Agricultural Machinery Adoption and Farmers’ Well-Being: Evidence from Jiangxi Province

Abstract

As a cornerstone of agricultural modernization, agricultural mechanization plays a pivotal role in driving rural revitalization and establishing agricultural competitiveness. Drawing upon the theoretical framework of happiness economics, this study investigates the impact, mechanisms, and heterogeneous effects of agricultural machinery adoption on farmers’ subjective well-being, utilizing comprehensive household survey data collected from Jiangxi Province in July 2023. The empirical results demonstrate a significant positive correlation between agricultural machinery adoption and farmers’ subjective well-being, a finding that remains robust after addressing endogeneity concerns through instrumental variable approaches. The mechanism analysis reveals that the enhancement of well-being is primarily mediated through facilitated transitions to non-agricultural employment. The purpose of the mechanism analysis is to explain why agricultural mechanization adoption improves farmers’ subjective well-being. This analysis finds that agricultural mechanization adoption improves farmers’ subjective well-being by helping them transition to non-agricultural employment more smoothly. Furthermore, heterogeneity analysis indicates that the beneficial effects are more substantial among male farmers, individuals with higher educational attainment, and younger demographic groups. These findings suggest that policy interventions should focus on enhancing innovation in agricultural machinery technology, optimizing subsidy programs for agricultural equipment, improving rural education systems, and facilitating the structural transformation of rural labor markets.

1. Introduction

Against the backdrop of China’s rapid economic growth and modernization, the material living standards of its citizens have significantly improved. However, beyond material prosperity, subjective well-being has increasingly become a crucial indicator for assessing national development and social progress. In recent years, the Chinese government has prioritized the comprehensive metric of “happiness”, integrating it into social governance and national development strategies. As the fundamental component of Chinese society, rural areas play a pivotal role in achieving common prosperity and advancing the construction of a modern socialist nation. Within this context, enhancing farmers’ subjective well-being has emerged as a central concern for both academia and policymakers.

The sustainable development of rural China has been constrained by multiple challenges, including widening income inequality, persistent labor shortages, and low agricultural productivity [1]. These structural issues not only hinder the long-term growth of the rural economy but also negatively impact farmers’ quality of life and overall well-being. In response, the Chinese government has implemented a series of policies under the Rural Revitalization Strategy, with agricultural mechanization serving as a key initiative. The widespread adoption of agricultural mechanization has demonstrated multifaceted benefits, not only improving agricultural productivity and reducing labor intensity but also fundamentally transforming farmers’ production methods, living standards, and socio-economic environments.

However, existing research has predominantly focused on economic factors, such as income growth and employment structure adjustments, in explaining farmers’ subjective well-being, with limited attention given to the role of technological transformation through agricultural mechanization. In practice, agricultural mechanization extends beyond labor efficiency improvements; it significantly influences farmers’ psychological well-being, family dynamics, and community engagement, thereby affecting their overall quality of life. Consequently, investigating the impact of agricultural mechanization on farmers’ subjective well-being not only deepens our understanding of the relationship between agricultural modernization and social transformation but also provides valuable insights for policy optimization and rural development strategies.

In the agricultural sector, the adoption of machinery has effectively substituted traditional production factors, yielding substantial positive effects on farmers’ income [2,3]. Mechanization reduces labor time in agricultural production, enhances efficiency, decreases manual labor requirements, and consequently increases farmers’ earnings [4]. Furthermore, the application of agricultural machinery contributes to higher crop yields, facilitates rural economic restructuring, and boosts agricultural income. Simultaneously, agricultural mechanization promotes the transfer of labor to non-agricultural sectors [5]. Empirical evidence suggests that the impact of agricultural mechanization is more pronounced among male laborers compared to their female counterparts [6].

2. Literature Review

Agricultural machinery adoption is a significant factor influencing farmers’ productivity, well-being, and overall rural development in China. Technological advancements in agriculture, such as the introduction of modern machinery, are crucial for improving efficiency, reducing labor costs, and enhancing food production in the country. However, the extent to which these technologies are adopted can vary depending on multiple factors, including infrastructural readiness, economic conditions, and government support. China has made substantial progress in adopting innovative technologies across various sectors, including agriculture. The employment of integrated innovative management technologies in agriculture is seen as a key factor in improving the efficiency and sustainability of agricultural practices. According to Guo et al. (2023) [7], the readiness of China’s macroeconomic environment to integrate these technologies is influenced by various factors, including the capacity of local governments, access to technological infrastructure, and financial support. This study highlights the importance of aligning technological adoption with local conditions, particularly in rural areas, where infrastructural and financial constraints often hinder large-scale adoption of advanced agricultural machinery.

Furthermore, the research by Guo et al. (2023) [7] discusses the role of government policies in promoting technological innovation. In the case of Jiangxi Province, policies targeting the modernization of agriculture are particularly relevant. The development of agro-tech ecosystems, combining governmental incentives with private sector innovation, is crucial for enhancing the adoption of agricultural machinery. Investment in agricultural development has been a critical factor in modernizing the sector. Zhang et al. (2022) [8] explored the investment trends in the agricultural economy sector and found that significant capital inflows into rural infrastructure and agricultural machinery are essential for boosting productivity. In Jiangxi Province, increased investment in agricultural mechanization has been observed, with particular emphasis on improving the accessibility and efficiency of machinery for smallholder farmers. These investments are necessary to bridge the technological gap between urban and rural areas, providing farmers with the tools needed for more efficient and sustainable agricultural practices. The trend of increasing investments also highlights the importance of integrating green technologies in agriculture [9]. With the government focusing on promoting sustainable farming practices, investments are increasingly directed toward environmentally friendly technologies, such as precision agriculture and energy-efficient machinery.

The development of rural infrastructure is fundamental to the adoption of agricultural machinery and the well-being of farmers. In rural territories such as Jiangxi Province, infrastructure includes transportation networks, storage facilities, and access to electricity and water [10]. The study by Li et al. (2022) [11] explores the impact of infrastructure on agricultural productivity, finding that improvements in infrastructure can significantly enhance the efficiency of agricultural machinery. Without adequate roads, storage facilities, and power sources, the benefits of modern machinery may be diminished. Rural infrastructural development also plays a key role in reducing rural poverty by improving access to markets and facilitating the efficient distribution of goods. In Jiangxi, ongoing government efforts to improve rural infrastructure are vital for ensuring that the benefits of agricultural mechanization are fully realized. For example, improvements in transportation networks allow for the easier movement of machinery and agricultural products, directly influencing farmers’ incomes and livelihoods. As China transitions toward a green economy, the role of sustainable agricultural practices becomes increasingly important. Zhang et al. (2022) discuss the adoption of ecological farming practices, which not only focus on the adoption of green technologies but also emphasize resource efficiency in agriculture. This includes the use of machinery that reduces the carbon footprint of agricultural activities and minimizes the environmental impact of farming. In Jiangxi Province, the push towards sustainable development has led to a greater emphasis on resource-efficient technologies. The use of energy-efficient machinery and the incorporation of green agricultural practices, such as organic farming, are seen as key strategies for improving the environmental sustainability of the region [12]. By adopting such practices, farmers can reduce reliance on harmful chemicals and lower the environmental costs of agriculture, leading to long-term benefits for both the local economy and the environment [13]. While the broader literature on agricultural machinery adoption and farmers’ well-being is insightful, this research specifically focuses on the context of Jiangxi Province. Based on the findings from previous studies and the current state of agriculture in Jiangxi, several conclusions can be drawn. First, the province’s continued investment in rural infrastructure and technological adoption is crucial for supporting agricultural mechanization. Specific programs, such as the “Smart Agriculture” initiative and rural electrification projects, have already begun to show positive effects on agricultural productivity. Additionally, government incentives that support the purchase of agricultural machinery, coupled with training programs for farmers, can significantly enhance the adoption rate in the region. These programs should be tailored to local needs, focusing on smallholder farmers who may face barriers to entry, such as high initial costs of machinery or insufficient knowledge of how to use advanced technologies.

In conclusion, the integration of innovative technologies, the development of rural infrastructure, and the promotion of a green economy are all critical components for improving agricultural productivity and farmers’ well-being in Jiangxi Province. By focusing on these specific areas, local governments can ensure that the benefits of agricultural mechanization are widespread, contributing to both economic development and environmental sustainability.

Previous research on farmers’ well-being has primarily focused on its conceptualization and determinants. Subjective well-being, defined as an individual’s comprehensive evaluation of their life circumstances based on self-established standards, encompasses both affective and cognitive dimensions [14]. Scholars have further conceptualized subjective well-being as not merely a global life evaluation but also an individual’s assessment of their current life situation from a life-value perspective [15], representing an equilibrium between positive and negative affective states [16]. This multidimensional and hierarchical construct [17] is characterized by three distinctive attributes: subjectivity, integrality, and relative stability [18].

The determinants of subjective well-being can be systematically classified into four primary dimensions: The first dimension is individual characteristics. Empirical studies reveal gender-based disparities in well-being, with women consistently reporting higher happiness levels than men [19,20]. The age–happiness relationship exhibits a distinctive U-shaped pattern [21,22], while educational attainment demonstrates a significant positive correlation with subjective well-being [23]. Physical health emerges as a crucial determinant of happiness [24,25], and marital status significantly influences well-being, with married individuals reporting greater happiness than their unmarried counterparts [26,27]. Parenthood substantially enhances life satisfaction [28,29], with daughters exerting a more pronounced positive effect on parental happiness than sons [30,31]. Employment status significantly affects well-being, where gainful employment enhances subjective well-being while unemployment diminishes it [32], potentially leading to psychological distress over prolonged periods. Notably, self-esteem exerts a more substantial influence on subjective well-being compared to other individual factors [33,34]. The second dimension is income. The income–happiness relationship presents a complex pattern. While absolute income generally shows a positive correlation with happiness [35,36,37,38], some studies suggest an inverted U-shaped relationship [39]. The Easterlin Paradox highlights the diminishing returns of income on happiness as economic development progresses [40]. Relative income position significantly affects well-being, with lower income groups reporting reduced happiness [41,42,43]. Income inequality demonstrates divergent effects across economic contexts [44,45]: while negatively impacting happiness in developed countries, it shows positive associations in developing economies [46]. The third dimension is social and environmental factors. Housing market dynamics differentially affect well-being, with rising property values enhancing homeowners’ wealth and well-being while increasing financial burdens for renters [47]. Environmental quality significantly influences rural residents’ well-being, where favorable ecological conditions enhance happiness [48,49], while noise pollution adversely affects nearby residents [50]. Similarly, air and water pollution negatively impact residents’ well-being [51]. Social capital plays a pivotal role in shaping well-being through its influence on social belonging [52], with social trust emerging as a particularly significant component in enhancing happiness levels [53]. The fourth dimension is political and institutional factors. Political systems significantly influence citizens’ well-being, with democratic governance showing positive correlations with subjective well-being [54]. Specific democratic mechanisms, including direct democracy [55], clean governance, and representative political systems [56], demonstrate significant positive effects on happiness. Conversely, economic instability, such as financial crises or shocks, can induce anxiety among rural populations, thereby reducing overall well-being [57]. Economic instability, such as financial crises or shocks, can induce anxiety among rural populations, thereby reducing overall well-being [58]. A study by Kim et al. (2016) found that in regions reliant on agriculture, economic downturns or changes in agricultural policies that lead to market instability can significantly affect the mental and physical health of farmers, with agricultural mechanization offering a potential solution for reducing dependency on unstable crop yields [59]. These shifts also highlight the interplay between agricultural policy, mechanization, and well-being, as countries that invest in modern agricultural technology tend to improve both their economic stability and the well-being of their rural populations, especially by improving food security and income levels [60].

While the existing literature provides a robust theoretical foundation, there remains a significant research gap in understanding the relationship between agricultural machinery adoption and farmers’ subjective well-being. As a pivotal element of modern agricultural development, agricultural mechanization plays a transformative role in reshaping farmers’ lifestyles and has profound implications for rural revitalization and well-being enhancement. Building upon the theoretical framework of happiness economics, this study employs field survey data collected during the implementation of the Rural Revitalization Strategy in Jiangxi Province in 2023. Methodologically, we utilize an ordered logistic regression model to examine the impact of agricultural machinery adoption on farmers’ subjective well-being. To further investigate the heterogeneity of these effects, we implement moderation effect analysis, focusing on key demographic variables, including gender and educational attainment. Additionally, we employ a mediation effect model to explore the underlying mechanisms through which non-agricultural employment and land scale operate in the context of agricultural mechanization. This study makes two substantive contributions to the existing literature. First, by focusing on Jiangxi Province as a representative agricultural region, it provides novel empirical evidence on the mechanisms and impacts of agricultural machinery adoption on farmers’ well-being, thereby advancing the theoretical understanding of subjective well-being in rural contexts. Second, through its examination of the mediating roles of non-agricultural employment and land scale, this study offers valuable policy insights for optimizing agricultural mechanization strategies and enhancing farmers’ welfare.

3. Theoretical Framework and Research Hypotheses

3.1. The Well-Being Implications of Agricultural Mechanization: A Theoretical Perspective

The theoretical framework of happiness economics suggests that subjective well-being is determined by a complex interplay of both economic and non-economic factors. This construct represents an individual’s comprehensive assessment of their life quality, with evaluation criteria being inherently subjective and varying across individuals. The determinants of subjective well-being exhibit substantial heterogeneity among individuals, and their perceptions of happiness standards are often markedly distinct. Moreover, temporal variations in subjective well-being can be observed even within the same individual.

From a theoretical perspective, subjective well-being is typically conceptualized through two distinct yet interrelated dimensions: cognitive and affective components. The cognitive dimension encompasses an individual’s overall assessment of life quality and satisfaction, while the affective dimension involves transient psychological states and emotional experiences, ranging from positive feelings such as joy to negative emotions like anxiety. Within the context of this study, farmers’ subjective well-being is operationalized as a dual-component construct comprising their cognitive evaluation of life satisfaction and affective psychological experiences.

The widespread adoption of agricultural machinery has fundamentally transformed traditional farming practices, leading to significant improvements in agricultural productivity, a reduction in labor intensity, and the expansion of per capita arable land utilization. This technological transformation has facilitated the reallocation of agricultural labor to non-agricultural sectors, where labor productivity substantially exceeds that of traditional farming. Consequently, the wage differential between non-agricultural employment and agricultural work has resulted in increased household income and improved living standards for rural families. Furthermore, the integration of agricultural machinery has generated multiple synergistic effects, including enhanced crop yields, structural transformation of the rural economy, and the acceleration of agricultural technological innovation, thereby promoting sustainable agricultural development. These multidimensional impacts demonstrate that agricultural mechanization influences various aspects of farmers’ livelihoods, ultimately affecting their subjective well-being through both direct and indirect channels. Based on this theoretical framework and empirical evidence, we propose the following hypothesis:

H1: 

The adoption of agricultural machinery has a significant positive impact on farmers’ subjective well-being.

3.2. The Mediating Role of Non-Agricultural Employment

On the one hand, Agricultural mechanization serves as a catalyst for labor reallocation through two interconnected mechanisms. Firstly, it significantly alleviates the labor intensity in traditional agricultural practices, facilitating the transition of rural labor to non-agricultural sectors. The adoption of mechanized equipment reduces dependence on manual labor in agricultural production processes. By automating various farming operations, mechanization enhances operational efficiency, reduces time allocation to agricultural activities, and diminishes the physical demands of farm work. This labor liberation effect creates substantial opportunities for farmers to pursue employment in non-agricultural industries. Simultaneously, agricultural mechanization contributes to productivity enhancement through improved crop yields, superior product quality, and reduced production costs, ultimately leading to increased agricultural profitability. This income effect not only elevates farmers’ living standards but also provides the necessary economic foundation for diversifying into non-agricultural employment opportunities. The dual mechanisms of labor liberation and income enhancement establish non-agricultural employment as a critical mediator in the relationship between agricultural mechanization and farmers’ well-being.

On the other hand, the transition to non-agricultural employment offers farmers substantial economic and social advantages compared to traditional agricultural reliance. Economically, it provides more stable and diversified income streams, which not only alleviate agricultural production costs and significantly enhance household living standards but also reduce vulnerability to agricultural risks such as crop failures and natural disasters. This financial resilience contributes substantially to improved subjective well-being. From a social perspective, non-agricultural employment enables farmers to transcend traditional agricultural identities, fostering personal and professional development. Through exposure to new vocational opportunities and skill acquisition, farmers enhance their employability and career prospects. Simultaneously, the expansion of social networks and the elevation of social status through non-agricultural work contribute to increased professional satisfaction and social belonging, further enhancing personal fulfillment. At the macroeconomic level, agricultural mechanization has catalyzed the structural transformation of rural economies from agriculture-dominated systems to diversified economic frameworks. This transformation has stimulated growth in the rural service and manufacturing sectors, enhancing economic vitality and labor absorption capacity. The resulting economic prosperity has facilitated improvements in rural infrastructure and social services, including education and healthcare systems, thereby elevating overall rural living conditions. These interconnected mechanisms suggest that agricultural mechanization indirectly enhances farmers’ subjective well-being through its facilitation of non-agricultural employment. Based on this theoretical framework, we propose the following hypothesis:

H2: 

Non-agricultural employment mediates the positive relationship between agricultural mechanization and farmers’ subjective well-being.

3.3. The Mediating Role of Land Scale

Agricultural mechanization demonstrates particular efficacy in large-scale farming operations, where it generates substantial economies of scale. The mechanization of agricultural processes reduces reliance on manual labor, and as the scale of cultivated land increases, the per-unit production costs experience a significant reduction. These cost efficiencies create strong incentives for farmers to expand their land management scale, thereby achieving enhanced economic returns.

The expansion of land scale facilitates improved production efficiency through mechanized operations and intensive management practices. Large-scale farming enables more effective distribution of fixed costs, enhances productivity, and substantially increases agricultural income. This income effect contributes to greater economic stability, enabling farmers to better manage life uncertainties and consequently improve their subjective well-being.

Furthermore, land scale expansion creates opportunities for farmers to engage in cooperative projects and agricultural partnerships, fostering enhanced interactions with other farmers, agricultural enterprises, and government agencies. These expanded social networks and positive institutional engagements not only generate economic benefits but also strengthen social cohesion, reduce rural isolation, and ultimately contribute to improved overall well-being. Based on this theoretical framework, we propose the following hypothesis:

H3: 

Land scale expansion mediates the positive relationship between agricultural mechanization and farmers’ subjective well-being.

4. Research Methodology

4.1. Data Sources and Collection

Jiangxi Province, a prominent agricultural region with a rich farming heritage, serves as an ideal case study for examining the relationship between agricultural mechanization and farmers’ well-being. The provincial government has prioritized agricultural mechanization as a strategic approach to enhance agricultural productivity and promote rural development. This policy context, coupled with the region’s agricultural significance, makes Jiangxi Province particularly representative for investigating the impact of agricultural machinery adoption on farmers’ subjective well-being. This study utilizes primary data collected through a comprehensive household survey conducted in Jiangxi Province in July 2023. To ensure geographical and economic representativeness, a multi-stage sampling approach was implemented. First, 10 counties (cities/districts) were selected based on key indicators, including economic development levels and geographical characteristics (Figure 1). Subsequently, within each selected county, farmers were randomly sampled from diverse village groups to capture regional variations. To ensure data reliability and validity, the research team implemented rigorous quality control measures. All survey personnel underwent standardized training to maintain consistency in data collection procedures. The household surveys were conducted through face-to-face interviews to minimize response bias and ensure data accuracy. From the initial collection of 560 questionnaires, 523 were retained as valid samples after excluding responses with substantial missing data or inconsistent responses, resulting in a high effective response rate of 93.39%. As demonstrated in Supplementary Material, the measurement showed adequate reliability (ICC = 0.82) and validity (r = 0.72 with SWLS) in our context.

Jiangxi Province, a prominent agricultural region with a rich farming heritage, serves as an ideal case study for examining the relationship between agricultural mechanization and farmers’ well-being. The provincial government has prioritized agricultural mechanization as a strategic approach to enhance agricultural productivity and promote rural development. This policy context, coupled with the region’s agricultural significance, makes Jiangxi Province particularly representative for investigating the impact of agricultural machinery adoption on farmers’ subjective well-being.

To ensure methodological rigor, this study utilizes primary data collected through a comprehensive household survey conducted in Jiangxi Province in July 2023. A multi-stage stratified random sampling approach was employed to enhance geographic and socioeconomic representativeness:

Stage 1—County Selection:

-

Ten counties (cities/districts) were purposefully selected from Jiangxi’s 100 counties using stratified sampling based on economic development levels (low/medium/high GDP per capita) and geographic characteristics (mountainous, plains, and lake regions) (Figure 1).

-

This stratification ensured coverage of diverse agricultural contexts and minimized regional selection bias.

Stage 2—Farmer Sampling:

-

Within each selected county, 5 villages were randomly chosen from administrative lists, followed by systematic random sampling of 10–15 households per village using local census registers.

-

The final sample included 523 valid responses from 560 initially collected surveys (93.39% effective rate), with incomplete or inconsistent responses excluded.

To assess county-level differences in well-being perceptions, multilevel regression analyses were conducted, controlling for household income, age, and farm size. Significant disparities emerged across counties (β = 0.18, p < 0.01), with farmers in high-mechanization counties (e.g., Nanchang County) reporting 12–15% higher well-being scores compared to those in less mechanized mountainous regions (e.g., Jinggangshan). These differences persisted even after adjusting for socioeconomic confounders, suggesting localized policy impacts or infrastructure disparities.

Data quality was prioritized as follows:

-

Standardized training for survey personnel to minimize interviewer bias.

-

Face-to-face interviews with structured questionnaires to ensure clarity and reduce non-response errors.

-

Post-survey validation via random callback checks (10% of samples) to confirm response accuracy.

Limitations and Transparency:

-

While the sampling strategy captured key geographic and economic variations, non-participation bias may exist due to reliance on voluntary household participation.

-

County-level findings should be interpreted cautiously, as unobserved factors (e.g., cultural attitudes) may contribute to well-being disparities.

4.2. Variable Selection and Measurement

(1) Dependent Variable

This study’s dependent variable is farmers’ subjective well-being, which represents an individual’s comprehensive evaluation and affective experience of their life circumstances. Following established measurement approaches in the literature, subjective well-being is operationalized through the survey question: “Overall, how would you rate your life happiness?” Responses are measured using a five-point Likert scale ranging from 1 (“very unhappy”) to 5 (“very happy”). The single-item measure was selected after pilot testing showed 92% response consistency with the multi-item Satisfaction With Life Scale (SWLS) in our context. This approach balances measurement rigor with survey feasibility, particularly given the older age profile (mean = 52.9) and educational constraints (38% primary school or below) of our sample.

(2) Independent Variable

The primary independent variable is agricultural machinery adoption, which is categorized into three distinct levels: traditional farming methods (coded as 1), partial mechanization (coded as 2), and full mechanization (coded as 3). This classification captures the varying degrees of mechanization in agricultural practices.

(3) Mediating Variables

This study examines two mediating variables: non-agricultural employment and land scale.Building upon the analysis of labor income disparities as a primary determinant of farmers’ well-being gaps, and following the methodological approach of Ruan et al. [61], non-agricultural employment is incorporated as a key mediating variable. Additionally, given the significant influence of land scale on agricultural income, this variable is selected for mechanism analysis. To mitigate data dispersion, a logarithmic transformation is applied to the land scale variable.

(4) Control Variables

Following the framework established by Luo et al. [62], control variables are selected across three dimensions: individual characteristics, household attributes, and environmental factors. Recognizing that household heads typically serve as primary decision-makers in agricultural production, individual characteristics are operationalized through four indicators: gender, age, educational attainment, and health status of the household head. Household characteristics are represented by income level, while environmental factors are measured through domestic water quality.

Table 1. Variable definition and descriptive statistics.

Variable Type	Variable Name	Variable Definition and Assignment	Mean	Std. Dev.
Dependent variable	Farmers’ subjective well-being	Very unhappy = 1; Relatively unhappy = 2; Neutral = 3; Relatively happy = 4; Very Happy = 5	3.688	0.756
Independent variable	Agricultural machinery application	Traditional farming = 1; Partial mechanization = 2; Full mechanization = 3	2.05	0.485
Mediating variable	Non-agricultural employment	Number of non-agricultural employed family members (persons)	1.925	1.579
	Land scale	Actual cultivated land area of household (mu)	5.147	15.644
Control variable	Gender	Female = 1; Male = 2	1.554	0.497
	Age	Years	52.952	14.701
	Education level	No schooling = 1; Primary school = 2; Junior high = 3; Senior high/Vocational high/Technical school = 4; College/University = 5	1.763	0.931
	Health status	Very unhealthy = 1; Relatively unhealthy = 2; Neutral = 3; Relatively healthy = 4; Very healthy = 5	3.721	0.982
	Household income level	Very poor = 1; Relatively poor = 2; Neutral = 3; Relatively good = 4; Very good = 5	2.876	0.679
	Drinking water quality	Very poor = 1; Relatively poor = 2; Neutral = 3; Relatively good = 4; Very good = 5	3.683	0.886

Note: 1 mu = 667 m² or 0.067 ha.

presents detailed variable definitions and corresponding descriptive statistics.

4.3. Model Specification

Given the ordinal nature of the dependent variable (farmers’ subjective well-being), an Ordered Logit model is employed for estimation. The baseline model is specified as follows:

$$H_{\mathrm{i}}={\beta }_0+{\beta }_{\mathrm{i}}{S}_{\mathrm{i}}+{\displaystyle \sum {\gamma }_j}{C}_{ij}+{\epsilon }_i$$

(1)

In the formula, H_i represents the subjective well-being level of farmer i, S_i represents the application of agricultural machinery by farmer i, and C_ij represents the jth control variable of farmer i, including personal characteristics, family characteristics, business characteristics, and other variables of the farmer. β₀ is a constant term, β_i and γ_j are estimated coefficients, and ε_i is a random perturbation term.

To examine whether the application of agricultural machinery influences farmers’ subjective well-being through non-agricultural employment and farm size, this study further constructs a mediation effect model based on Equation (1). In this framework, non-agricultural employment and land scale are introduced as mediating variables. The mediation model is specified as follows:

$$Mi={\sigma }_0+{\sigma }_1{S}_i+{\sigma }_2{C}_{ij}+{\epsilon }_2$$

(2)

$$Hi={\phi }_0+{\phi }_1{S}_i+{\phi }_2{M}_i+{\phi }_3{C}_{ij}+{\epsilon }_3$$

(3)

In the formula, ${\mathrm{M}}_{\mathrm{i}}$ is the mediator variable, representing the non-agricultural employment situation and agricultural land scale of farmer $\mathrm{i}$; both σ₀ and φ₀ are constant terms; σ₁, σ₂, φ₁, φ₂, and φ₃ are coefficients to be estimated; and both ε₂ and ε₃ are random perturbation terms.

4.4. Endogeneity and Robustness Tests

Potential endogeneity concerns may arise from measurement errors, omitted variable bias, and reverse causality. To address these issues, this study employs an instrumental variable (IV) approach, following established methodologies in the literature. Specifically, the interaction term between the average agricultural machinery adoption rate at the township level (excluding individual farmers) and a geographical factor, such as the distance to the nearest agricultural machinery service center, is used as the instrument for agricultural machinery adoption. This instrument is justified based on its ability to influence agricultural machinery adoption at the township level while being uncorrelated with the error term in the main regression model. The exclusion restriction is satisfied because the distance to the nearest service center affects machinery adoption primarily through its impact on the availability and convenience of maintenance services rather than through direct factors influencing farm-level productivity or other variables related to the dependent variable. To ensure the robustness and reliability of the empirical findings, two additional tests are implemented: (1) Alternative Dependent Variable Specification. Following Tan et al. [63] and considering data availability, life satisfaction is employed as an alternative proxy for subjective well-being. This approach allows for testing the consistency of results across different well-being measures. (2) Alternative Model Specification. Given the ordinal nature of the dependent variable, the analysis is replicated using an Ordered Probit (Oprobit) model. This alternative specification enables the examination of the stability of coefficient signs and significance levels across different estimation methods, thereby enhancing the robustness of the findings.

5. Results and Analysis

5.1. The Relationship Between Agricultural Machinery Adoption and Farmers’ Subjective Well-Being

The descriptive statistics reveal that the mean subjective well-being score across the sample is 3.688 (Table 1), indicating a moderate level of well-being with substantial potential for improvement. This finding underscores the importance of identifying factors that could enhance farmers’ subjective well-being. A cross-tabulation analysis of agricultural machinery adoption levels and subjective well-being reveals significant disparities across different farming methods. Among fully mechanized households (n = 75), 56 farmers reported being “relatively happy” and 7 reported being “very happy”, collectively representing 84.00% of this group. In contrast, partially mechanized farmers (n = 399) showed lower levels of reported happiness, with only 187 and 48 farmers indicating “relatively happy” and “very happy” statuses, respectively, totaling 58.90%. Traditional farming households (n = 49) demonstrated the lowest happiness levels, with only 19 and 8 farmers reporting “relatively happy” and “very happy” statuses, constituting 55.11% of this group (

Table 2. Cross-analysis of application of agricultural machinery and farmers’ subjective well-being.

Subjective Well-Being	Traditional Farming		Partial Mechanization		Full Mechanization
	Count	(%)	Count	(%)	Count	(%)
Very unhappy	1	2.04	1	0.25	0	0
Relatively unhappy	3	6.12	19	4.76	2	2.67
Neutral	18	36.73	144	36.09	10	13.33
Relatively happy	19	38.78	187	46.87	56	74.67
Very happy	8	16.33	48	12.03	7	9.33
Total	49	100	399	100	75	100

). These findings suggest a positive association between agricultural mechanization and farmers’ subjective well-being, as evidenced by the higher proportion of farmers reporting elevated happiness levels among those utilizing agricultural machinery compared to traditional farming practitioners. The gradient pattern observed across mechanization levels provides preliminary evidence supporting the hypothesized relationship between agricultural modernization and enhanced well-being.

5.2. The Impact of Agricultural Machinery Adoption on Farmers’ Subjective Well-Being

The empirical analysis utilizes Stata 17.0 to estimate the model through a stepwise regression approach. This methodological strategy involves first estimating a baseline model with only the dependent variable (farmers’ subjective well-being), followed by the incremental inclusion of control variables to assess the robustness of the estimation results.

Table 3. Benchmark model estimation results.

Variable Name	Farmers’ Subjective Well-Being
	Model 1		Model 2
	Coef.	Std. Err.	Coef.	Std. Err.
Agricultural machinery	0.440 **	0.171	0.544 ***	0.179
Gender			0.118	0.180
Age			−0.011	0.007
Education level			−0.049	0.103
Health status			0.344 ***	0.104
Household income level			0.810 ***	0.136
Drinking water quality			0.181 *	0.099
N	523		523
Pseudo-R2	0.0056		0.0694

Standard errors in parentheses * p < 0.1, ** p < 0.05, *** p < 0.01.

presents the regression results examining the impact of agricultural machinery adoption on farmers’ subjective well-being.

Model 1 reveals a statistically significant positive coefficient for agricultural machinery services at the 5% level (β = 0.440, p < 0.05), indicating that the adoption of agricultural machinery significantly enhances farmers’ subjective well-being. This finding suggests that mechanized farming practices contribute substantially to improving farmers’ happiness levels. The robustness of this relationship is confirmed in Model 2, which incorporates control variables. The coefficient for agricultural machinery services remains significantly positive at the 1% level (β = 0.544, p < 0.01), demonstrating the stability of the estimated effect. These results provide strong evidence that agricultural mechanization exerts a positive influence on farmers’ well-being, independent of other potentially confounding factors. This positive relationship can be explained through multiple mechanisms. First, agricultural machinery adoption mitigates labor constraints by addressing both labor shortages and quality issues in agricultural production. Second, it enhances production efficiency while reducing operational costs, leading to improved economic outcomes. Third, the reduced physical burden associated with mechanized farming contributes to better working conditions. Collectively, these factors enhance farmers’ satisfaction with both their occupational and living circumstances, thereby improving their overall subjective well-being. Therefore, Hypothesis H1 is supported.

5.3. Endogeneity Test Analysis

The instrumental variable (IV) estimation results provide crucial insights into addressing potential endogeneity concerns. In the first-stage regression, the coefficient for the instrumental variable (distance to nearest agricultural machinery service center) is positive and statistically significant at the 1% level (

Table 4. Test results of the instrumental variable method.

Variable	First Stage		Second Stage
	Agricultural Machinery		Subjective Well-Being
	Coef.	Std. Err.	Coef.	Std. Err.
Rate	0.992 ***	0.030
Agricultural machinery			0.332 ***	0.125
Gender	−0.050 **	0.025	0.072	0.102
Age	0.001	0.001	−0.005	0.004
Education level	0.032 **	0.015	−0.036	0.061
Health status	0.010	0.014	0.198 ***	0.058
Household income level	−0.014	0.018	0.459 ***	0.076
Drinking water quality	0.014	0.014	0.099 *	0.056
N	523		523
F-value	162.52	0.000
Weak IV test			10.82	0.001

Standard errors in parentheses * p < 0.1, ** p < 0.05, *** p < 0.01.

). Meanwhile, the first-stage F-statistic exceeds the critical value of 10, confirming that the variable passes the weak instrument test, indicating a strong correlation between the selected instrumental variable and agricultural machinery adoption. The second-stage regression results reveal that the coefficient of agricultural machinery adoption is positive and statistically significant at the 1% level. This suggests that after accounting for endogeneity issues, agricultural machinery adoption does indeed enhance farmers’ well-being. These findings further validate the reliability of our conclusions.

5.4. Robustness Test Analysis

The robustness checks confirm the stability of our findings through two alternative approaches: (1) Alternative Measurement Specification. After modifying the operational definition of the dependent variable (subjective well-being), the estimated coefficient of agricultural machinery adoption remains statistically significant at the 1% level (p < 0.01) with a positive effect size (

Table 5. Robustness test results.

Variable	Farmers’ Subjective Well-being
	Change Dependent Variable		Change Estimation Model
	Coef.	Std. Err.	Coef.	Std. Err.
Agricultural machinery	0.456 ***	0.184	0.302 ***	0.103
Control variables	Controlled		Controlled
Sample size	523		523
R2	0.0843		0.0689

Standard errors in parentheses *** p < 0.01.

). (2) Model Re-specification. When employing alternative estimation methods tailored to the distributional characteristics of the outcome variable, the positive association between agricultural machinery use and farmers’ well-being persists (β = 0.302, p < 0.01). These consistent results across multiple empirical strategies demonstrate the robustness of our core conclusions regarding the welfare-enhancing effects of agricultural mechanization.

5.5. Heterogeneity Analysis: Moderation Effects

Table 6. Adjustment effect regression results.

	Farmers’ Subjective Well-Being
Variable	Model 1 (Gender)	Model 2 (Education)	Model 3 (Age)	Model 4 (Income Level)	Model 5 (Regional Differences)
Agricultural machinery	0.016 (0.360)	0.023 (0.313)	0.023 (0.313)	0.023 (0.313)	0.115 (0.400)
Machinery_gender	0.391 * (0.232)
Machinery_edu		0.309 * (0.153)
Machinery_age			−0.024 ** (0.010)
Family_income_level				0.652 *** (0.245)
Regional_differences					0.412 * (0.186)
Control variables	Controlled	Controlled	Controlled	Controlled	Controlled
N	523	523	523	523	523
Pseudo-R2	0.0718	0.0728	0.0737	0.0835	0.0796

Standard errors in parentheses * p < 0.1, ** p < 0.05, *** p < 0.01.

presents the regression results examining the heterogeneity of effects across different farmer characteristics, including gender, educational attainment, age, family income level, and regional differences within Jiangxi Province. Five distinct models are estimated to investigate potential moderation effects. The results of Model 1 reveal significant gender heterogeneity, with agricultural mechanization having a more substantial positive impact on male farmers’ subjective well-being compared to their female counterparts (β = 0.391, p < 0.05). This disparity is likely due to male farmers’ typically higher involvement in operating and maintaining complex agricultural machinery, leading to greater efficiency gains and well-being improvements. Model 2 shows that the positive effects of agricultural mechanization are more pronounced among farmers with higher educational attainment (β = 0.309, p < 0.05). This suggests that educated farmers are better able to utilize advanced agricultural technologies, demonstrating better learning abilities and technical proficiency in machinery operation. The results of Model 3 reveal significant age-related heterogeneity, with younger farmers experiencing greater improvements in well-being from agricultural mechanization (β = −0.024, p < 0.01). Younger farmers are likely more adaptable to new technologies and more willing to adopt modern agricultural practices, resulting in greater benefits. Model 4 introduces family income level as an additional moderating factor. The results show that higher family income levels are associated with greater improvements in well-being (β = 0.652, p < 0.01). This indicates that wealthier farmers may have better access to and greater ability to adopt advanced agricultural technologies, leading to more significant improvements in well-being. Finally, Model 5 explores regional differences within Jiangxi Province. The results show significant regional heterogeneity, with farmers from certain regions benefiting more from agricultural mechanization (β = 0.412, p < 0.05). This effect could be due to varying levels of infrastructure, access to resources, and the implementation of mechanization policies across different regions. These findings emphasize the importance of considering various demographic and socio-economic factors, such as gender, education, age, family income, and regional differences, when designing agricultural mechanization policies. The benefits of mechanization are not uniformly distributed, and understanding these variations can help tailor policies to maximize their effectiveness across different farmer groups.

5.6. Mechanism Analysis: Pathways of Agricultural Mechanization’s Impact on Farmers’ Well-Being

Building on the theoretical framework, this study examines two potential mechanisms through which agricultural mechanization may enhance farmers’ subjective well-being: (1) facilitating non-agricultural employment and (2) expanding agricultural land scale. A stepwise regression approach within a mediation analysis framework is employed to investigate these pathways.

5.6.1. Non-Agricultural Employment Mechanism

The mediation analysis reveals a significant positive relationship between agricultural machinery adoption and non-agricultural employment (β = 0.329, p < 0.05) (

Table 7. Mediation effect test results.

Variable	Non-Agricultural Employment		Farmers’ Subjective Well-Being		Land Scale		Farmers’ Subjective Well-Being
	Coef.	Std. Err.	Coef.	Std. Err.	Coef.	Std. Err.	Coef.	Std. Err.
Agricultural machinery	0.329 **	0.144	0.512 ***	0.180	−1.109	1.442	0.546 ***	0.179
Non-agricultural employment			0.120 **	0.055
Land scale							0.002	0.006
Control variables	Controlled		Controlled		Controlled		Controlled
Constant	−0.478	0.669	-	-	4.644	6.694	-	-
Pseudo-R2	0.040		0.0734		0.020		0.0695

Standard errors in parentheses ** p < 0.05, *** p < 0.01.

). This suggests that mechanization effectively facilitates labor transfer from agricultural to non-agricultural sectors. When incorporating non-agricultural employment into the baseline model, both agricultural machinery adoption (β = 0.512, p < 0.01) and non-agricultural employment (β = 0.120, p < 0.05) maintain statistical significance. The reduction in the agricultural machinery coefficient compared to the baseline model indicates that non-agricultural employment partially mediates the relationship between mechanization and well-being. Therefore, Hypothesis H2 is supported.

5.6.2. Land Scale Mechanism

Contrary to theoretical expectations, the analysis reveals no significant relationship between agricultural machinery adoption and land scale expansion (β = −1.109, p > 0.10). When land scale is included in the baseline model, agricultural machinery adoption remains significantly positive (β = 0.546, p < 0.01), while land scale shows no significant effect on subjective well-being (β = 0.002, p > 0.10). This suggests that land scale expansion does not serve as a viable mechanism for enhancing farmers’ well-being in this context. The absence of a land scale effect may be attributed to several factors. First, while mechanization may enable land expansion for some households, agriculture remains a high-risk sector vulnerable to both natural and market uncertainties. Second, the relatively low comparative income advantages in agriculture may limit the well-being benefits of land expansion. Third, the potential benefits of increased land scale may be offset by higher operational risks and management complexities. These findings highlight non-agricultural employment as the primary mechanism through which agricultural mechanization enhances farmers’ well-being while challenging the assumed importance of land scale expansion in this relationship. Therefore, Hypothesis H3 is not supported.

The model controlled for other variables, and these control variables did not show significant effects on the outcomes. The pseudo-R² values indicate the explanatory power of the model: the model explains 4.0% of the variance in “Non-agricultural Employment”, 7.34% in “Farmers’ Subjective Well-being”, and 2.0% in “Land Scale”. In conclusion, the adoption of agricultural machinery has a meaningful impact on farmers’ well-being and non-agricultural employment, but land scale does not significantly influence agricultural machinery adoption, contrary to theoretical expectations.

6. Conclusions and Policy Implications

6.1. Research Conclusions

This study yields several important findings regarding farmers’ well-being and the impact of agricultural mechanization. First, the average subjective well-being score of 3.688 indicates moderate overall happiness levels among farmers, with 62.14% reporting being “relatively happy” or “very happy”. However, the remaining 37.86% reporting lower happiness levels suggests substantial potential for well-being improvement through targeted interventions.

Second, the analysis reveals significant disparities in subjective well-being between farmers who have adopted agricultural mechanization and those who have not. Farmers utilizing mechanized practices consistently report higher levels of subjective well-being, highlighting the transformative potential of agricultural modernization.

Third, agricultural mechanization demonstrates a substantial positive impact on farmers’ subjective well-being through multiple pathways. By bridging smallholder farmers with modern agricultural practices, mechanization reduces labor intensity, lowers production costs, and increases farm income, collectively contributing to enhanced well-being.

Fourth, the heterogeneity analysis reveals important demographic and regional differences in the impact of agricultural mechanization. These findings suggest that demographic and regional characteristics significantly influence the effectiveness of agricultural mechanization in enhancing farmers’ well-being.

These conclusions underscore the importance of agricultural mechanization as a catalyst for improving rural well-being while highlighting the need for targeted policies that consider demographic differences in technology adoption and its impacts.

6.2. Policy Implications

Based on the empirical findings, this study proposes a comprehensive policy framework to enhance farmers’ subjective well-being through agricultural mechanization and rural development:

(1) Enhancing Agricultural Machinery Innovation. To strengthen innovation-driven capabilities in agricultural machinery, a multi-pronged approach is recommended: increase R&D investment through dedicated government funds and private sector incentives; foster industry-university-research collaboration through innovative platforms; attract and cultivate high-end technical talent through targeted policies; promote the integration of advanced technologies (AI, IoT, autonomous systems).

(2) Optimizing Agricultural Machinery Subsidies. To improve the accessibility and affordability of agricultural machinery, expand subsidy coverage to include energy-efficient and intelligent equipment and adjust subsidy amounts based on market conditions and farmer needs. For instance, in mountainous southern Jiangxi (e.g., Ji’an, Ganzhou), we propose raising agricultural machinery purchase subsidies to 60% for smallholder farmers yet to adopt mechanization, coupled with a “First-Tractor Incentive Program”. Strengthen supervision mechanisms to ensure transparency and fairness. Prioritize support for smallholder farmers and emerging agricultural sectors.

(3) Improving Farmer Education and Skills Development. To enhance human capital in rural areas, upgrade rural education infrastructure and teaching quality; expand vocational training programs tailored to modern agriculture; develop digital learning platforms for continuous education; and establish feedback mechanisms to align training with farmer needs. Collaborating with Jiangxi Agricultural University, we propose establishing “Agri-Mechanization + Literacy” dual-upgrade courses at village activity centers. Participants completing the program will receive vocational skill certification and be eligible for Jiangxi Province’s “New Era Professional Farmers” support policy. For farmers under 40, we recommend full reimbursement of “Agricultural Machinery Operator License” training fees, with priority enrollment in Jiangxi’s “One College Student Per Village” initiative to enhance youth participation.

(4) Facilitating Labor Market Transition. To promote non-agricultural employment opportunities, provide targeted vocational training for diverse industries; develop comprehensive labor transfer policies with mobility support; strengthen social security systems for transitioning workers; and support non-agricultural entrepreneurship through financial and technical assistance. For example, agricultural machinery cooperatives are supported to expand services such as maintenance, leasing, and processing of agricultural products, and the “agricultural machinery service complex” is piloted in the Poyang Lake plain area (such as Yugan county and Duchang County). For each local farmer employed, the county finance subsidies enterprises 1000 CNY/year.

(5) Inclusion of Marginalized Farmers. First, targeted mechanisms should be implemented: ① Low-literacy farmers: An “Agricultural Machinery Operation Video Library” system with QR codes affixed to equipment (pilot implementation in Suichuan County demonstrated a 73% increase in adoption rates). ② Low-income households: Tiered subsidies (75% subsidy for households with annual incomes < 25,000 CNY). ③ Gender equity: Exclusive female training cohorts (successfully piloted in Ji’an City’s Women’s Agricultural Machinery Team).

(6) The implementation follows a phased approach, initially targeting 15 priority counties in 2025 with a full provincial rollout scheduled for 2027, allowing for dynamic adjustment of subsidy tiers informed by real-time monitoring data (as showed in

Table 8. Regional adaptation framework.

Region Type	Key Intervention	Implementation Modality	Budget Allocation
Northern Jiangxi Plain Region	Smart machinery hubs	Enterprise/cooperative PPP model	60% provincial funds
Southern Jiangxi Mountainous Region	Small equipment leasing	Village collective-operated	80% subsidy + 20% credit
Poyang Lake Ecological Zone	Green tech packages	Eco-compensation mechanism	50% central eco-funds

) and tiered subsidy criteria optimized through real-time analytics from Jiangxi’s Agricultural Big Data Platform.

These policy recommendations aim to create a synergistic effect between technological advancement, human capital development, and labor market transformation, ultimately contributing to sustainable improvements in farmers’ well-being and rural development.