The Impact of Multi-Subjective Governance on Tea Farmers’ Green Production Behavior Based on the Improved Theory of Planned Behavior

Abstract

This study constructs a research framework to examine the decision-making process of tea farmers’ green production behavior based on the improved theory of planned behavior, incorporating external environmental factors such as government regulation, market mechanisms, industrial organization-driven environmental factors, and community governance. A structural equation model was employed to empirically analyze the influence paths and underlying mechanisms of multi-subjective governance on tea farmers’ green production behavior using survey data from 872 tea farmers in the main tea-producing areas of Fujian Province. The results showed that (1) government regulation, market mechanisms, and community governance significantly and directly impact the decision-making of tea farmers’ green production behavior, with path coefficients of 0.676, 0.686, and 0.373, respectively, and market mechanisms also indirectly act on green production behavior through perceptual behavioral control, with a path coefficient of 0.459. (2) The market mechanisms had the greatest influence on the decision-making of tea farmers’ green production behavior (total utility of 0.830), followed by government regulation (total utility of 0.676), community governance (total utility of 0.373), and finally, industrial organization-driven factors (total utility of 0.046), indicating that the market organization and the government departments are the most important external environmental forces affecting the decision-making of tea farmers’ green production behavior. The results provide valuable references for achieving effective multi-subjective governance and guiding/regulating tea farmers’ green production behavior. While strengthening the incentives and constraints of government regulations on tea farmers’ green production behavior, it is important to fully leverage the roles of market mechanisms, industrial organization-driven factors, and community governance in the governance of tea farmers’ green production behavior.

1. Introduction

1.1. Research Background

Green agricultural production refers to reaching the goals of resource conservation, recycling, and environmental and ecological protection in the process of agricultural production and management. It aims to minimize the harm to the agro-ecological environment and the threat to the quality and safety of agricultural products posed by irrational agricultural production methods [1]. In recent years, Chinese agriculture has presented a “dual negative externalities” feature in terms of food safety and environmental pollution issues, and the core issue it involves is how to regulate farmers’ behavior to achieve green production [2]. Thus, guiding and regulating farmers’ green production behavior has become an important action to promote the transformation and development of agriculture towards sustainability. There still exists the improper and non-standard use of fertilizers and pesticides by farmers in their pursuit of agricultural production benefits, and that has led to the excessive application of fertilizers and pesticides beyond the economically optimum level. It poses a significant obstacle to the green and sustainable development of Chinese agriculture [3]. The current problems of agricultural product quality and safety are becoming increasingly complex. It is difficult to fulfill their roles and functions for the traditional single-mode governance models such as government regulation, market-based models, industrial organization, or community self-governance. It is essential to address the scientific problem of how to accurately and effectively leverage the functions and roles of multiple governance entities including government departments, market organizations, industrial organizations, and community organizations. They will influence farmers’ psychological cognition and promote green production as their conscious behavior.

Therefore, this study focuses on 872 tea farmers in Fujian Province, China, and is based on the improved theory of planned behavior (TPB), which includes psycho-cognitive factors such as behavioral attitudes, subjective norms, and perceptual behavioral control, as well as external factors such as government regulation, market mechanisms, industrial organization-driven factors, and community governance. It aims to construct a theoretical analysis framework for the decision-making of tea farmers’ green production behavior under a multi-subjective governance. Using a structural equation model, it empirically analyzes the mechanisms and pathways through which multi-subjective governance influences tea farmers’ psychological cognition, willingness for green production, and behavioral choices. Moreover, this study aims to clarify the intrinsic logic between external factors and psychological mechanisms in the decision-making of tea farmers’ green production behavior, providing scientific references for guiding and regulating tea farmers’ green production behavior.

1.2. Literature Review

Farmers’ green production behavior is a behavior that enables the sustainability and improvement of agricultural productivity [4]. It is not only conducive to reducing the use of fertilizers and pesticides, and protecting the agricultural ecological environment [5], but also ensuring food safety and increasing the farmers’ welfare [6]. As a micro-subject in agricultural production and management, farmers’ green production behavior is influenced not only by internal factors such as individual characteristics, family production characteristics, and cognitive characteristics, but also by external factors such as government regulation, market mechanisms, industrial organization-driven factors, and community governance.

Government regulation is a direct way to influence farmers’ green production behavior [7]. Studies by Li et al. [8] and Liu et al. [9] have shown that government guidance and regulation have a positive effect on farmers’ green production behavior. Huang et al. [10] found in their empirical research that government incentive policies have significant normative effects on farmers’ compliance with waiting periods, reading label instructions, and excessive pesticide use. Shen et al. [11] discovered that the intensity of production supervision and punishment has a significant positive impact on farmers’ use of biological pesticides. However, research by Zhang et al. [12] revealed that restrictive regulation has no significant effect on farmers’ green production behavior. Given the small-scale and decentralized nature of farmers in China, government regulation incurs high costs [13]. Relying solely on traditional, single government regulation models and limited supervisory resources is increasingly difficult to regulate the production behavior of the large number of agricultural production entities [14]. Strengthening the regulatory powers of various social stakeholders such as markets, industrial organizations, and communities has become an important governance measure to standardize and constrain farmers’ green production behavior [15]. Market mechanisms, as an essential factor in the external environment, play a role in influencing farmers’ decision-making regarding green production behavior and complement government regulation [16]. Wang et al. [17] and Yi et al. [18] pointed out that market incentive mechanisms, primarily based on price mechanisms, ensure that farmers obtain higher profits by increasing the selling prices of green agricultural products, thereby influencing farmers’ adoption of green production behavior. Montalvo [19] also found that agricultural product market prices significantly influence farmers’ adoption of green production behavior. Yu [20] argued that quality inspection, as a means of regulating the safety of agricultural products, acts as a constraint and limitation on farmers’ “non-green” production behavior, and strengthening quality inspection can enhance farmers’ safety production behavior. Industrial organizations, acting as the bridge for information communication between farmers, the government, and the market [21], possess advantages in resources such as technology, information, capital, and social networks, which can overcome the shortcomings of small-scale and dispersed farming operations [5] and play a vital role in promoting green development in agriculture. Furthermore, they have a complementary or substitute effect on government regulation, effectively reducing the costs of government regulation [22]. Yuan et al. [23] argued that industry organizations, through the provision of technical training and guidance, agricultural input procurement, and quality supervision of agricultural products, can drive and promote green production among farmers. Zhu et al. [24] also found that the involvement of industry chains can directly stimulate farmers to adopt green production behavior and has a marginal incremental effect. Community governance has a significant promoting effect on farmers’ green production behavior [25] and partially compensates for the shortcomings of government regulation of farmers’ green production behavior, alleviating government regulation pressure, and transferring government regulation costs. At the same time, the realization of the endogenous potential of community governance also requires government guidance and incentives, and the two play a complementary and regulatory role in constraining farmers’ production behavior [26]. Bailey et al. [27] argued that community supervision of producer behavior has a significant advantage in rural food safety risk governance. Xu et al. [28] have confirmed that peer supervision among farmers effectively influences their reputation demands, thereby significantly promoting their pro-environmental behavior choices. Xu et al. [29] pointed out that community-provided green production services are more effective in promoting the continuous improvement of farmers’ green production levels compared to the implementation of strict regulatory measures. Wang et al. [30] provided evidence that the demonstration effect of technology adoption by rural neighbors positively influences the adoption of green production behavior by farmers. In summary, the effective governance of farmers’ green production behavior requires the involvement of social forces and the guidance of the entire society in collaborative governance. It is essential to harness the complementary or alternative role of diverse entities such as the market, industry organizations, and communities in relation to government regulation. This approach can help improve the efficiency of government regulation.

The existing literature has made significant contributions to understanding the factors influencing farmers’ green production behavior from different research perspectives and theoretical foundations. These studies provide valuable insights and offer many areas that can be learned from. However, there are still a few limitations: First, most existing research focuses on analyzing the influence of a single or two external forces on farmers’ green production behavior, such as the government, market, industry organizations, or communities. The mechanisms by which these external forces affect farmers’ green production behavior have not been clearly delineated. Second, many studies on farmers’ green production behavior mainly investigate the adoption behavior of individual technologies or specific types of green production technologies. It is challenging to apply these findings to the decision-making process of adopting other green production technologies. Therefore, it is necessary to encompass a range of agricultural green production technologies in the research scope of farmers’ green production behavior. Addressing these limitations and considering a broader range of external forces and diverse green production technologies will contribute to a more comprehensive understanding of farmers’ green production behavior.

2. Theoretical Framework and Hypotheses

The theory of planned behavior (TPB), as one of the main theories in the field of social psychology for explaining and predicting individual behavior, has been widely applied in studying farmers’ green production behavior [31,32]. It provides a comprehensive pathway of “cognition → intention → behavior” and exhibits strong explanatory and predictive power in the adoption decisions made by farmers regarding green production behavior [33]. Therefore, scholars have applied the TPB to elucidate the intrinsic logic of farmers’ green production behavior decisions and found that farmers’ behavioral attitudes, subjective norms, and perceived behavioral control directly influence their intention to engage in green production, thereby affecting their choices of green production behavior [34]. However, the theory has a prerequisite assumption, which requires the external environment to be consistent and stable, and it controls the influence of the external environment on behavioral decision-makers [35]. To meet this prerequisite assumption, some scholars have introduced external environmental variables, such as policy regulations, market incentives, industrial organization, and community governance [35], which have enhanced the explanatory and predictive power of farmers’ intentions and behavior towards green production.

Moreover, the individual behavior of farmers is the result of the combined influence of internal and external factors in specific contexts. Farmers, as the main actors of green production implementation, have their own willingness to engage in green production as the most direct factor influencing the occurrence of their behaviors [36]. Farmers’ willingness to engage in green production reflects their subjective probability of choosing to implement green production behavior. The stronger the willingness, the greater the probability of selecting green production behavior [33]. However, farmers’ willingness to engage in green production is influenced by their behavioral attitudes, perceptions of external norms, and cognitive control ability regarding the choice to implement green production behavior [31]. Additionally, farmers’ behavioral attitudes, subjective norms, and perceived behavioral control towards selecting green production behavior are also influenced by external environmental factors, such as government regulation, market mechanism, industrial organization-driven factors, and community governance. Furthermore, the decision of tea farmers to engage in green production behavior is also directly influenced by perceived behavioral control. It is evident that external factors influence farmers’ behavioral attitudes, subjective norms, and perceived behavioral control, thereby affecting their willingness and behavior choices towards green production.

Therefore, this study expands the theory of planned behavior (TPB) by incorporating the “environment → cognition” decision process. External environmental variables, including government regulation, market mechanisms, industrial organization-driven factors, and community governance are introduced to construct a theoretical analytical framework for tea farmers’ decision-making process regarding green production behavior, consisting of the “external environment → psychological cognition → behavioral intention → behavior choice”. The concrete behavior model is shown in Figure 1.

2.1. The Impact of Government Regulation on Tea Farmers’ Psychological Cognition and Green Production Behavior

Government regulation reflects the government departments’ use of relevant regulatory policies to externally constrain tea farmers’ green production behavior, influencing their psychological cognition through guiding, incentivizing, and constraining measures, and further impacting their behavioral attitudes, subjective norms, and perceived behavioral control. In terms of guiding regulatory measures, the government promotes the importance and advantages of agricultural green production to tea farmers through the dissemination of green production policies, publicizing, and education and training, enhancing their cognition of the economic, social, and ecological benefits of agricultural green production technologies [16]. It also helps reduce the learning costs and risks associated with adopting agricultural green production technologies [37] and facilitates farmers’ understanding and mastery of agricultural green production technologies [10], thereby improving tea farmers’ behavioral attitudes and perceived behavioral control towards green production, leading to a more positive attitude towards engaging in green production. In terms of incentive regulatory measures, research has shown that the government can effectively promote farmers to choose green production behavior by using material subsidies and cash subsidies [37]. These incentive policies can reduce the marginal costs of tea farmers’ green production [16] and ensure they receive minimum returns or compensation [8], enhancing their profit expectations from green production. This encourages tea farmers to choose to implement green production behavior, thereby strengthening their subjective norms and perceived behavioral control. In terms of constraining regulatory measures, the government regulates tea farmers’ production behavior by enacting relevant regulations and policies. If tea farmers violate the policy objectives, their actions will be subject to administrative penalties or financial sanctions. Faced with the additional costs of punishment, their economic rationality will compel them to choose to implement green production behavior [38], thereby playing a normative role in tea farmers’ green production behavior [39]. Therefore, we propose Hypothesis 1 (H1):

Hypothesis 1 (H1). 

Government regulation has a significant positive impact on tea farmers’ green production behavior.

2.2. The Impact of Market Mechanisms on Tea Farmers’ Psychological Cognition and Green Production Behavior

Market mechanisms reflect the influence of market entities on tea farmers’ adoption of green production technologies through supply and demand relationships and price mechanisms, thereby inducing them to actively choose to implement green production behavior [40,41]. This is primarily achieved through means such as quality-based incentives, quality testing constraints, and information dissemination services to guide and regulate tea farmers’ production behavior. The market price mechanism for high-quality and high-price products plays an important role in ensuring the quality and safety of agricultural products [42]. Tea buyers establish a market discrimination mechanism for high-quality and high-priced tea products by setting strict standards for the tea’s quality. Tea farmers who aim to gain higher profits must adopt green agricultural production technologies to ensure that their tea products meet the purchase standards. Under the constraints of sales regulations, tea farmers consciously comply with the tea quality requirements specified by tea buyers, leading to a more positive attitude towards green production and influencing their subjective norms [43]. Behind the high-quality and high-price strategy for agricultural products are various limiting conditions, such as the ecological environment, compliance with green production standards, and meeting the requirements of quality safety testing [11]. Therefore, through quality safety testing, opportunistic behavior in unsafe production by farmers is suppressed [44], and the risk costs of “non-green” production are increased, which encourages the formation of environmental attitudes among farmers [35], thus increasing the likelihood of choosing to implement green production behavior. It is evident that tea buyers conduct quality and safety testing on tea, which imposes requirements on tea farmers, thereby creating constraints and strengthening their subjective norms and perceived behavioral control in adopting green agricultural production technologies, thus influencing the choice and implementation of green production behavior.

In terms of information dissemination services, agricultural input dealers are one of the important channels for farmers to obtain information on green agricultural production technologies, which to some extent influences farmers’ green production behavior. Due to the lack of knowledge among most farmers regarding green agricultural production technologies, such as biological pesticides, commercial organic fertilizers, and proper use of pesticides, some farmers rely on the recommendations and promotions from agricultural input dealers to acquire information and know-how on green agricultural production technologies [45], thereby encouraging them to engage in green production [46]. Therefore, the behavioral attitudes of tea farmers can be influenced by the propaganda and recommendations of agricultural input dealers. After acquiring information and knowledge on green agricultural production technologies, tea farmers can enhance their behavioral control abilities, thereby impacting their green production behavior. Therefore, we propose Hypothesis 2 (H2):

Hypothesis 2 (H2). 

Market mechanisms have a significant positive impact on tea farmers’ green production behaviors.

2.3. The Impact of Industrial Organization-Driven Factors on Tea Farmers’ Psychological Cognition and Green Production Behavior

Industrial organization-driven factors reflect how farmers engage in agricultural production activities by joining cooperatives or establishing beneficial relationships with agricultural enterprises, following contractual arrangements made by industrial organizations [47]. This includes receiving technical training and guidance, purchasing agricultural inputs, and supervising the production quality of agricultural products. This incentive mechanism, which is tied to interests, can promote farmers’ engagement in green production [23]. The organizational relationship between farmers and industrial cooperatives can influence farmers’ green awareness [5], thus impacting their willingness to engage in green production and determining their production behavior choices. Industrial cooperatives provide tea farmers with agricultural green production technology training and guidance, which can increase tea farmers’ level of awareness and influence their attitudes towards green production behavior. At the same time, it reduces the learning costs and transaction costs of adopting agricultural green production technologies [48], strengthens tea farmers’ control over green production behavior, and promotes the adoption of agricultural green production technologies. Industrial organizations establish strict tea production standards and conduct rigorous quality and safety inspections and supervision of the tea production process. This process also enhances tea farmers’ level of green production awareness and strengthens their attitudes and subjective norms towards green production. It benefits industrial organizations in regulating tea farmers’ production behavior through comprehensive participatory supervision [49]. Additionally, industrial organizations provide tea farmers with unified and low-priced services for purchasing green agricultural inputs, such as highly efficient and low-toxicity pesticides and organic fertilizers. This not only reduces the production costs and resource constraints of tea farmers [3], but also enhances their level of green production, strengthens their perceived behavioral control, and reduces the misuse of chemical fertilizers and pesticides [50]. Therefore, we propose Hypothesis 3 (H3):

Hypothesis 3 (H3). 

Industrial organization-driven factors have a significant positive impact on driving tea farmers’ green production behavior.

2.4. The Impact of Community Governance on Tea Farmers’ Psychological Cognition and Green Production Behavior

Community governance reflects the direct constraints on tea farmers’ green production behavior through top–down supervision by community organizations in the form of governance guidelines, mutual supervision among farmers within the community, and guidance through neighborhood demonstration effects. Community organizations, possessing official legitimacy, authority, and enforcement, assist the government or its delegated agencies in promoting, guiding, and supervising farmers’ green production behavior, thus influencing farmers’ behavioral cognition and subjective norms, and promoting their adoption of green production behavior. In rural communities characterized by acquaintances, reputation is important to farmers [28]. Mutual monitoring among neighboring farmers has a significant impact on farmers’ behavior as a form of social supervision [51]. Therefore, constraining tea farmers’ green production behavior through the monitoring role of neighboring farmers can enhance their subjective norms and subsequently influence their choices of green production behavior. Close communication between tea farmers and their neighbors not only allows them to access information and methods regarding agricultural green production, reducing information search costs, but also enables tea farmers to have a more intuitive understanding of the effectiveness of agricultural green production technologies. Therefore, based on the “learning effect” and “imitation effect” formed among tea farmers, as well as considerations of pursuing greater profits [52], the neighborhood demonstration effect is internalized into tea farmers’ attitudes towards green production behavior. This enhances their control over green production and subsequently influences their decisions regarding green production behavior. Therefore, we propose Hypothesis 4 (H4):

Hypothesis 4 (H4). 

Community governance has a significant positive impact on tea farmers’ green production behavior.

2.5. The Impact of Tea Farmers’ Psychological Cognition on Their Willingness to Engage in Green Production

The attitude of farmers’ green production behavior reflects their level of awareness and intention to engage in green production. The higher the level of farmers’ awareness and positive evaluation of green production, the greater their willingness to implement it. Wang et al. [53] found that the economic benefits, ecological benefits, and responsibility benefits perceived by farmers in green production have a more direct impact on their willingness to participate. For tea farmers, choosing to engage in green production behavior not only increases tea sales revenue and improves tea quality and safety, but also enhances the ecological environment of tea gardens, resulting in significant economic, social, and ecological benefits. The higher the level of value awareness that tea farmers have towards green tea production, the more evident their attitude towards green production becomes, and the higher their willingness to adopt green production. The subjective norms under which farmers adopt green production behavior refer to the decision-making process influenced by government policies and regulations, market conditions, support or opposition from industry organizations, and the acknowledgement or discrimination from neighboring farmers [54]. The promotional guidance from agricultural extension personnel, market incentives for high-quality and high-price products, the application of green production technologies by tea cooperatives, and the impact of neighboring tea farmers’ adoption of green production technologies all play important roles in tea farmers’ decision-making process to adopt green production behavior. They enhance the subjective norms towards engaging with green production, resulting in a tendency to choose green production in the tea production process. The perceived behavioral control of farmers’ adoption of green production behavior refers to the constraints they may face in terms of resources, policies, risks, personal abilities, and experiences when making decisions regarding green production. When farmers lack resources, abilities, opportunities, or experiences that make it difficult for them to choose green production, their willingness to engage in such behavior will weaken [35]. For tea farmers, the higher their confidence is in acquiring knowledge and information on agricultural green production technologies and their ability to control green production risks and the resources required for green production, the higher their willingness to engage in green production is. Therefore, we propose Hypothesis 5 (H5):

Hypothesis 5 (H5). 

Tea farmers’ cognitive perception has a significant positive impact on their willingness to engage in green production.

2.6. The Impact of Tea Farmers’ Perceived Behavioral Control and Intentions to Engage in Green Production on Green Production Behavior

According to the TPB, perceived behavioral control directly influences individuals’ behavioral intentions and behavior implementation [55]. Previous studies have demonstrated that perceived behavioral control has a direct impact on farmers’ production behavior [54]. Therefore, when tea farmers have a stronger ability to control the resources needed for implementing green production behavior, such as technology, information, funds, and labor, their perceived behavioral control over green production becomes stronger, thus directly influencing their choice to engage in green production. Studies conducted by scholars have also confirmed the direct impact of farmers’ green production intentions on behavior implementation. For example, Zhao et al. [30] utilized the theory of planned behavior and a structural equation model to study the decision-making mechanism of green production in new agricultural entities, and they found evidence for the explanatory and predictive role of behavioral intentions in behavior implementation. Wang et al. [35] also conducted an empirical analysis using the theory of planned behavior to examine the key influencing factors and pathways of farmers’ green production intentions and behavior. The results indicated that farmers’ green production intentions positively promoted green production behavior. Therefore, we propose Hypotheses 6 (H6) and Hypotheses 7 (H7):

Hypothesis 6 (H6). 

Tea farmers’ perceived behavioral control has a significant positive impact on their green production behavior.

Hypothesis 7 (H7). 

Tea farmers’ green production intention has a significant positive impact on their green production behavior.

3. Research Method

3.1. Questionnaire Design

The dependent variable in this study was green production behavior, abbreviated as GPB. Since no single agricultural green production technology can comprehensively reflect tea farmers’ green production behavior, this study referred to the approach adopted by Xie et al. [56] to measure the tea farmers’ green production behavior by investigating the actual number of agricultural green production technologies adopted by the sample of tea farmers. The total score of tea farmers’ adoption of green production technologies was categorized as low adoption (assigned a value of 1) for scores ranging from 0 to 2, moderate adoption (assigned a value of 2) for scores ranging from 3 to 5, and high adoption (assigned a value of 3) for scores ranging from 6 to 8.

Independent variables: based on the above theoretical analysis, the potential independent variables in this study include government regulation (GP), market mechanisms (MG), industrial organization-driven factors (OG), community governance (CG), behavioral attitudes (AT), subjective norms (SN), perceived behavioral control (PBC), and green production intention (GPI). Government regulation (GP) includes 5 items: promotion and training, policy subsidies, government supervision, pesticide residue testing, and government penalties. Market mechanisms (MG) includes 4 items: formulation of procurement standards, constraints on sales rules, tea purchase testing, and dissemination of technical information. Industrial organization-driven factors (OG) includes 4 items: promotion of green production technologies, formulation of production standards, production quality supervision, and agricultural input purchasing services. Community governance (CG) includes 3 items: supervision by village committees, supervision by farmers, and neighborhood demonstrations. Behavioral attitudes (AT), drawing on the research of Zhao et al. [35] and Wang et al. [53], includes 3 items: economic value, social value, and ecological value. Subjective norms (SN), drawing on the research of Hou et al. [54], Wang et al. [53], and Wang et al. [57], includes 4 items: government influence, market influence, organizational influence, and community influence. Perceived behavior control (PBC), drawing on the research of Cao et al. [34], Xie et al. [56], Zhao et al. [35], and Wang et al. [57], includes 4 items: awareness of the hazards of excessive use of fertilizers and pesticides, degree of policy subsidies, level of technological understanding, and information acquisition ability. In measuring green production intention (GPI), due to the differences in the usability and usefulness of green technologies in the fertilization and pesticide application stages of tea farming, separate measurement scales were developed for tea farmers’ intention of green fertilization and green pesticide application. The variables of tea farmers’ green production intention were measured based on their adoption of 5 green pesticide technologies and 3 green fertilization technologies in the tea production process.

Control variables. In the tea production process, individual, household, and production-related characteristics of tea farmers can influence their choices and implementation of green production behaviors. Drawing on relevant studies by scholars such as Geng [58], Cheng et al. [59], Luo et al. [16], He et al. [32], and Wang et al. [48], this study selected tea farmers’ educational level, risk preferences, tea labor force, non-part-time, tea-planting experience, tea-planting scale, and degree of tea garden concentration as control variables.

The data in this study were collected from the household survey of tea farmers in Anxi County, Wuyishan City, and Fuding City, Fujian Province from July to August 2022. The location of the sample areas in this study is shown in Figure 2. The specific sampling process was as follows: firstly, tea-producing townships or streets were selected in the three counties or cities for the investigation; secondly, three administrative villages were selected in each township (street); finally, 15–25 tea farming households were randomly selected in each administrative village for the survey. This survey used a combination of household interviews and questionnaire surveys. The survey team members filled out the questionnaires based on the responses of the surveyed tea farmers. A total of 960 questionnaires were distributed, with 960 questionnaires collected. After excluding questionnaires with contradictions and omissions, the final dataset consisted of 872 valid questionnaires, resulting in a questionnaire validity rate of 90.83%. The description and assignment of all variables are shown in

Table 1. Description of variables and descriptive statistics.

Variable Type	Variable Name	Measurement Question Items	Variable Description	Average Value	Standard Deviation
Dependent Variable	Green Production Behavior (GPB)	Degree of Adoption of Green Production Behavior by Tea Farmers	1 = Low adoption; 2 = medium adoption; 3 = high adoption rate	1.833	0.818
Independent Variable	Government Regulation (GP)	Advocacy Training GP1	Number of times per year that relevant departments, such as government agricultural extension stations, carry out awareness raising and training for green production technologies	0.881	0.889
		Policy Subsidies GP2	Government subsidies for tea farmers to adopt agricultural green production techniques: 1 = very small; 2 = relatively small; 3 = average; 4 = relatively large; 5 = very large	1.323	0.663
		Government Regulation GP3	Do the relevant local government departments regulate the application of pesticides and fertilizers by tea farmers? 1 = yes, 0 = no	0.929	0.257
		Pesticide Residue Testing GP4	Do the local government authorities test tea for pesticide residues? 1 = yes, 0 = no	0.750	0.434
		Government Punishment GP5	The local government imposes and enforces strict penalties on tea farmers for non-green production: 1 = strongly disagree; 2 = do not really agree; 3 = fairly strongly agree; 4 = quite strongly agree; 5 = strongly agree	3.450	0.708
	Market Mechanisms MG	Acquisition Criteria Development MG1	Tea buyers set strict tea quality acquisition standards: 1 = strongly disagree; 2 = do not really agree; 3 = fairly strongly agree; 4 = quite strongly agree; 5 = strongly agree	4.406	0.673
		Sales Rules Binding MG2	Tea farmers consciously comply with the quality requirements of tea agreed by the tea market or the purchaser: 1 = strongly disagree; 2 = do not really agree; 3 = fairly strongly agree; 4 = quite strongly agree; 5 = strongly agree	4.360	0.604
		Tea Acquisition Testing MG3	Do tea buyers conduct quality and safety tests on tea leaves? 1 = yes, 0 = no	0.700	0.459
		Technical Information Dissemination MG4	Do agro-dealers disseminate information on green agricultural production techniques to tea farmers? 1 = yes, 0 = no	0.534	0.499
	Industrial Organization-Driven Factors OG	Green Production Technology promotion OG1	Do tea cooperatives provide training and guidance to tea farmers on agricultural green production techniques? 1 = yes, 0 = no	0.452	0.498
		Establishing production Standards OG2	Do tea cooperatives set strict tea production standards for tea farmers? 1 = yes, 0 = no	0.452	0.498
		Production Quality Supervision OG3	Do tea cooperatives conduct strict quality and safety inspections and supervision of tea farmers’ tea production processes? 1 = yes, 0 = no	0.454	0.498
		Agricultural Purchase Service OG4	Do tea cooperatives provide tea farmers with low-toxicity and high-efficiency pesticides, organic fertilizers, and other agricultural purchasing services? 1 = yes, 0 = no	0.222	0.416
	Community Governance CG	Village Council Supervision CG1	Strong supervision of green production of tea farmers by village committees: 1 = strongly disagree; 2 = not very much agree; 3 = fairly strongly agree; 4 = quite strongly agree; 5 = strongly agree	3.094	0.624
		Farmers’ Supervision CG2	Adoption of green production behaviors by surrounding tea farmers has a supervisory effect on other tea farmers: 1 = strongly disagree; 2 = do not really agree; 3 = average; 4 = mostly agree; 5 = strongly agree	3.853	0.665
		Neighborhood Demonstration CG3	Demonstration effect from the adoption of green production technology by the neighbors and the exchange of learning among tea farmers: 1 = strongly disagree; 2 = d not really agree; 3 = average; 4 = mostly agree; 5 = strongly agree	3.149	0.694
	Behavioral Attitude AT	Economic Value AT1	Do you think the adoption of agricultural green production technology will increase the income from tea sales: 1 = strongly disagree; 2 = do not really agree; 3 = average; 4 = mostly agree; 5 = strongly agree	3.720	0.621
		Social Value AT2	Do you think the quality and safety of tea will be improved after adopting agricultural green production technology: 1 = strongly disagree; 2 = do not really agree; 3 = average; 4 = mostly agree; 5 = strongly agree	4.388	0.541
		Ecological Value AT3	Do you think the ecological environment of tea plantations will be improved after adopting agro-green production technology: 1 = strongly disagree; 2 = do not really agree; 3 = average; 4 = mostly agree; 5 = strongly agree	4.417	0.551
	Subjective Norms SN	Government Impact SN1	Your adoption of green production techniques in agriculture is influenced by the relevant government departments: 1 = yes, 0 = no	0.516	0.500
		Market Impact SN2	Your adoption of green agricultural production techniques is influenced by the fact that tea can fetch a higher price: 1 = yes, 0 = no	0.658	0.475
		Organizational Impact SN3	Your adoption of green agricultural production techniques is influenced by tea economic cooperatives: 1 = yes, 0 = no	0.445	0.498
		Community Impact SN4	Your adoption of agricultural green production techniques is influenced by the surrounding tea farmers: 1 = yes, 0 = no	0.447	0.498
	Perceptual Behavioral Control PBC	Overdose Hazard Perception PBC1	You know the danger of excessive use of chemical fertilizers and pesticides: 1 = almost no impact; 2 = little impact; 3 = fair; 4 = serious; 5 = very serious	3.750	0.606
		Level of Policy Subsidies PBC2	The extent to which you think government subsidies have influenced the adoption of green production technologies in agriculture: 1 = almost no impact; 2 = little impact; 3 = fair; 4 = serious; 5 = very serious	3.571	0.666
		Level of Technical Understanding PBC3	How much do you know about green production technologies in agriculture? 1 = do not know; 2 = know a little; 3 = average; 4 = know quite well; 5 = know very well	2.887	0.825
		Information Acquisition Capability PBC4	How easy it is for your family to get information on green production technologies in agriculture? 1 = difficult; 2 = relatively difficult; 3 = average, 4 = relatively easy; 5 = very easy	3.417	0.912
	Green Production Intention GPI	Biogenic Pesticides GPI1	Your family is willing to use bio-sourced pesticides: 1 = yes, 0 = no	0.860	0.347
		Ecological Regulation Technology GPI2	Your home is willing to adopt ecological regulation technology: 1 = yes, 0 = no	0.739	0.440
		Biological Control Technology GPI3	Your family is willing to use biological control technology: 1 = yes, 0 = no	0.216	0.412
		Physical and Chemical Induction and Control Technology GPI4	Your family is willing to use physical and chemical induction and control technology: 1 = yes, 0 = no	0.628	0.484
		Scientific Medication Technology GPI5	Your family is willing to adopt scientific medication techniques: 1 = yes, 0 = no	0.998	0.048
		Green Manure Intercropping Technology GPI6	Your family is willing to adopt green manure intercropping technology: 1 = yes, 0 = no	0.667	0.472
		Soil Testing and Fertilization Technology GPI7	Your family is willing to adopt soil testing and fertilization technology: 1 = yes, 0 = no	0.186	0.389
		Organic Fertilizer Technology GPI8	Your family is willing to adopt organic fertilizer technology: 1 = yes, 0 = no	0.915	0.279
Control Variables	Education	Educational Level	1 = did not go to school; 2 = primary school; 3 = junior high school; 4 = high school or junior college; 5 = college and above	3.353	1.074
	Risk	Risk Preferences	1 = do not like to take risks; 2 = general; 3 = love adventure	1.837	0.807
	Labor	Tea Labor	Number of tea farmers’ households engaged in tea growing labor force	2.626	0.852
	Non-Part-Time	Non-Part-Time Degree	Tea income/total household income	0.656	0.271
	Experience	Tea Planting Experience	1 = 5 year and below; 2 = 6–10 year; 3 = 11–15 year; 4 = 16–20 year; 5 = more than 20 years	3.993	0.965
	Scale	Tea Planting Scale (hm2)	Tea farmers’ household tea planting area	1.928	5.202
	Concentration	Degree of Tea Garden Concentration	1 = very fragmented; 2 = more dispersed; 3 = general; 4 = more focused; 5 = very concentrated	2.608	0.912

.

3.2. Analytical Method

The decision-making of tea farmers’ green production behavior is influenced by internal and external factors. However, their behavioral attitudes, subjective norms, and perceived behavior control are latent variables that cannot be observed directly. In order to empirically test the impact paths and underlying mechanisms, as well as estimate the corresponding effects of multi-subjective governance on tea farmers’ psychological cognition, behavioral willingness, and behavior choices in green production, a structural equation model (SEM) was adopted as the main research method to construct an empirical test model for tea farmers’ green production behavior decision-making under multi-subjective governance. The SEM combines factor analysis and path analysis, which can be used to validate the relationships among observed variables, latent variables, and disturbance or error variables in the model, thus estimating the effects of independent variables on dependent variables, including total effects, direct effects, and indirect effects [60]. The SEM for tea farmers’ green production behavior decision-making under multi-subjective governance is depicted in Figure 3.

The mathematical expression of the SEM is as follows:

$$\mathrm{GP}={\mathsf{\alpha }}_{11}{\mathrm{GP}}_1+{\mathsf{\alpha }}_{12}{\mathrm{GP}}_2+{\mathsf{\alpha }}_{13}{\mathrm{GP}}_3+{\mathsf{\alpha }}_{14}{\mathrm{GP}}_4+{\mathsf{\alpha }}_{15}{\mathrm{GP}}_5+{\mathsf{\epsilon }}_1$$

(1)

$$\mathrm{MG}={\mathsf{\alpha }}_{21}{\mathrm{MG}}_1+{\mathsf{\alpha }}_{22}{\mathrm{MG}}_2+{\mathsf{\alpha }}_{23}{\mathrm{MG}}_3+{\mathsf{\alpha }}_{24}{\mathrm{MG}}_4+{\mathsf{\epsilon }}_2$$

(2)

$$\mathrm{OG}={\mathsf{\alpha }}_{31}{\mathrm{OG}}_1+{\mathsf{\alpha }}_{32}{\mathrm{OG}}_2+{\mathsf{\alpha }}_{33}{\mathrm{OG}}_3+{\mathsf{\alpha }}_{34}{\mathrm{OG}}_4+{\mathsf{\epsilon }}_3$$

(3)

$$\mathrm{CG}={\mathsf{\alpha }}_{41}{\mathrm{CG}}_1+{\mathsf{\alpha }}_{42}{\mathrm{CG}}_2+{\mathsf{\alpha }}_{43}{\mathrm{CG}}_3+{\mathsf{\epsilon }}_4$$

(4)

$$\mathrm{AT}={\mathsf{\alpha }}_{51}{\mathrm{AT}}_1+{\mathsf{\alpha }}_{52}{\mathrm{AT}}_2+{\mathsf{\alpha }}_{53}{\mathrm{AT}}_3+{\mathsf{\beta }}_{51}\mathrm{GP}+{\mathsf{\beta }}_{52}\mathrm{MG}+{\mathsf{\beta }}_{53}\mathrm{OG}+{\mathsf{\beta }}_{54}\mathrm{CG}+{\mathsf{\epsilon }}_5$$

(5)

$$\mathrm{SN}={\mathsf{\alpha }}_{61}{\mathrm{SN}}_1+{\mathsf{\alpha }}_{62}{\mathrm{SN}}_2+{\mathsf{\alpha }}_{63}{\mathrm{SN}}_3+{\mathsf{\alpha }}_{64}{\mathrm{SN}}_4+{\mathsf{\beta }}_{61}\mathrm{GP}+{\mathsf{\beta }}_{62}\mathrm{MG}+{\mathsf{\beta }}_{63}\mathrm{OG}+{\mathsf{\beta }}_{64}\mathrm{CG}+{\mathsf{\epsilon }}_6$$

(6)

$$\mathrm{PBC}={\mathsf{\alpha }}_{71}{\mathrm{PBC}}_1+{\mathsf{\alpha }}_{72}{\mathrm{PBC}}_2+{\mathsf{\alpha }}_{73}{\mathrm{PBC}}_3+{\mathsf{\alpha }}_{74}{\mathrm{PBC}}_4+{\mathsf{\beta }}_{71}\mathrm{GP}+{\mathsf{\beta }}_{72}\mathrm{MG}+{\mathsf{\beta }}_{73}\mathrm{OG}+{\mathsf{\beta }}_{74}\mathrm{CG}+{\mathsf{\epsilon }}_7$$

(7)

$$\begin{array}{c}\mathrm{GPI}={\mathsf{\alpha }}_{81}{\mathrm{GPI}}_1+{\mathsf{\alpha }}_{82}{\mathrm{GPI}}_2+{\mathsf{\alpha }}_{83}{\mathrm{GPI}}_3+{\mathsf{\alpha }}_{84}{\mathrm{GPI}}_4+{\mathsf{\alpha }}_{85}{\mathrm{GPI}}_5+{\mathsf{\alpha }}_{86}{\mathrm{GPI}}_6+{\mathsf{\alpha }}_{87}{\mathrm{GPI}}_7+{\mathsf{\alpha }}_{88}{\mathrm{GPI}}_8\\ +{\mathsf{\beta }}_{81}\mathrm{AT}+{\mathsf{\beta }}_{82}\mathrm{SN}+{\mathsf{\beta }}_{83}\mathrm{PBC}+{\mathsf{\epsilon }}_8\end{array}$$

(8)

$$\mathrm{GPB}={\mathsf{\beta }}_{91}\mathrm{GP}+{\mathsf{\beta }}_{92}\mathrm{MG}+{\mathsf{\beta }}_{93}\mathrm{OG}+{\mathsf{\beta }}_{94}\mathrm{CG}+{\mathsf{\beta }}_{95}\mathrm{PBC}+{\mathsf{\beta }}_{96}\mathrm{GPI}+{\mathsf{\epsilon }}_9$$

(9)

Equations (1)–(9) represent the following variables: GP, MG, OG, and CG represent government regulation, market mechanisms, industrial organization-driven factors, and community governance, respectively. AT, SN, and PBC represent behavioral attitudes, subjective norms, and perceived behavior control, respectively. GPI and GPB represent green production intention and green production behavior, respectively. GP₁–GP₅, MG₁–MG₄, OG₁–OG₄, CG₁–CG₃, AT₁–AT₃, SN₁–SN₄, PBC₁–PBC₄, and GPI₁–GPI₈ are observed variables in the SEM. α represents the path coefficients between observed variables and latent variables, β represents the path coefficients between latent variables, and ε represents the residual term.

4. Results

4.1. Reliability and Validity Tests

To ensure the reliability and validity of the research findings, the reliability and validity of the potential independent variable scales in the structural equation model of tea farmers’ green production behavior decision-making were tested using SPSS 21.0 and AMOS 24.0 software. According to

Table 2. Results of reliability and validity tests of potential independent variables.

Potential Independent Variables	Number of Questions	Cronbach’s α	KMO Value	Bartlett’s Sphericity Test	p
Government Regulation	5	0.693	0.748	412.870	0.000
Market Mechanisms	4	0.614	0.649	440.564	0.000
Industrial Organization-Driven Factors	4	0.927	0.784	2232.337	0.000
Community Governance	3	0.822	0.700	485.297	0.000
Behavioral Attitude	3	0.867	0.713	702.349	0.000
Subjective Norms	4	0.659	0.615	348.831	0.000
Perceptual Behavioral Control	4	0.851	0.808	827.313	0.000
Green Production Intention	8	0.688	0.736	479.731	0.000
Overall Scale		0.951	0.938	1314.512	0.000

, the overall Cronbach’s α coefficient of the scale was 0.951, and the Cronbach’s α coefficient of each latent variable ranged from 0.614 to 0.927, which was greater than 0.6. This indicates that the sample data used in this study passed the reliability test, and the scale had a high level of internal consistency. KMO and Bartlett tests were used to examine the validity of the measurement model for the eight latent variables. The overall KMO value of the scale was 0.938, and the Bartlett’s test of sphericity yielded a p-value of 0.000. The KMO values of each latent variable in the model ranged from 0.615 to 0.808, exceeding the threshold of 0.5. The Bartlett’s test of sphericity yielded a p-value of 0.000 for each latent variable, indicating a good validity of the sample data of the observed variables.

4.2. Model Fitness Test

The structural equation model of tea farmers’ green production behavior decision-making under multi-subjective governance established in the previous text was tested for goodness-of-fit using AMOS 24.0 statistical software. Based on the modification indices, the initial model was revised and tested. The absolute fit index, incremental fit index, and parsimonious fit index, totaling 11 indicators, all met the criteria for the goodness-of-fit test, as shown in

Table 3. Results of the overall model fitness test.

Suitability Index	Inspection Index	Model Estimates	Adaptation Standards	Test Results
Absolute Suitability Index	CMIN/DF	2.860	<3	Accepted
	RMSEA	0.065	<0.08	Accepted
	GFI	0.901	>0.9	Accepted
Value-Added Suitability Index	NFI	0.937	>0.9	Accepted
	RFI	0.914	>0.9	Accepted
	IFI	0.943	>0.9	Accepted
	TLI	0.927	>0.9	Accepted
	CFI	0.942	>0.9	Accepted
Minimalist Suitability Index	PGFI	0.682	>0.5	Accepted
	PNFI	0.735	>0.5	Accepted
	PCFI	0.778	>0.5	Accepted

CMIN/DF is the chi-square to degrees of freedom ratio, RMSEA is the root-mean-square error of approximation, GFI is the goodness-of-fit index, NFI is the normed fit index, RFI is the relative fit index, IFI is the incremental fit index, TLI is the Tacker–Lewis index, CFI is the comparative fit index, PGFI is the parsimony goodness-of-fit index, PNFI is the parsimony-adjusted normed fit index, PCFI is the parsimony comparative fit index.

. The overall fit of the modified model was good, indicating that the actual survey data were aligned with the theoretical model established in this study.

4.3. Model Estimation Results

The structural equation model analysis was performed on the modified model of tea farmers’ green production behavior decision-making, and the estimated results of the model parameters are shown in

Table 4. Estimation results of the decision-making model of tea farmers’ green production behavior.

Paths	Unstandardized Estimated Coefficients	Standard Error (S.E.)	Critical Ratio (C.R.)	p	Standardized Estimated Coefficients
Structural equations
AT ← GP	0.385	0.139	2.771	**	0.345
SN ← GP	0.424	0.146	2.895	**	0.414
PBC ← GP	0.051	0.031	1.636	0.107	0.062
GPB ← GP	0.691	0.153	4.517	***	0.676
AT ← MG	0.434	0.058	7.483	***	0.458
SN ← MG	0.497	0.088	5.646	***	0.521
PBC ← MG	0.331	0.102	3.242	**	0.314
GPB ← MG	0.796	0.168	4.736	***	0.686
AT ← OG	0.167	0.101	1.648	0.099	0.069
SN ← OG	0.502	0.155	3.237	**	0.418
PBC ← OG	0.120	0.068	1.764	0.078	0.060
GPB ← OG	0.057	0.078	0.73	0.466	0.046
AT ← CG	0.244	0.119	2.051	*	0.246
SN ← CG	0.371	0.111	3.346	**	0.368
PBC ← CG	0.143	0.076	1.880	0.079	0.124
GPB ← CG	0.398	0.118	3.377	**	0.373
GPI ← AT	0.682	0.095	7.183	***	0.619
GPI ← SN	0.457	0.091	5.024	***	0.454
GPI ← PBC	0.411	0.088	4.670	**	0.409
GPB ← PBC	0.473	0.118	4.008	***	0.459
GPB ← GPI	−0.560	1.063	−0.527	0.598	−0.089
Measurement equation
GP1 ← GP	1.000				0.720
GP2 ← GP	0.464	0.049	9.521	***	0.454
GP3 ← GP	0.143	0.020	7.288	***	0.356
GP4 ← GP	0.399	0.033	12.241	***	0.592
GP5 ← GP	0.864	0.053	16.284	***	0.779
MG1 ← MG	1.000				0.125
MG2 ← MG	1.775	0.142	12.499	***	0.797
MG3 ← MG	2.009	0.161	12.476	***	0.602
MG4 ← MG	2.050	0.164	12.502	***	0.846
OG1 ← OG	1.000				0.590
OG2 ← OG	2.057	0.138	14.879	***	0.996
OG3 ← OG	2.025	0.133	15.206	***	0.981
OG4 ← OG	1.857	0.128	14.472	***	0.899
CG1 ← CG	1.000				0.725
CG2 ← CG	1.058	0.064	16.492	***	0.793
CG3 ← CG	0.837	0.049	16.972	***	0.670
AT1 ←AT	1.000				0.936
AT2 ← AT	0.817	0.047	17.542	***	0.877
AT3 ← AT	0.862	0.047	18.449	***	0.909
SN1 ← SN	1.000				0.666
SN2 ← SN	1.112	0.075	14.833	***	0.786
SN3 ← SN	0.827	0.070	11.760	***	0.558
SN4 ← SN	0.526	0.075	7.057	***	0.353
PBC1 ← PBC	1.000				0.795
PBC2 ← PBC	0.925	0.066	13.978	***	0.671
PBC3 ← PBC	1.514	0.070	21.730	***	0.884
PBC4 ← PBC	1.589	0.079	20.176	***	0.839
GPI1 ← GPI	1.000				0.355
GPI2 ← GPI	2.121	0.295	7.183	***	0.591
GPI3 ← GPI	1.995	0.310	6.439	***	0.594
GPI4 ← GPI	2.161	0.345	6.266	***	0.548
GPI5 ← GPI	2.031	0.321	6.328	***	0.567
GPI6 ← GPI	2.048	0.329	6.223	***	0.536
GPI7 ← GPI	1.564	0.260	6.028	***	0.495
GPI8 ← GPI	0.655	0.146	4.496	***	0.288
Control variables
Education	0.527	0.125	4.219	***	0.498
Risk	0.104	0.034	3.081	**	0.108
Labor	−0.005	0.028	−0.175	0.861	−0.005
Non-Part-Time	0.367	0.064	5.739	***	0.329
Experience	0.031	0.024	1.307	0.191	0.039
Scale	0.057	0.027	2.092	*	0.080
Concentration	0.052	0.024	2.160	*	0.061

***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. GP is government regulation, MG is market mechanisms, OG is industrial organization-driven factors, CG is community governance, AT is behavioral attitudes, SN is subjective norms, PBC is perceived behavioral control, GPI is green production intention, GPB is green production behavior.

. The standard errors of the model parameters were within a reasonable range, and the estimated results also passed the significance test. The verification of the hypotheses and path coefficients of the tea farmers’ green production behavior decision-making model under multi-subjective governance are shown in Figure 4.

4.3.1. Analysis of Factors Influencing Tea Farmers’ Green Production Behavior

The path coefficients of government regulations → behavioral attitude, government regulations → subjective norms, and government regulations → green production behavior were 0.345, 0.414, and 0.676 respectively, all of which passed the significance test at the 5% level. This indicates that government regulations significantly and positively influence tea farmers’ behavioral attitude, subjective norms, and green production behavior, supporting research hypothesis H1. The path of government regulations → perceived behavior control did not pass the significance test. The possible reason was the limited subsidies and information supply from the government regarding the adoption of green production technologies by tea farmers. In the actual survey, only 21.6% of the sampled tea farmers indicated that they received government subsidies for adopting green production technologies, while 42.7% of the sampled tea farmers reported not obtaining agricultural green production technology information and training from government departments. Only 29.6% of the sampled tea farmers reported participating in training provided by government departments only once. In addition, there is a phenomenon of “relative institutional failure” in the constraints within government regulations [8]. Although the government has introduced policies that prohibit tea farmers from using highly toxic pesticides, it does not directly guide tea farmers in selecting and implementing green production behavior.

The path coefficients of market mechanisms → behavioral attitude, market mechanisms → subjective norms, and market mechanisms → green production behavior were 0.458, 0.521, and 0.686, respectively, all of which were significant at the 1% level in the significance test. The path coefficient of market mechanisms → perceived behavioral control was 0.314, passing the significance test at the 0.05 level, indicating that market mechanisms significantly and positively promote tea farmers’ psychological cognition and green production behavior. Research hypothesis H2 was proved.

The path coefficient of industrial organization-driven factors → subjective norms was 0.418, passing the significance test at the 5% level, indicating that industrial organization-driven factors significantly positively affect tea farmers’ subjective norms. However, the paths of industrial organization-driven factors → behavioral attitude, industrial organization-driven factors → perceived behavioral control, and industrial organization-driven factors → green production behavior did not pass the significance test, indicating that research hypothesis H3 was not supported. The possible reason is that the role of current tea cooperative organizations in the tea production process has gradually weakened, and the phenomenon of “hollow organizations” is more pronounced. Tea farmers have limited access to services such as technical training, information provision, and green agricultural input purchases from tea cooperative organizations. This resulted in the failure of industrial organization-driven factors to significantly influence tea farmers’ behavioral attitudes and perceived behavioral control, as well as the failure to motivate and constrain tea farmers’ green production behavior. This result was consistent with the research conclusions of scholars such as Li [61] and He et al. [32].

The path coefficient of community governance → behavioral attitude was 0.246, passing the significance test at the 10% level, and the path coefficients of community governance → subjective norms and community governance → green production behavior were 0.368 and 0.373, respectively, both significantly passing the significance test at the 5% level, indicating that community governance can significantly and positively promote tea farmers’ behavioral attitudes, subjective norms, and green production behavior, supporting research hypothesis H4. However, the path of community governance → perceived behavioral control did not pass the significance test, rejecting research hypothesis H4c. A possible reason for this was that the role of village-level organizations in promoting agricultural green production technology and providing information services is very limited. In the actual survey, only 13.1% of tea farmers stated that they learned agricultural green production technology through village committee-organized technical training. Additionally, 21.3% of tea farmers indicated that they obtained agricultural green production technology information through village committee channels to obtain information on agricultural green production technology, thus resulting in the promotion effect of community governance on tea farmers’ perceptual behavior control not being significant.

The path coefficients of behavioral attitude → green production intention and subjective norms → green production intention were 0.619 and 0.454, respectively, both significantly passing the significance test at the 1% level, and the path coefficient of perceived behavioral control → green production intention was 0.409, passing the significance test at the 5% level, indicating that tea farmers’ psychological cognition significantly and positively influences their intentions to engage in green production, supporting research hypothesis H5.

The path coefficient of perceived behavioral control → green production behavior was 0.459, passing the significance test at the 1% level, indicating that tea farmers’ perceived behavioral control significantly and positively promotes their actual green production behavior, supporting research hypothesis H6. However, the path from green production intention to green production behavior did not pass the significance test, with a path coefficient of −0.089, indicating a discrepancy between tea farmers’ green production intentions and their actual behavior, suggesting that research hypothesis H7 was not supported. Although the sample of tea farmers have a strong intention to adopt agricultural green production technologies, their actual adoption rate is relatively low.

From Table 4, it can be seen that tea farmers’ educational level, risk preferences, non-part-time status, tea-planting scale, and degree of tea garden concentration all have a significant positive impact on green production behavior, indicating that tea farmers with higher educational levels, higher risk preferences, non-part-time status, larger tea planting scale, and higher degree of tea garden concentration are more inclined to adopt green production technologies and are more likely to engage in green production behavior. However, tea labor and tea planting experience did not pass the significance test. This may be due to the increasing trend of agricultural labor workers leaving the agricultural sector and engaging in other industries, as well as the continuous rise in labor costs. Tea farmers often rely on their own abundant family labor resources and continue to use traditional tea planting methods to reduce labor input costs. They can also seek additional family income through off-farm work during the off-season, which results in a lack of motivation for tea farmer households to adopt green production technologies, which is consistent with the findings of Geng et al. [58]. Furthermore, the survey revealed that 71.3% of the sampled tea farmer households had been involved in tea planting for more than 15 years. The longer the tea planting experience, the more likely tea farmers are to possess inertia thinking, relying on their own experience for tea production and being less willing to try new green production technologies. Therefore, the non-significant result for tea planting experience was consistent with the research findings of Cheng et al. [59].

4.3.2. Analysis of the Effect of Multi-Subjective Governance on Tea Farmers’ Green Production Behavior

The direct effects, indirect effects, and total effects among the potential variables in the decision-making model of tea farmers’ green production behavior are shown in

Table 5. Direct, indirect, and total effects among variables of the decision-making model of tea farmers’ green production behavior.

Influence Pathways	Direct Effects	Indirect Effects	Total Effects
Government regulation → Green production behavior	0.676	-	0.676
Market mechanisms → Perceptual behavioral control → Green production behavior	-	0.144	-
Market mechanisms → Green production behavior	0.686	0.144	0.830
Industrial organization-driven factors → Green production behavior	0.046	-	0.046
Community governance → Green production behavior	0.373	-	0.373
Green production intention → Green production behavior	−0.089	-	−0.089

Note: The total effect is the sum of the direct and indirect effects associated with the latent variable in the model, and the indirect effect is the sum of the products obtained by multiplying the path coefficients of all direct effects.

. The influence pathways of tea farmers’ green production behavior indicated that government regulation and community governance have significant direct effects on the decision-making of tea farmers’ green production behavior, with a direct effect of government regulation on green production behavior of 0.676 and a direct effect of community governance of 0.373. Market mechanisms not only have significant direct effects on the decision-making of tea farmers’ green production behavior, but also have significant indirect effects. In other words, besides directly influencing the decision-making of green production behavior, market mechanisms also indirectly affect green production behavior through perceived behavioral control. The indirect effect of market mechanisms → perceived behavioral control → green production behavior was 0.144, and the direct effect of market mechanisms → green production behavior was 0.686. The impact of industrial organization-driven factors on the decision-making of tea farmers’ green production behavior was not significant, with a direct effect of only 0.046. The impact of green production intention on tea farmers’ green production behavior was not significant, with a direct effect of −0.089. This indicated a discrepancy between tea farmers’ green production intention and behavior, as their intention did not effectively translate into actual behavior.

The total effects of government regulation, market mechanism, and community governance on the adoption decision-making of tea farmers regarding green production behavior were 0.676, 0.830, and 0.373 in this model, respectively. However, industrial organization-driven factors did not produce a significant impact on the adoption decision-making of tea farmers related to green production behavior. Therefore, it can be concluded that market mechanisms had the greatest influence on the adoption decision-making of tea farmers’ for green production behavior, followed by government regulation and community governance, while industrial organization-driven factors had the least influence. Market organizations and government departments were the most influential external environmental forces in the governance of tea farmers’ green production behavior. As a high-yielding perennial cash crop, the quality, productivity, and stability of tea are closely related to the site conditions, cultivation techniques, and pest control techniques of tea trees [62]. However, due to its high asset specificity and production costs, market mechanisms and government regulation can effectively reduce the costs and risks of tea farmers in acquiring and using green production technologies to a large extent, enabling them to obtain higher returns and thus promoting their positive engagement in green production behavior.

The green production behavior of tea farmers is a driving force for achieving the green transformation and development of the tea industry, as well as an important action initiative to realize the goal of “green mountains and clear waters are as valuable as mountains of gold and silver” in tea production areas. Under the background of increasing resource and environmental constraints, some tea farmers still lack the enthusiasm to choose and implement green production behavior during tea planting and production processes, consequently affecting the green and sustainable development of the tea industry. Tea farmers, as rational economic agents, make rational choices to implement green production behaviors based on their endowment capabilities and psychological cognition under specific external environmental constraints, aiming to maximize welfare utility after fully weighing the risks and benefits. Multi-subjective governance provides guidance, incentives, and constraints for tea farmers to choose and implement green production behaviors, while their own endowment capabilities provide possibilities or constraints for tea farmers to choose and implement green production behaviors. Tea farmers exhibit different behavioral attitudes, subjective norms, and perceived behavioral control under multi-subjective governance, which affects their level of psychological cognition towards green production, and thus influences their intentions and behavior choices regarding green production.

However, this study revealed a negative effect (−0.089) of tea farmers’ intention to engage in green production on their actual behavior, indicating a disconnect between their intentions and actions. This discrepancy can be attributed to the ineffective synergistic effects of multi-subjective governance in governing tea farmers’ green production behavior or the constraints imposed by the farmers’ own endowment capabilities, resulting in divergent perceptions regarding green production. Consequently, tea farmers may exhibit inconsistent intentions and behaviors towards green production during the tea production process. This result is consistent with the research conclusions of scholars such as Yu et al. [63] and Xu et al. [64], which indicated that the divergence between farmers’ green awareness and their intentions to adopt green production technologies leads to inconsistent behavior. The actual investigation revealed that tea farmers face constraints in adopting agricultural green production technologies due to factors such as limited government support and subsidies, high implementation costs and risks, the failure to reflect the value of high-quality tea, and the difficulty of technology implementation. Therefore, even when tea farmers have strong intentions for green production, it does not necessarily translate into actual actions.

Tea farmers, as rational economic agents, engage in agricultural production and management behavior driven by rationality, with maximizing benefits as a significant decision-making criterion. However, due to limitations in their knowledge and the asymmetry of information, tea farmers may have weak abilities in risk-taking and control. Moreover, agricultural green production incorporates various modern production elements, such as technology, talent, information, and machinery, which pose challenges to traditional tea farmers [65]. As a result, some tea farmers may exhibit inconsistencies between their intentions and behaviors. Despite the implementation of various measures by relevant government departments, such as the introduction of green agricultural subsidy policies, the establishment of agricultural product quality and safety supervision systems, the promotion of modern agricultural technology innovation and dissemination systems, and the implementation of agricultural social service systems, to alleviate the discrepancies between tea farmers’ intentions and behaviors in green production, small-scale household farmers still constitute the majority of tea farmers in China. When making decisions to implement green production behaviors, they are influenced by not only their own endowments but also various external environmental forces. These external environmental forces mainly come from stakeholders such as government departments, market organizations, industry organizations, and community organizations. The level of collaborative governance and its effects on tea farmers’ green production behavior impose an influence, to a certain extent, on the conversion of intention into actual green production behavior. In general, the higher the degree of collaboration in multi-subjective governance, the more effective it is in guiding and regulating the implementation of green production behavior among tea farmers. Therefore, enhancing the collaborative governance among multiple stakeholders, including government departments, market organizations, industry organizations, and community organizations, can effectively resolve the dilemma of the discrepancy between tea farmers’ intentions and their actual production behavior in green production.

Compared to previous research findings, this paper mainly provides supplementary and extended analysis in the following areas. It introduces external environmental variables such as government regulation, market mechanisms, industrial organization-driven factors, and community governance into the TPB framework, expanding it to construct a theoretical analysis framework for the decision-making process of tea farmers’ green production behavior, which follows the sequence “external environment → psychological cognition → behavioral intention → behavioral choice”. This framework thoroughly examines the transmission mechanisms between multi-subjective governance and tea farmers’ psychological cognition, intention, and behavior from various dimensions. It aimed to address the shortcomings of existing studies that predominantly analyzed tea farmers’ green production behavior from different single perspectives.

5. Conclusions and Recommendations

5.1. Conclusions

This study used survey data from 872 tea farmers in the main tea-producing areas of Fujian Province, China. A structural equation model was used to test the research hypotheses and analyze the impact pathways and effects of multi-subjective governance on tea farmers’ psychological cognition, green production intention, and behavioral choices. The conclusions of the study are as follows:

• Government regulations have a positive influence on tea farmers’ attitudes, subjective norms, and green production behavior, but their perceived behavioral control is not strengthened by government regulations. Market mechanisms have a positive impact on tea farmers’ attitudes, subjective norms, perceived behavioral control, and green production behavior. Industrial organization-driven factors have a significant positive promoting effect on tea farmers’ subjective norms, but their attitudes and perceived behavioral control are not enhanced by industrial organization-driven factors. Community governance also has a positive impact on tea farmers’ attitudes, subjective norms, and green production behavior, but its impact on tea farmers’ perceived behavioral control is not significant. Enhancing tea farmers’ attitudes, subjective norms, and perceived behavioral control can significantly increase their intentions for green production, and tea farmers’ perceived behavioral control also plays a significant positive role in green production behavior. However, there is an inconsistency between tea farmers’ green production intentions and their actual behaviors;
• Government regulations and community governance significantly and directly influence tea farmers’ decision-making regarding green production behavior. Market mechanisms not only directly impact tea farmers’ decision-making regarding green production behavior, but also indirectly affect green production behavior through perceived behavioral control. Industrial organization-driven factors do not have a significant impact on tea farmers’ decisions regarding green production behavior. From the perspective of the effects of multi-subjective governance on tea farmers’ green production behavior, market mechanisms have the greatest overall impact on tea farmers’ decisions regarding green production behavior (total effect = 0.8230), followed by government regulations (total effect = 0.676), then community governance (total effect = 0.373), and finally industrial organization-driven factors (total effect = 0.046). This shows that market mechanisms and government regulations play key roles in tea farmers’ decision-making regarding green production behavior.

5.2. Recommendations

• It is necessary to establish a multi-subjective and multi-channel agricultural green production technology promotion system, actively mobilizing diverse stakeholders such as the market, community, and industry organizations to participate in the dissemination and training process of tea farmers’ green production behavior governance. This will broaden the channels for tea farmers to acquire technical information and emphasize the popularization of knowledge and techniques related to green tea production, aiming to enhance tea farmers’ awareness of green production. Through multi-subjective and multi-channel dissemination and training, as well as the promotion of knowledge and relevant technologies for green tea production, the costs of information searching, learning, and time in adopting green production behavior can be reduced for tea farmers. This will increase their attention to tea quality and safety, as well as their understanding and adoption of green production technologies;
• It is necessary to improve and perfect the agricultural subsidy mechanism that is oriented towards green production. This can be achieved by expanding the scope of agricultural green subsidies and increasing the level of policy subsidies. This will address the issue of insufficient continuous investment in green agricultural inputs and equipment for tea farmers, and enhance the willingness and motivation of tea farmers who have not adopted green production technologies to do so. At the same time, it is important to optimize the design of the agricultural subsidy mechanism and develop targeted subsidy strategies based on the differences in tea farmers’ resource endowments. This will incentivize tea farmers with different characteristics to actively adopt green production behaviors, and achieve the desired effect of targeted incentives. In addition, it is also necessary to improve the market supervision system for tea quality and safety. This includes the strict regulation of the agricultural input sales market, improvement of the market access system for tea and the standardization system for tea quality, the establishment of a market identification mechanism for high-quality and reasonably priced tea products, and the promotion of tea farmers’ selection of green production behaviors;
• Fully leverage the collaborative participation of diverse stakeholders in restraining and supervising tea farmers’ green production behaviors. First, increase government oversight of tea farmers’ green production behaviors through measures such as agricultural input regulation, cultivation and production monitoring, market sampling inspections, and continuous special campaigns to address excessive pesticide residues in tea. Strictly crack down on the illegal use of banned or restricted pesticides in tea gardens to impose penalties and losses on tea farmers practicing non-green production, thereby incentivizing the adoption of green production behaviors. Second, strengthen the market regulation role of tea buyers in relation to tea farmers’ green production behaviors. Enhance quality inspections of tea products at the terminal market by tea buyers and encourage tea farmers to comply with agreed-upon tea quality standards through contracts with tea buyers. Third, strengthen the development of industry organizations and continuously innovate and improve the mechanisms that connect the interests of industry organizations with tea farmers. Use both formal and informal institutional arrangements of industry organizations to regulate and restrain tea farmers’ green production behaviors. Fourth, emphasize the coordination and cooperation between government regulation and community supervision. The relevant government departments can delegate certain regulatory powers concerning tea farmers’ green production behaviors to communities, improving the implementation mechanisms for community supervision to ensure effective community oversight. Establish a reward and punishment mechanism for tea farmer supervision, relying on neighboring tea farmers’ supervision of green tea production to enhance the efficiency of regulating tea farmers’ green production behaviors. Additionally, utilize the neighborhood demonstration effect as a complementary measure to supervise and restrain tea farmers’ green production behaviors.