out. Therefore, in order to eliminate the interference of the sample self-selection problem on the results of this paper, the differences in personal and family characteristics among the final sample farmers are eliminated by matching the samples of whether or not to participate in land transfer according to the personal characteristics of the household head, such as gender, age, education level, and health status, and other household characteristics, such as the size of the farmer’s family and the proportion of household workers in the first stage. The final sample is then obtained based on the 1:3 nearest neighbour matching method, and model (1) is used for the robustness tests using these sample data.

3.4.2. Robustness Test of the Size of the Land Transferred on the Sustainable Livelihoods of Rural Households

In this study, the Tobit model is used to test the robustness of the effect of the size of the land transferred on the sustainable livelihoods of rural households. Since the explained variable is the sustainable livelihood index and the data are between 0 and 1, the problem of the regression relationship between the size of the land transferred and sustainable livelihood is a truncated regression problem, and the Tobit model can deal with the above data and effectively avoid the problem of inconsistency and bias in parameter estimation. Therefore, this paper uses the Tobit model for the robustness test.

3.5. Regression Decomposition Method

The estimation results of Equations (1) and (2) reflect only the marginal effects of each variable on the sustainable livelihoods of rural households. However, the contribution of each of the variables, such as whether or not land was transferred out; the size of the land transferred; the household head’s gender, age, education level, and health status; and the size of the household and the share of household labour, to the difference in sustainable livelihoods of farm households, and which factor is the main contributor to this difference, need further study. This study used regression decomposition methods to measure the relative contribution of each variable to the sustainable livelihoods of rural households. Based on the estimates from Equations (1) and (2), the study measured the contribution of each variable to the difference in the sustainable livelihoods of rural households using the improved regression decomposition method of Wan [88]. The detailed steps are described in Appendix A.2 in Appendix A.

4. Results

4.1. Descriptive Statistics

The outcomes of descriptive statistics for the key variables are shown in

Table 3. Descriptive statistics of key variables in the sample of 650 households.

Variables	Mean	Standard Deviation	Min	Max
Sustainable Livelihood Index	0.176	0.073	0.038	0.503
Land transfer out or not	0.271	0.445	0	1
The size of the land transferred	1.189	2.558	0	14
Gender of the head of household	0.888	0.316	0	1
Age of the head of household	59.157	11.280	23	87
Education level of household head	3.149	1.220	1	6
Health status of the head of household	3.558	1.218	1	5
Family size	4.569	1.974	1	11
Proportion of family workers	0.007	0.053	0	0.600

. The mean value of the explained variable, the sustainable livelihood index, was 0.176, indicating that its mean value in the sample was 0.176. Whether land was transferred out and the area transferred had mean values of 0.271 and 1.189, respectively. These were explanatory factors, and they demonstrated that approximately 27.1% of the farmers in the study sample transferred their land, and the sample’s average transferred area value was 1.189 acres.

4.2. Impact of Land Transfer out or Not on the Sustainable Livelihoods of Rural Households

4.2.1. Univariate Test

Table 4. The univariate tests of land transfer out or not.

Variables	Land Not Transferred Out 1		Land Transfer Out 2		Z-Test
	Mean	Median	Mean	Median
Sustainable Livelihood Index	0.162	0.150	0.200	0.198	34.849 ***
Gender of the head of household	0.877	1.000	0.922	1.000
Age of the head of household	58.154	58.000	57.972	59.000	0.131
Education level of household head	3.099	3.000	3.101	3.000	1.468
Health status of the head of household	3.591	4.000	3.419	3.000	0.248
Family size	4.626	5.000	4.391	4.000	1.062
Proportion of family workers	0.005	0.000	0.010	0.000	0.890

¹ The number of households that have not been transferred out of their land is 474. ² The number of households whose land has been transferred out is 176. Note: *** indicate significance at the 1% levels.

reports the results of the univariate tests of land transfer or not. Among them, the sustainable livelihood level of farmers who transferred their land was significantly higher than that of farmers who did not transfer their land.

4.2.2. Main Test Results and Analysis

We empirically tested whether land transfer affects the sustainable livelihood of rural households according to model (1). The results for this are shown in

Table 5. Regression results from the modelling of land transfer out or not and sustainable livelihoods of rural households.

Variables	(1)	(2)
	Sustainable Livelihood Index	Sustainable Livelihood Index
Land transfer out or not	0.039 ***	0.036 ***
	(5.847)	(5.883)
Gender of the head of household		−0.013
		(−1.415)
Age of the head of household		−0.001 ***
		(−3.140)
Education level of household head		0.001
		(0.259)
Health status of the head of household		0.005 **
		(2.424)
Family size		−0.016 ***
		(−10.941)
Proportion of family workers		0.177 ***
		(3.474)
constant	0.166 ***	0.277 ***
	(53.325)	(12.657)
N	650	650
r2_a	0.056	0.251

t statistics in parentheses. ** p < 0.05, *** p < 0.01.

. Column (1) shows the results of whether land transfer out or not affects farmers’ sustainable livelihoods without controlling for other factors, and column (2) illustrates the results after adding the control variables. The magnitude of the effect on the sustainable livelihood level of rural households was 3.6%.

4.3. Impact of the Size of the Land Transferred on the Sustainable Livelihoods of Rural Households

We investigated whether the size of the land transferred has an impact on the ability of farming families to continue their way of life, as assumed by model (2). The findings are provided in

Table 6. Regression results of modelling the size of the land transferred and sustainable livelihoods of rural households at different scales.

Variables	(1)	(2)	(3)	(4)	(5)
	Sustainable Livelihood Index	Sustainable Livelihood Index	Small-Scale	Medium-Scale	Large-Scale
The size of the land transferred	0.007 ***	0.006 ***	0.005	0.024 **	−0.001
	(4.624)	(4.601)	(0.310)	(2.385)	(−0.201)
Gender of the head of household		−0.012	0.013	−0.021	−0.048
		(−1.297)	(0.223)	(−0.699)	(−1.161)
Age of the head of household		−0.001 ***	0.002	−0.001	−0.003 **
		(−3.420)	(1.456)	(−1.025)	(−2.587)
Education level of household head		0.000	0.017	0.005	−0.015
		(0.115)	(1.607)	(0.630)	(−1.622)
Health status of the head of household		0.004 **	0.008	0.004	0.013
		(2.120)	(1.348)	(0.726)	(1.057)
Family size		−0.016 ***	−0.020 ***	−0.018 ***	−0.026 ***
		(−10.736)	(−2.865)	(−3.842)	(−4.095)
Proportion of family workers		0.161 ***	0.025	0.192 *	0.363 ***
		(3.324)	(0.269)	(1.960)	(4.630)
constant	0.169 ***	0.286 ***	0.084	0.232 ***	0.526 ***
	(56.104)	(13.055)	(0.823)	(2.761)	(6.046)
N	650	650	60	67	49
r2_a	0.050	0.243	0.166	0.268	0.315

t statistics in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

. Without accounting for the effects of other variables, the results are shown in column (1), followed by the results after the addition of the control variables in column (2), and the differential effects of the size of the land transferred on the sustainable livelihoods of transfer households of various sizes in columns (3) to (5). It can be seen that after taking into account the possible impact of other factors on the sustainable livelihoods of rural households, the degree of impact of area transferred on the sustainable livelihoods of farmers decreased. This was still statistically significantly positive at the 1% level, with an average increase of 0.6% in the sustainable livelihoods of rural households for every 1 unit increase in the area of transfer.

The positive effect of the area transferred on the sustainable livelihoods of medium-sized transferring households was observed. The underlying mechanism may be that this scale of farming is more risk-resistant and less vulnerable to livelihoods than large-scale transfer households, and that, compared with large- and small-scale transfer households, medium-scale transfer households are able to engage in both agricultural work and non-farming production. Further, their sources of income may be diversified and the labour capacity of family members fully utilised. This may enable senior individuals who are still capable of productive work to earn income through labour while allowing young adults to make money by engaging in off-farm production. At the same time, medium-scale streaming households have social ties in both urban and rural areas, thus increasing livelihood sustainability.

4.4. Robustness Tests Results

4.4.1. Robustness Test of Land Transfer out on the Sustainable Livelihoods of Rural Households

The final 440 sample was obtained based on the 1:3 nearest neighbour matching method, and the results shown in

Table 7. Land transfer out or not and sustainable livelihoods of rural households under Propensity Matching Score approach.

Variables	(1)	(2)
	Sustainable Livelihood Index	Sustainable Livelihood Index
Land transfer out or not	0.037 ***	0.033 ***
	(5.077)	(5.171)
Gender of the head of household		−0.017
		(−1.307)
Age of the head of household		−0.001 **
		(−2.331)
Education level of household head		0.000
		(0.112)
Health status of the head of household		0.007 **
		(2.551)
Family size		−0.017 ***
		(−9.624)
Proportion of family workers		0.101
		(1.378)
_cons	0.167 ***	0.282 ***
	(37.209)	(9.968)
N	440	440
r2_a	0.053	0.254

t statistics in parentheses. ** p < 0.05, *** p < 0.01.

were obtained after empirical testing using this sample data via model (1). This clearly showed that land transfer out or not still had a significant enhancing effect on farmers’ sustainable livelihoods, and the conclusions of this study are therefore solid.

4.4.2. Robustness Test of the Size of the Land Transferred on the Sustainable Livelihoods of Rural Households

In this study, we re-ran the regression using the Tobit model, and the outcomes are displayed in

Table 8. Tobit model test.

	(1)	(2)
	Sustainable Livelihood Index	Sustainable Livelihood Index
The size of the land transferred	0.007 ***	0.006 ***
	(5.97)	(5.80)
Gender of the head of household		−0.012
		(−1.44)
Age of the head of household		−0.001 ***
		(−3.33)
Education level of household head		0.000
		(0.12)
Health status of the head of household		0.004 **
		(2.06)
Family size		−0.016 ***
		(−11.93)
Proportion of family workers		0.161 ***
		(3.20)
Constant	0.169 ***	0.286 ***
	(54.75)	(13.05)
Observations	650	650
F test	3.89 × 10−9	0
r2_a	.	.
F	.	.

t-statistics in parentheses. *** p < 0.01, ** p < 0.05.

. The size of the land transferred was found to have less of an influence on farmers’ sustainable livelihood levels after other factors that may affect such livelihoods were taken into account. This, however, still significantly increased farmers’ sustainable livelihood levels at the 1% level, which is consistent with the results of the previous model. Therefore, the conclusions of the current study are reliable.

4.5. Results of Regression Decomposition Method

The results of the regression decomposition of each variable are shown in

Table 9. Regression decomposition results.

Variables	Coefficient of Variation	Contribution Rate (%)
Land transfer or not	0.1581	12.4158
The size of the land transferred	0.0782	10.0976
Gender of the head of household	−0.0563	1.6247
Age of the head of household	−0.1210	6.4775
Education level of household head	0.0086	0.4490
Health status of the head of household	0.0822	2.5952
Family size	−0.4404	63.6356
Proportion of family workers	0.1223	2.7045
All	−0.1683	100

. From the results of the regression decomposition, whether land was transferred out or not, the size of the land transferred and household size were the main variables that contributed to the total variation in the sustainable livelihoods of rural households, while the contribution of other variables was found to be low. In terms of the positive coefficient of variation, land transfer or not and area transferred were the main variables affecting the sustainable livelihood of rural households, with 12.42% and 10.10%, respectively. Possible reasons for this are that land transfer can increase farming income and enable the efficient use of agricultural land, which can provide more opportunities for farmers to enter other fields of work [89]. This can lead to increases in the total financial capital of rural households. As farmers’ incomes increase, their investment in durable consumer goods, housing, and children’s education increases accordingly, which in turn leads to an increase in physical, social, and human capital. Overall, after land transfer, rural households change their livelihood strategies based on their existing livelihood capital status to adapt to the impact of land transfer and ensure their livelihoods are more sustainable.

It is noteworthy that for the negative coefficient of variation, the impact of household size on the sustainable livelihoods of rural households was larger, with a contribution rate of 63.64%. This may be explained by the fact that as household size increases, more arable land and resources are needed to sustain livelihoods, and insufficient arable land area and resources can lead to increased pressure on arable land. This affects the yield and quality of crops, thus affecting the sustainability of the livelihoods of rural households. At the same time, if the household size is too large and the household labour force is insufficient, this may lead to unworkable or inefficient agricultural production, making sustainable development difficult to achieve.

5. Discussion

Land transfer has a considerable favourable impact on the ability of rural households to continue their way of life, both with and without the addition of control factors. This supports Research Hypothesis 1, set out in this paper. This finding both converges and diverges from the findings of existing studies, where Zhao et al. [74] found a positive net effect of farmland transfer behaviour on the total index of rural households’ livelihood capital, and Wu & Wu [71] also concluded that through participation in homestead transfer, farmers’ livelihood capital was accumulated to a certain extent.

However, the results of this study differ from those of Zhang [81], who contended that farmers who receive land transfers are only able to maintain their current levels of subsistence but are still some way off from achieving the ideal of sustainable development. Possible explanations for this include the fact that after land is transferred out, farmers lose ownership of their property, which can erode their land rights and threaten the stability of their livelihoods. This is true even though they can diversify their income sources by engaging in non-farm production. At the same time, the production scale of farmers is restricted and risks increase, making it possible to maintain only their original livelihood level rather than growing it.

There was a substantial positive effect of the size of the land transferred on farmers’ sustainable livelihoods, both without and after including the control factors. This finding supports hypothesis 2 of this study. Similar to findings in earlier studies, the expansion of the size of the land transferred was found to enhance the sustainability of farmers’ livelihoods, and land transfer has a beneficial effect on resolving issues with low land utilisation, land fragmentation, and barriers to the adoption of new technologies and labour mobility brought about by smallholder production [90]. This may be connected to the ability of land transfer to maximise the use of rural resources, advance modern agriculture, raise the standard of the agricultural sector, and advance farmers’ income growth and lifestyles [91].

Unlike previous studies, this study referred to Wan’s [88] study and used his improved regression decomposition method to measure the contribution of each variable to the difference in sustainable livelihoods of rural households. In an area comparable to that used in the present study, Yang et al. [55] examined the relationship between livelihood capital and land transfer for different types of farmers in a sustainable livelihood framework. Additionally, Liu et al. [92] investigated the differences in changes in farmers’ livelihood capital under different land-transfer practices. Together, their results revealed how land transfer and farming livelihoods are interconnected, further providing evidence that the two subjects cannot be examined independently.

The current study has some limitations. First, the sample used in the study was limited to 650 rural households in Hubei Province. To address this, future research should be conducted across different regions. Second, both land transfers in and out are considered directions of land transfer. The current study, however, focused on the impact of land transfer out on the sustainable livelihoods of rural households, while the impact of land transfer in on the sustainable livelihoods of rural households was not covered. Future research should therefore examine the sustainable lifestyles of rural households by combining land transfers in and out.

Compared with previous studies, the current study contributes in the following ways: Specifically, this study used a combination of qualitative and quantitative methods, created an indicator system to assess the livelihood sustainability of land-transfer farmers, and employed data from field research to assess the livelihood sustainability of land-transfer farmers on a quantitative level. In terms of the study’s content, it described the sustainable livelihood traits of post-land-transfer farmers, built a sustainable livelihood framework for farmers based on the variations in farmers’ livelihoods preceding and following land transfer, and localised and implemented the sustainable livelihood analysis framework to create a sustainable agriculture program. The regression decomposition method was used as the study’s research design to examine the extent to which each factor affects the sustainability of rural households’ livelihoods.

6. Conclusions

The current study used cross-sectional data from a sample of 650 households in Hubei Province, China. It adopted a multiple linear regression model combined with regression decomposition to measure the differential contribution of each factor to the sustainable livelihoods of farmers. The following conclusions were drawn from the analysis of the results: First, land transfer contributes to improvements in the sustainability of livelihoods among rural households. Second, the larger the area of land transferred, the higher the level of sustainability in the livelihoods of rural households. Third, land transfer out, the size of the land transferred, and farm household size represent the main factors contributing to differences in the level of sustainability in the livelihoods of rural households in Hubei Province.

The third finding differs from all previous studies. One reason for this could be that some current studies that measured the impact of land transfer also measured the marginal impact of land transfer without comparing the magnitude of the contribution of each factor. Of course, land transfer may not be the most important determinant of the increased sustainable livelihood levels of rural households in other regions, but this does not prevent land transfer from being an important factor that impacts the level of sustainability of rural households’ livelihoods.

Based on the analysis presented above, this study includes three policy implications. The first of these is the need to strengthen land transfer policies. Land transfer management needs to be upgraded with the aim of guaranteeing the reasonability and sustainability of land transfers. The government should establish a land transfer management system, regulate land transfer behaviour, strengthen the supervision of land transfer, and protect the legitimate rights and interests of farmers. At the same time, it should also strengthen land-use planning and reasonably plan land use to avoid the wastage of land resources and duplicate construction.

The second is the need to develop agricultural industrialisation. In order to increase farmers’ incomes and reduce risks, agricultural industrialisation needs to be developed. The government should formulate agricultural industrialisation policies and support agricultural industrialisation enterprises to drive rural economic development. At the same time, farmers should be encouraged to organise into agricultural cooperatives to understand the scale and intensification of agricultural production.

Third, the flow of rights in terms of land-contract management should be promoted. To protect land resources and the livelihoods of farmers, it is necessary to promote the flow of these rights. The government should strengthen how land-contract management rights and transfers are regulated, as well as encourage farmers to transfer their idle land. At the same time, it is also important that they safeguard the land-contract management rights of farmers. Additionally, the management of land-transfer contracts should be reinforced to ensure their legality and enforceability.