Does Farmers’ Participation in Skills Training Improve Their Livelihood Capital? An Empirical Study from China

Abstract

With the rapid development of China’s rural economy, rural collective economic organizations have played a significant role in increasing farmers’ income and promoting rural revitalization. This study aims to explore the impact of farmers’ participation in skills training organized by rural collective economic organizations on their livelihood capital and the underlying mechanisms. Using tracking survey data from rural households in Chongzhou City, Sichuan Province, in 2023, the paper employs empirical analysis methods, including OLS and mediation effect analysis. The results show that participation in skills training organized by rural collective economic organizations enhances farmers’ livelihood capital. The effectiveness of public service delivery by village committees, the network role of farmers’ cooperatives, and the linkage effect of leading agricultural enterprises in driving rural development act as mediating factors. Additionally, the impact of skills training on farmers’ livelihood capital varies according to household characteristics and the attributes of the rural collective economic organizations involved. Therefore, this paper proposes the following policy recommendations: (1) Further strengthen the public service and resource allocation functions of the village committees. (2) Support and optimize the operating entities such as farmer cooperatives and leading agricultural industrialization enterprises. (3) Address the training needs of different groups and enhance the focus and support of grassroots governments for skills training in collective economic organizations.

1. Introduction

Agriculture, rural areas, and farmers’ issues have long been key concerns for both the Party and the State, representing urgent challenges that require resolution [1]. These issues are fundamental to the comprehensive construction of a modern socialist country and the realization of Chinese-style modernization. Since 2014, the No. 1 Central Document has consistently called for the “strengthening of rural collective economies and broadening their development paths” for nine consecutive years. In June 2024, the “Law of the People’s Republic of China on Rural Collective Economic Organizations” was officially promulgated. This law explicitly states that rural collective economic organizations must provide services such as technical support and information to assist members’ production and business activities. Additionally, the law emphasizes that these organizations should support and collaborate with village committees in implementing self-governance under the leadership of the village Party organization, while also playing an active role in supporting other rural economic and social organizations in accordance with the law [2,3]. Under the guidance of this law, rural collective economic organizations will be better positioned to provide social services and improve infrastructure, enabling members to benefit from the services and welfare provided, thereby improving their livelihood capital.

Chongzhou City, as the first county-level area in the province to implement a performance evaluation system for collective economic organizations, stands as a model of rural collective economies, particularly in driving agricultural integration, promoting economic development, and establishing exemplary cases in areas such as sustainable income generation for farmers, industrial advancement, and enhanced governance efficiency [4]. To further improve farmers’ livelihood capabilities and promote rural common prosperity, Chongzhou City has, based on the genuine needs of its farmers, provided tailored skills training through the village Party committees and local collective economic organizations. These initiatives have significantly contributed to the economic and social benefits of the adage “teaching a man to fish is better than giving him a fish”. Specifically, the skills training offered by rural collective economic organizations in Chongzhou City encompasses a wide array of subjects, including crop cultivation techniques, animal husbandry, pest and disease control, soil improvement, the cultivation of new crops, and the maintenance and operation of agricultural machinery. Unlike other forms of social skills training, the rural collective economic organizations in Chongzhou City monitor and assess the entire learning and practical process of farmers, providing visual evaluations of their proficiency in various skills. For farmers facing learning difficulties, customized support plans are created, with designated personnel providing step-by-step, individualized training to help manage and enhance their sustainable development capacity.

Undoubtedly, rural collective economic organizations, as a vital force in rural economic development, possess significant potential to enhance farmers’ livelihood capital. Through the training and guidance provided by these organizations, they facilitate the growth of various forms of livelihood capital—such as natural, material, human, social, and financial capital. This process is a crucial avenue for farmers to achieve self-development and serves as an essential strategy for advancing rural modernization and promoting common prosperity [5]. The objective of this study is to examine the impact of farmers’ participation in skills training organized by rural collective economic organizations on their livelihood capital, using micro-survey data from households in pilot villages of the “Rural-Community Integration” project in Chongzhou City, Sichuan Province, collected in 2023, as the empirical basis. The significance of this study and the issues it aims to address include: analyzing the impact of skills training by rural collective economic organizations on farmers’ livelihood capital, providing empirical evidence to clarify their interrelationship; exploring the mechanisms through which skills training enhances farmers’ livelihood capital, thereby offering a theoretical basis and practical guidance for targeted reforms and strategic development of these organizations; and assessing the varied principles by which skills training can enhance the livelihood capital of farmers with distinct characteristics, providing a guide for improving their livelihood capital.

2. Literature Review

The livelihood activities of farm households are mainly expressed as the means or ways of acquiring basic material resources, and the livelihood capital of farm households includes the labor capacity possessed by individuals, the assets needed for family life, and the social activities in which individuals participate, which encompasses the various resources and capacities used by farm households to maintain and improve their living conditions, which is not only the key to the sustainable development of the countryside, but also an important part of the improvement of the living conditions of the households [6]. There are many factors affecting the livelihood capital of farm households, most of which are included in the livelihood capital framework proposed by the UK Department for International Development (DFID), which mainly centers on the five dimensions of natural capital, physical capital, human capital, financial capital, and social capital [7]. For example, the withdrawal of homesteads can improve the livelihood capital of farm households in terms of the level of physical and social capital [8]. The increase in the scale of transferring out farmland can improve the physical and financial capital of farm households, but it will make the level of natural and social capital of farm households decrease [9]. The increase in family size, on the other hand, favors the enhancement of natural, financial, human, and social capital of farm households [10]. The development of rural lodging leads to a decline in the natural capital of farm households but leads to an increase in all the rest of the capital [11].

As an important part of China’s rural development, rural collective economic organizations, through the integration of resources, improve production efficiency, increase farmers’ income, help to narrow the gap between the rich and the poor, drive the balanced development of rural society, and promote the realization of the goal of common prosperity. Different scholars have different understandings of rural collective economic organizations, and most scholars understand rural collective economic organizations as economic organizations that take collectively owned land as the basic means of production and implement a two-tier management system based on family contracting and combining unification and division [12,13]. However, due to policy support, rural collective economic organizations are also evolving, which brings about in-depth research in the academic community on the supervisory and management mechanism of rural collective economic organizations [14,15], the criteria for identifying the members [16,17], and the status of the legal person [18,19] and other aspects. Under the general trend of reform and development, each rural collective economic organization, relying on its resource advantages, has played an active role in driving farmers to increase their incomes by formulating development plans in line with local conditions [20,21], which has given rise to a number of studies on the role of rural collective economic organizations in promoting the common prosperity of farming households [22,23].

Skills training carried out by rural collective economic organizations is one of the important ways to improve the endogenous development power of farm households, promote comprehensively increasing the income farm households, and realize the common prosperity of farmers and rural areas [24]. With reference to academic research, skills training carried out by rural collective economic organizations can solve the problem of insufficient vocational skills of low-income groups in rural areas [25]; it can enhance the level of human capital of farm households, thus expanding their livelihood transition capacity and livelihood space [26], and it can increase the economic support for the old age farm households [27]. At the same time, it reduces farmers’ dependence on traditional agriculture and expands their sources of income [28], which in turn ensures the stability of non-farm employment income [29]. In addition, it can also help farmers understand the importance of environmental protection and sustainable agricultural practices, so that they can learn how to utilize natural resources more rationally and conserve energy [30], thus narrowing the ability gap between rural and urban populations [31].

In conclusion, despite the wealth of research on rural collective economic organizations and livelihood capital, as well as extensive studies on the impacts of skills training on farmers, integrating these elements within a unified theoretical and empirical framework remains largely unexplored. Given the critical importance placed on enhancing the capabilities of rural collective economic organizations and exploring their diversified development in China, investigating the effects of skills training conducted by these organizations on farmers’ livelihood capital is not only theoretically relevant but also vitally practical. This study uses Chongzhou City, Sichuan Province as a representative case, employing survey data to conduct an empirical analysis based on the livelihood capital assessment framework.

3. Research Analysis and Hypotheses

Amidst the rapid development of rural collective economic organizations, the implementation of skills training has become a vital mechanism for enhancing the skill levels of farmers and increasing their income. Yet, questions remain about whether such training can elevate farmers’ livelihood capital and the specific channels or pathways through which this enhancement occurs, necessitating further investigation and validation.

According to human capital theory, human capital can be categorized into affective and cognitive types, with cognitive human capital typically obtained through formal education or skills training [32]. Participation in skills training conducted by rural collective economic organizations enables farmers to learn new skills, directly boosting their cognitive human capital. Human capital comprises two main types: productive and non-productive capital, with non-productive capital primarily focused on planning and decision-making abilities [33]. Skills training helps farmers acquire new knowledge, enhancing their decision-making skills, and thereby increasing their non-productive human capital. Social capital, a critical informal institution in rural areas, revolves around the Chinese concept of “guanxi,” or social networks [34]. From the viewpoint of political science and social organizations, social capital is the capital formed by supportive behaviors and norms among group members [35]. Participation in skills training strengthens trust and network relationships within the community, directly boosting farmers’ social capital and thus reinforcing their individual livelihood capital. Given these insights, the study posits the following hypothesis:

H1: 

Skills training by rural collectives boosts farm households’ livelihood capital and well-being significantly.

Under the “two brands, one team” rural collective economic development model prevalent in Chongzhou City, the implementation of skills training by rural collective economic organizations is profoundly enhancing the managerial capabilities and decision-making efficiency of village committee members. Specifically, specialized training in financial management, project management, and public relations for members of these organizations significantly improves the effectiveness of village committees in executing their economic development roles. According to the theory of health needs, education and health are vital components of human capital and are also fundamental to livelihood capital. Education enhances productivity, whereas health capital is crucial for maintaining employable hours in the labor market [36]. The bolstered capacity of village committees to manage rural collective economic development is apparent in their ability to devise customized skills training programs that increase the knowledge and skill levels of farmers. Additionally, it is reflected in the emphasis placed on regular healthcare engagement among members of the collective economic organizations, thereby elevating the health status of farmers. Thus, the amplified effectiveness of village committees in their economic functions significantly boosts the farmers’ livelihood capital. Based on these findings, this paper proposes the following hypothesis:

H2: 

Skills training by rural collectives enhances farmers’ livelihood capital through improved village governance.

Farmer Professional Cooperatives, as voluntary mutual aid economic organizations formed by farmers, are open to membership beyond the local community without being restricted by administrative boundaries. Members might engage in skills training facilitated by rural collective economic organizations, although some may be precluded due to time or geographic constraints. Those who participate in this training gain additional insights into organizational management and team collaboration, enhancing the cooperative’s internal management efficiency and decision-making quality, which in turn ensures a more robust operation within external environments and fosters sustainable development of the cooperative. According to social network theory, people in social contexts form networks through their relationships and leverage these networks for the exchange of information and resources [37]. Social capital theory suggests that stronger ties within the network yield greater social capital [38]. Consequently, Farmer Professional Cooperatives enable farmers who have undergone skills training to establish connections with a broader network of peers, suppliers, customers, and other stakeholders, creating platforms for information exchange and resource trading, thus enhancing the livelihood capital of these trained farmers. Based on these considerations, the study proposes the following hypothesis:

H3: 

Skills training by rural collectives enhances farmers’ livelihood capital through the improved social networks of farmers’ professional cooperatives.

Skills training offered by rural collective economic organizations also targets emerging business entities within the jurisdiction, including leading agricultural industrialization enterprises, family farms, and professional cooperatives. Through targeted training in areas such as business management, strategic planning, and market analysis, this training enhances the management capacity and operational efficiency of these leading enterprises, improving their business conditions and market competitiveness. As a result, these enterprises can create more job opportunities for farmers, raise their non-agricultural income, and strengthen the profit-sharing mechanisms between enterprises and farmers. Additionally, as the operational performance and market position of these leading enterprises improve, they typically gain easier access to financial resources, which can benefit farmers through mechanisms like supply chain finance. This, in turn, enhances farmers’ financial capital, further strengthening their livelihood capital. Based on these findings, the paper proposes the following hypothesis (Figure 1):

H4 

: Skills training by rural collectives enhances farmers’ livelihood capital through strengthening the role of leading agricultural industrialization enterprises in linking up with farmers.

4. Study Design and Variable Descriptions

4.1. Data Sources

The data analyzed in this study is derived from a micro-survey conducted in 2023 in the “Village-Community Integration” pilot villages in Chongzhou, Sichuan Province. The survey targeted a variety of stakeholders, including members of rural collective economic organizations, officials from village committees, leaders of agricultural industrialization enterprises, family farms, and cooperatives—representing emerging business entities. The survey was carried out across eleven towns and streets, including Liaojia, Baitou, Guansheng, and Daoming, all part of the “Village-Community Integration” initiative. A total of 510 questionnaires were distributed to members of rural collective economic organizations, with 418 returned. After careful examination of the responses, no significant data omissions or logical inconsistencies were found, resulting in 418 valid samples and a 100% response validity rate. Among the respondents, males slightly outnumbered females, with a near 1:1 male-to-female ratio. In terms of age distribution, nearly 70% of the respondents were aged 46 or older. Regarding educational attainment, over 60% of the respondents had no higher than a middle school education, reflecting the generally low educational level in rural Sichuan. Thus, the sample is considered to be highly representative of the local population (

Table 1. Basic characteristics of agricultural households.

Basic Characteristics of Farm Households		Sample Size	Proportion (%)
Gender	man	224	53.59
	woman	194	46.41
Age	18–25 years	12	2.87
	26–35 years	55	13.16
	36–45 years	73	17.46
	46–55 years	117	27.99
	56+	161	38.52
Highest level of education	uneducated	9	2.15
	secondary schools	71	16.99
	junior high school	175	41.87
	high school	86	20.57
	vocational school	27	6.46
	university and above	50	11.96

). The subsequent data analysis was completed using Stata 18.

4.2. Variables

4.2.1. Dependent Variables

Livelihood capital is selected as the dependent variable in this study. In terms of human capital, factors such as education level, skills, and health determine a farmer’s competitiveness in the labor market [39,40]. In the domain of social capital, a farmer’s social network, degree of organizational participation, and community support are crucial for accessing resources and information, as well as for mitigating risks [41,42]. Regarding financial capital, improvements in personal income, profit dividends from collectives, and the ability to secure loans provide essential financial support for the farmer’s livelihood and development [43,44]. In terms of physical capital, public facilities, housing, and water infrastructure owned by the collective to which the farmer belongs form an important material foundation for their life [45,46,47]. With respect to natural capital, access to more natural resources, such as land and water, offers farmers greater livelihood opportunities, enabling them to escape relative poverty [48]. Based on the aforementioned quantitative studies on livelihood capital by both domestic and international scholars, and considering the availability of data, the indicator system in these studies has been adjusted. A set of livelihood capital measurement indicators specifically tailored to local farmers has been developed (

Table 2. Measurement indicators, assignments, and weights of farmers’ livelihood capital in Chongzhou City.

Livelihood Capital Measurement Indicators		Weight	Assignment
human capital	Did you learn a lot of new things	5.823%	No is 0, Yes is 1
	Whether a skill is learned	11.602%	No is 0, Yes is 1
social capital	Did you meet more interesting people	10.246%	No is 0, Yes is 1
	Has the sense of collective belonging become stronger	6.156%	No is 0, Yes is 1
financial capital	Whether there has been a significant improvement in income after the establishment of village collective economic organizations	17.878%	No is 0, Yes is 1
	Whether the collective economic organization of the village in which it is located receives dividends	8.414%	0 if no dividend, 1 if dividend has not yet started, 2 if dividend is available
physical capital	Whether or not there are any buildings such as houses in the village community	7.057%	No is 0, Yes is 1
	Availability of public facilities in the village community	13.386%	No is 0, Yes is 1
	Whether there are water facilities in the village collectively	16.562%	No is 0, Yes is 1
natural capital	Whether the village collectively owns land property	2.876%	No is 0, Yes is 1

).

In the process of calculating livelihood capital, in order to effectively overcome the information overlap between indicators and the subjectivity of human determination of indicator weights, and to give the weight values of the given indicators a high degree of credibility, this study adopts the entropy method to determine the weights of each indicator.

As indicators in the livelihood capital evaluation system vary in dimensions and magnitudes, it is important to standardize them before entropy–method calculation. The formula is as follows:

$$x_{ij}^{\prime }={\displaystyle \frac{x_{ij}-x_{min}}{x_{max}-x_{min}}}$$

Step 2: Calculate the proportion $p_{ij}$ of the index value of the i-th sample for the j-th index:

$$p_{ij}={x^{\prime }}_{ij}/\sum _{i=1}^n{x^{\prime }}_{ij}$$

Step 3: Calculate the entropy value $e_j$ for the j-th evaluation indicator:

$$e_j=-{\displaystyle \frac{1}{lnn}}\sum _{i=1}^n{p}_{ij}ln{p}_{ij}$$

Step 4: Calculate the weight $w_j$ of the j-th evaluation indicator:

$$w_j=(1-e_j)/\sum _{j=1}^m(1-e_j)$$

Step 5: Calculate the livelihood capital LC of the farmer:

$$LC=\sum _{j=1}^m{w}_j{p}_{ij}$$

In the equation, $x_{ij}$ represents the data of the j-th measurement indicator for the i-th sample, and LC refers to the livelihood capital index of the farmer.

4.2.2. Explanatory Variables

Skill training is selected as the explanatory variable. The training programs were organized by the village-level collective economic organizations in Chongzhou, with participation on a voluntary basis, primarily involving local farmers. The training content covered a range of topics, including planting techniques, animal husbandry, pest and disease control, agricultural machinery maintenance, land conservation, water resource management, financial management, business management, market analysis, e-commerce, and legal regulations. In this study, the variable “skill training” is assigned values based on the survey question, “Has the village-level collective economic organization organized skill training or other related programs (e.g., legal education, digital literacy)”? The values are as follows: 0 for “none”, 1 for “rarely”, 2 for “occasionally”, and 3 for “frequently”. A higher value indicates that skill training organized by the village-level collective economic organization is more frequent.

4.2.3. Control Variables

Considering the characteristics of individual farmers and the development of village collective economy, this paper chooses livelihood security, policy implementation effect, whether to serve in the village council, financial support, policy support, participation in labor, and the development of village collective economy as control variables (

Table 3. Variable meanings and descriptive statistics.

Variable Type	Symbol	Variable Name	Variable Meaning and Assignment
Dependent variable	LC	Livelihood capital	Farmers’ livelihood capital.
Independent variable	ST	Skills training	Whether village-level collective economic organizations have organized skills training or other relevant training: Not at all = 0; Hardly = 1; Occasionally = 2; Often = 3.
Control variables	LS	Livelihood security	Whether joining a village-level collective economic organization no longer worries about livelihood security: No = 0; Yes = 1.
	PIE	Policy implementation effect	How effective do you think the implementation of regulations or policy documents on the management of village collective economic organizations is: bad = 0; unclear = 1; average = 2; good = 3.
	SVC	Whether to serving in the village council	Whether serving in a government commune or village council: Yes = 1; No = 0.
	FS	Financial support	What do you think about national and local financial support for the development of village collective economy: little = 0; cannot say = 1; generally = 2; a lot = 3.
	PS	Policy support	What do you think about national and local policy support for the development of village-level collective economy: No importance = 0 Can’t say = 1 General = 2 Emphasis = 3.
	PL	Participation in labor	Whether to participate in labor: Yes = 1; No = 0.
	VED	Village collective economic development	How is the development of the collective economy of the village in which it is located: bad = 0; unclear = 1; average = 2; good = 3.

).

4.3. Modeling

Considering that the dependent variable livelihood capital is a continuous variable, the OLS model is constructed in the paper to analyze the impact of skill training conducted by rural collective economic organizations on the livelihood capital of farm households, and the baseline model is set as follows:

$$Y_i={\alpha }_0+{\alpha }_1{X}_i+\beta {control}_i+{\epsilon }_i$$

(1)

where the explanatory variable $Y_i$ denotes the livelihood capital of the ith farm household, the explanatory variable $X_i$ denotes the frequency of skills training conducted by the rural collective economic organization in the village of the ith farm household, ${\alpha }_0$, ${\alpha }_1$ denote the intercept term and the to-be-estimated coefficients of the explanatory variable X, ${control}_i$ denotes the control variable, $\beta$ denotes the to-be-estimated coefficients of the control variable, and ${\epsilon }_i$ denotes the random error term.

5. Empirical Analysis

5.1. Benchmark Regression

Table 4. Benchmark regression results.

	(1) OLS	(2) WLS	(3) FGLS
Variable	LC	LC	LC
ST	0.044 ***	0.043 ***	0.043 ***
	(0.015)	(0.014)	(0.002)
LS	0.094 ***	0.096 ***	0.088 ***
	(0.021)	(0.022)	(0.003)
PIE	0.043 ***	0.047 ***	0.049 ***
	(0.016)	(0.014)	(0.003)
SVC	−0.034	−0.040 *	−0.023 ***
	(0.024)	(0.021)	(0.003)
FS	0.007	0.005	0.005 ***
	(0.013)	(0.011)	(0.002)
PS	0.006	0.002	0.002
	(0.018)	(0.015)	(0.003)
PL	0.048 *	0.039	0.043 ***
	(0.027)	(0.027)	(0.004)
VED	0.033 **	0.029 **	0.033 ***
	(0.014)	(0.012)	(0.002)
Breusch–Pagan	15.61 **
N	418	418	418

Note: * p < 0.1, ** p < 0.05, *** p < 0.01. Robust standard errors are in parentheses.

presents the baseline regression results. Column (1) displays the results from the OLS model, where the coefficient for skills training is positive and statistically significant at the 1% level. The Breusch–Pagan heteroscedasticity test indicates the presence of heteroscedasticity, prompting the use of Weighted Least Squares (WLS) and Feasible Generalized Least Squares (FGLS) to address this issue. Column (2) shows the WLS estimates, where the coefficient for skills training decreases slightly after adjusting for residual weighting but remains significantly positive at the 1% level. Column (3) presents the FGLS estimates, with the coefficient for skills training continuing to be significantly positive at the 1% level. These findings suggest that participation in skills training organized by rural collective economic organizations significantly enhances farmers’ livelihood capital, thus supporting H1.

5.2. Robustness Tests

5.2.1. Endogeneity Test

Endogeneity issues may lead to biased estimates in the results mentioned above. To address the potential impact of omitted variables, measurement errors, and other issues on the estimates, this study employs the instrumental variables (IV) method to handle endogeneity. An ideal instrument must satisfy both relevance and exogeneity conditions. On one hand, the quality and frequency of skills training are influenced by the overall competence of village cadres, which makes the competence of village cadres highly correlated with the skills training organized by rural collective economic organizations. On the other hand, the competence of village cadres is not directly related to the livelihood capital of farmers, thus satisfying the exogeneity condition for the instrument. Consequently, the competence of village cadres (measured based on the overall quality of village cadres in the village—if low, assigned a value of 0; average, assigned a value of 1; high, assigned a value of 2) is chosen as the instrument. Furthermore, the heteroscedasticity-generated instrumental variable method is employed [49]. This method assumes that when the residuals from regressing endogenous variables on other exogenous variables exhibit heteroscedasticity, the product of these residuals and de-centered exogenous variables can serve as an effective instrument.

The results of the endogeneity test are shown in

Table 5. Endogenous treatment results.

	IV-Ols		Lewbel-IV	Standard IV+ Lewbel-IV	Heckman Two-Stage		CMP
Variable	(1)ST	(2) LC	(3) LC	(4) LC	(5)LC_Dummy	(6) LC	(7)ST	(8)LC
ST		0.177 **	0.105 ***	0.113 ***		0.033 ** (0.015)		0.590 *** (0.135)
		(0.074)	(0.028)	(0.028)
Comprehensive quality of village cadres	0.347 *** (0.077)				0.383 ** (0.157)		0.575 *** (0.075)
sense of participation					0.237 **
					(0.115)
IMR						−0.113 **
						(0.055)
Control	YES	YES	YES	YES	YES	YES	YES	YES
Kleibergen-Paap rk LM	17.532 ***		27.783 ***	36.398 ***
Cragg-Donald Wald F	22.926		24.359	23.777
Sargan-Hansen			0.148	0.173
Breusch–Pagan			68.75 ***
atanhrho_12							−0.297 *** (0.113)
N	418	418	418	418	418	418	418	418

Note: ** p < 0.05, *** p < 0.01, Robust standard errors are in parentheses; the same below.

. Columns (1) and (2) report the estimates with the instrument variable “village cadres’ overall competence.” Both the instrument variable and the explanatory variables have significant coefficients. The Kleibergen-Paap rk LM statistic is significant at the 1% level, effectively rejecting the null hypothesis of under-identification of the instrument. The Cragg-Donald Wald F statistic exceeds the 10% critical value for weak instruments, ruling out the possibility of weak instruments. Column (3) presents the estimates using the Lewbel heteroscedasticity-generated instrument variable (HGIV) method. Column (4) shows the estimates considering both the Lewbel instrument and the village cadres’ competence as the instrument. As seen in Table 5, the estimated coefficient for skills training is significantly positive at the 1% level in all models. Additionally, the Breusch–Pagan heteroscedasticity test indicates heteroscedasticity in the residuals from regressing endogenous variables on exogenous variables, satisfying the prerequisite for using the Lewbel method. Columns (7) and (8) present the estimates using the CMP model. The coefficients of the instrument variables and explanatory variables are both significantly positive, confirming that after addressing endogeneity, the positive effect of skills training on livelihood capital remains intact.

Given the potential selection bias in the survey conducted with farmers, this study employs the Heckman two-stage model to address this issue. First, a dummy variable LC_dummy is created by taking the median livelihood capital of all farmers. If a farmer’s livelihood capital is greater than the median, the value is set to 1; otherwise, it is set to 0. In the first stage, LC_dummy is used as the dependent variable, and based on the instrument “village cadres’ overall competence”, the variable “sense of participation” (whether the farmer has a strong sense of participation in the collective economic organization, where no participation is assigned a value of 0, somewhat indifferent is 1, moderate participation is 2, and strong participation is 3) is added as an exclusion restriction variable in a Probit regression. The inverse Mills ratio (IMR) is then obtained. In the second stage, the IMR from the first stage is included as an additional control variable and the model is re-estimated. The results, as shown in columns (5) and (6) of Table 5, indicate that in the first stage, both village cadres’ competence and the sense of participation are significantly and positively correlated with LC_dummy at the 5% level. In the second stage, after controlling for sample selection bias, the coefficient for skills training remains significantly positive, consistent with the baseline regression results.

5.2.2. Other Robustness Tests

To further ensure the reliability of the findings, the empirical model is tested for robustness in three ways in the paper. The specific regression results are shown in

Table 6. Robustness test results.

Variable	(1) LC	(2) LC	(3) LC	(4) LC *
ST	0.040 ***	0.031 **		0.030 ***
	(0.015)	(0.014)		(0.011)
ST *			0.121 ***
			(0.029)
Control	YES	YES	YES	YES
N	359	322	418	418

Note: * p < 0.1, ** p < 0.05, *** p < 0.01. Robust standard errors are in parentheses.

.

First, the dataset is constrained by removing extreme samples, including farmers with educational levels of “no formal education” and “college or above”. Additionally, the top and bottom 5% of livelihood capital values are excluded, creating a new dataset. The estimation results are presented in columns (1) and (2). Second, the explanatory variable is replaced by reassigning values to the core explanatory variable, skills training. Farmers with “no participation” or “minimal participation” are assigned a value of 0, while those with “occasional” or “frequent” participation are assigned a value of 1. This binary variable is named ST* to test the robustness of the results. The estimation results are shown in column (3). As seen in Table 6, the estimated coefficient for skills training is significant and positive, which is consistent with the baseline analysis results. Third, the dependent variable is replaced to further test the robustness of the results. Based on the entropy method, a new livelihood capital index is calculated by combining the TOPSIS method, and it is named LC*. As shown in column (4), after replacing the dependent variable, the impact of skills training on farmers’ livelihood capital remains significant and positive. Therefore, the results are robust with respect to the choice of dependent variable.

5.3. Mechanism

To further explore the mechanism through which skills training provided by rural collective economic organizations influences farmers’ livelihood capital, this study adopts village council (VC: What do you think is the effect of the village council in exercising the collective economic function: bad = 0, fair = 1, good = 2), farmer cooperatives (FC: What is the role of the professional farmer cooperatives in your village in terms of increasing the accumulation of village collective economic organizations and carrying out construction in all aspects: none = 0, bad = 1, average = 2, good = 3), and leading enterprises (LE: For the village collective economic organization to increase accumulation, carry out all aspects of construction, etc., how to play the role of the leading enterprises of agricultural industrialization in your village: none = 0, bad = 1, average = 2, good = 3) as mediating variables to examine the mediation effect. Given that the Generalized Structural Equation Model (GSEM) can uncover causal relationships that traditional correlation analysis may overlook, as well as differentiate between direct and indirect effects, the study employs GSEM to test the underlying mechanism.

As shown in

Table 7. Mechanism analysis results.

	Mechanisms for the Role of Village Councils		Mechanisms for the Role of Cooperatives		Mechanisms for the Role of Leading Enterprises
Variable	(1) VC	(2) LC	(3) FC	(4) LC	(5) LE	(6) LC
ST	0.346 ***	0.036 ***	0.224 **	0.039 ***	0.298 **	0.039 ***
	(0.101)	(0.014)	(0.09)	(0.014)	(0.095)	(0.014)
VC		0.061 ***
		(0.022)
FC				0.023 **
				(0.010)
LE						0.019 **
						(0.009)
Control	YES	YES	YES	YES	YES	YES
N	418	418	418	418	418	418

Note: ** p < 0.05, *** p < 0.01. Robust standard errors are in parentheses.

, in column (1), the coefficient for skills training is significantly positive at the 1% level. Column (2) presents the results with the inclusion of the mediating variable, village committees. The coefficient for village committees is significantly positive at the 1% level, and the coefficient for skills training remains significantly positive at the 1% level. This suggests that skills training by rural collectives enhances farmers’ livelihood capital through improved village governance, thus supporting Hypothesis H2.

In column (3), the coefficient for skills training is significantly positive at the 5% level. Column (4) shows the estimates with the inclusion of the mediating variable, cooperatives. The coefficient for cooperatives is significantly positive at the 5% level, and the coefficient for skills training remains significantly positive at the 1% level. This indicates that skills training by rural collectives enhances farmers’ livelihood capital through improved social networks of farmers’ professional cooperatives, thus supporting Hypothesis H3.

In column (5), the coefficient for skills training is significantly positive at the 5% level. Column (6) presents the results with the inclusion of the mediating variable, leading enterprises. The coefficient for leading enterprises is significantly positive at the 5% level, and the coefficient for skills training remains significantly positive at the 1% level. This suggests that skills training by rural collectives enhances farmers’ livelihood capital through strengthening the role of leading agricultural industrialization enterprises in linking up with farmers, thus supporting Hypothesis H4.

5.4. Heterogeneity Analysis

5.4.1. Heterogeneity Analysis of Farm Households

Farmers with different characteristics may have varying attitudes toward skills training organized by rural collective economic organizations, differ in the benefits they gain from the training, and experience different effects in applying what they have learned to enhance their livelihood capital. To clarify the impact of skills training provided by rural collective economic organizations on farmers’ livelihood capital across different characteristics, this study conducts a heterogeneity analysis based on farmers’ age, risk preferences, and their level of understanding of village collective economics.

Farmers’ ages are categorized into three groups: 18–25 years, 26–45 years, and 46 years and above. Risk preferences are categorized based on the survey question, “If the collective economic organization encourages you to invest in new industries, would you be willing to participate?” Farmers who answer “not willing” and “somewhat hesitant” are categorized as risk-averse, while those who answer “very willing” are categorized as risk-prone. The level of understanding of collective economics is based on the survey question, “Do you understand what village collective economics or village collective economic organizations are?” Farmers who answer “unclear” are categorized as having low understanding, and those who answer “very clear” are categorized as having high understanding. The results of the heterogeneity analysis are shown in

Table 8. Heterogeneity farm households.

	Age of Farmers			Risk Appetite of Farmers		Farmers to Collective Economic Organizations Level of Understanding
	18–25 Years	26–45 Years	46+	Risk Avoidance	Risk Appetite	Low Level of Understanding	High Level of Understanding
Variable	LC	LC	LC	LC	LC	LC	LC
ST	0.017	0.051 *	0.037 **	0.052 **	0.042 **	0.035 **	0.038 **
	(0.307)	(0.028)	(0.017)	(0.021)	(0.019)	(0.02)	(0.019)
Control	YES	YES	YES	YES	YES	YES	YES
N	12	128	278	119	299	78	340

Note: * p < 0.1, ** p < 0.05. Robust standard errors are in parentheses.

.

As shown in Table 8, the impact of skills training on livelihood capital varies across different age groups. For farmers aged 18–25, the effect is not significant. However, for farmers aged 26–45 and those over 46, the impact is significantly positive, with the coefficient for the former being notably larger than that for the latter. A possible explanation for this is that farmers aged 18–25 are often in the early stages of their careers, may still be receiving formal education, or are exploring different career paths. They may also be more inclined to migrate to cities for job opportunities rather than stay in rural areas relying on agriculture for their livelihoods. Given that their life and career directions are still uncertain, they may not be as motivated to absorb and apply agricultural training as older age groups. On the other hand, farmers aged 26–45 typically have more established careers and clearer life goals. With greater responsibilities in both family and career, they have a more urgent need to enhance their livelihood capital, which provides stronger motivation to transform the knowledge and skills acquired through training into productive capabilities, thereby significantly improving their livelihood capital. Farmers over 46 years old may be more conservative and less receptive to new technologies. However, they have accumulated rich experience in agricultural production, and skills training may help them update outdated knowledge and improve traditional farming methods. Despite this, due to physical and cognitive decline, their ability to convert new skills into practical productivity may be less effective than that of middle-aged farmers.

As shown in Table 8, in the different risk preference groups, the impact of skills training on farmers’ livelihood capital is significantly positive regardless of whether the farmers are risk-averse or risk-prone, but the coefficient for risk-averse farmers is larger than that for risk-prone farmers. A possible explanation for this is that risk-averse farmers tend to adopt relatively stable, lower-risk technologies and methods, which generally offer more predictable returns. After acquiring new skills, risk-averse farmers are likely to apply them more cautiously and focus on maximizing returns through careful management, thereby fully utilizing the skills. In contrast, risk-prone farmers are more inclined to experiment with various new technologies and innovative methods. While this increases the potential for higher returns, it may also lead to shallow applications of each technology due to the diversion of energy and resources. Moreover, the frequent switching of technologies or strategies by these farmers may prevent them from consistently gaining stable benefits from a single training program.

As shown in Table 8, in the groups with different levels of understanding of collective economic organizations, the impact of skills training on livelihood capital is significantly positive for farmers with both low and high levels of understanding, but the coefficient for the former is smaller than that for the latter. A possible explanation is that farmers with a higher level of understanding of rural collective economic organizations have a deeper comprehension of the organization’s structure, operations, and potential economic opportunities. As a result, they are able to more quickly grasp the practical application of the training content and effectively filter and utilize information directly related to their production activities, optimizing their decision-making processes and improving economic benefits. In contrast, farmers with a lower level of understanding of rural collective economic organizations lack sufficient information to fully assess and leverage the opportunities offered by the training. This slows down the speed at which they absorb and apply the training information, and they may also fail to fully exploit the new skills to explore new production or market opportunities. Therefore, although skills training still brings positive effects, its impact is significantly weaker for farmers with a lower level of understanding of rural collective economic organizations.

5.4.2. Analysis of the Heterogeneity of Rural Collective Economic Organizations

Rural collective economic organizations, as the primary entities responsible for vocational training, play a pivotal role in enhancing farmers’ livelihood capital. To examine how vocational training conducted by these organizations, with varying characteristics, impacts farmers’ livelihood capital, this study performs a heterogeneity analysis based on two dimensions: the level of attention given to the development of rural collective economic organizations and the extent to which these organizations safeguard farmers’ rights and interests. The first dimension is classified according to the survey question, “How do you perceive the local government’s attitude toward the development of the collective economy in village collectives (village-level collective economic organizations)”? Responses of “not concerned” and “unclear” are categorized as low attention, while “neutral” and “concerned” are categorized as high attention. The second dimension is based on the survey question, “Do the managers of village-level collective economic organizations or the government actively protect the rights of farmers or address the challenges they face”? Responses of “not very active” and “unclear” are classified as low activity, while “neutral” and “very active” are classified as high activity. The results of the heterogeneity analysis for rural collective economic organizations are presented in

Table 9. Heterogeneity analysis of rural collective economic organizations.

	Attention to the Development of Rural Collective Economic Organizations		Degree of Activeness of Rural Collective Economic Organizations in Safeguarding the Rights and Interests of Farming Households
	Low Priority	High Priority	Low Motivation	High Motivation
Variable	LC	LC	LC	LC
ST	−0.003	0.049 ***	0.034	0.041 ***
	(0.022)	(0.015)	(0.043)	(0.016)
Control	YES	YES	YES	YES
N	21	397	29	389

Note: *** p < 0.01. Robust standard errors are in parentheses.

.

As illustrated in Table 9, within the groups categorized by the level of attention given to the development of rural collective economic organizations, farmers in areas where these organizations receive low attention do not show a significant effect of vocational training on their livelihood capital. In contrast, in areas where these organizations receive higher attention, vocational training has a significantly positive impact on farmers’ livelihood capital. One possible explanation is that the varying levels of attention result in differences in resource allocation. Rural collective economic organizations with higher attention tend to have more resources—such as funding, equipment, and qualified instructors—that can be directed toward vocational training programs. This resource advantage enhances the quality and effectiveness of training, offering farmers more effective learning opportunities and enabling them to more effectively enhance their livelihood capital.

As shown in Table 9, within the groups categorized by the level of activity in safeguarding farmers’ rights by rural collective economic organizations, in areas where organizations exhibit low levels of activity, the effect of vocational training on farmers’ livelihood capital is not significant. In contrast, in areas where organizations exhibit high levels of activity, vocational training has a significantly positive impact on farmers’ livelihood capital. A possible explanation is that in an environment where farmers’ rights are actively protected, farmers, feeling that their rights are respected and safeguarded, are more willing to invest time and effort in vocational training. Moreover, they are better able to apply the skills they acquire to practical production activities, thus improving their livelihood. Furthermore, rural collective economic organizations that place a high emphasis on safeguarding farmers’ rights often have effective feedback and adjustment mechanisms in place. This allows training programs to be adjusted promptly based on farmers’ feedback, making the content more aligned with farmers’ actual needs and local conditions, which in turn enhances the adaptability and effectiveness of the training.

6. Conclusions and Policy Implications

This study draws on micro-survey data from rural households in pilot villages under the “District-Society Integration” program in Chongzhou City, Sichuan Province, col-lected in 2023, as the empirical quantitative foundation. It evaluates the impact of skill training organized by rural collective economic organizations on farmers’ livelihood capital, as well as the underlying mechanisms at play. The key findings of the study are as follows:

(1)

Participation in skill training organized by rural collective economic organizations significantly enhances farmers’ livelihood capital. This conclusion remains robust even after addressing potential endogeneity issues in the model.

(2)

The mechanisms through which skill training impacts farmers’ livelihood capital are as follows:

First, participation strengthens farmers’ livelihood capital by enhancing the public service functions of village committees.

Second, it improves farmers’ livelihood capital by strengthening the social networks of farmers’ professional cooperatives.

Third, it boosts farmers’ livelihood capital by reinforcing the role of leading agricultural industrialization enterprises in linking farmers to the industry.

(3)

Further heterogeneity analysis reveals that:

Farmers aged 26–45 and those over 46 are more likely to enhance their livelihood capital through skill training, with the former benefiting more than the latter.

Risk-averse farmers derive greater benefits from skill training in improving their livelihood capital than those who are more risk-tolerant.

Farmers with a higher level of familiarity with rural collective economic organizations experience more significant improvements in livelihood capital through skill training than those with less familiarity.

Farmers in rural collective economic organizations that receive greater attention are more likely to experience substantial gains in livelihood capital through skill training.

Rural collective economic organizations that are more proactive in safeguarding farmers’ rights exhibit a stronger impact on enhancing farmers’ livelihood capital through skill training.

The policy implications drawn from this study are as follows:

(1)

Strengthen the public service and resource allocation functions of village committees. The government should support these committees in providing tailored skill training programs through rural collective economic organizations, optimizing both the content and methods of training, and enhancing their role in coordinating and managing skill training projects.

(2)

Support and optimize farmers’ professional cooperatives and leading agricultural industrialization enterprises. Fiscal subsidies, tax incentives, and other measures should be implemented to foster their healthy and sustainable development, enabling them to offer more market opportunities and technical support to farmers, thereby enhancing farmers’ livelihood capital.

(3)

Address the training needs and outcomes of different groups. Based on factors such as age, risk preference, and familiarity with the collective economy, specialized skill training should be developed around local leading and characteristic industries. Rural collective economic organizations should be guided in designing differentiated training strategies. In particular, for farmers aged 26–45, greater investment should be made in skill training, while more targeted content should be provided to risk-averse farmers and those with less understanding of rural collective economic organizations.

(4)

Increase the attention and support from grassroots governments for skill training initiatives led by collective economic organizations. In policy design and resource allocation, appropriate emphasis should be placed. Through policy guidance and financial support, grassroots governments should continue to encourage rural collective economic organizations to actively participate in and promote skill training projects, ensuring that farmers have equitable access to training opportunities and relevant support, while continually enhancing the human capital in rural areas.