Can Symbiotic Relationship Promote Green Technology Innovation of Agricultural Enterprises? A Study Based on the Empirical Evidence of Chinese Agricultural Listed Companies

Abstract

Agricultural green innovation has become a key driver of China’s agricultural development in the modern era. Maintaining a strong symbiotic relationship is crucial for enhancing awareness and resource allocation capabilities related to green technology innovation within enterprises, ultimately fostering green development. This study utilizes balanced panel data from Chinese agricultural listed companies in the Shanghai and Shenzhen stock markets between 2011 and 2020 to investigate the impact of symbiotic relationships on green technology innovation capability. Through empirical testing and a moderated mediating effect model, the research explores how organizational green awareness and dynamic capabilities influence this relationship. The findings indicate that a positive symbiotic relationship can significantly enhance green technology innovation in agricultural enterprises by improving their green cognition and dynamic abilities. In terms of the heterogeneity of the promotion of green technology innovation by enterprise symbiosis, the symbiotic relationship has a significant direct promotion effect on both progressive and joint green technology innovation. By confirming the effectiveness of firm symbiosis in promoting green technology innovation, this study provides new practical guidance for developing countries on how to promote the development of green technology innovation.

1. Introduction

Countries worldwide are currently striving to balance the pressing issues of productive consumption and ecological protection through initiatives such as low emissions, energy conservation, and recycling [1]. As the largest developing nation, China has put forth the ‘dual carbon’ objective to reach the peak of CO₂ emissions by 2030 and achieve carbon neutrality by 2060 during the 75th session of the United Nations General Assembly [2]. The establishment of this goal signifies China’s commitment to advancing the green transformation of economic and social development by steadfastly following the path of green, low-carbon, circular, and sustainable growth [3]. In recent years, while China’s overall economic level has continued to improve, issues related to ecological resource constraints and insufficient environmental capacity remain prominent [4]. The long-term reliance on the excessive use of chemical fertilizers and pesticides to enhance agricultural production has adversely affected both human health and the natural environment [5,6]. This approach can no longer meet the new demands of China’s agricultural economic development in the current context. Additionally, rough economic development has led to significant waste during the process of social transformation [7]. Consequently, there is an urgent need for green technological innovation to enhance environmental governance capabilities. As a critical measure to promote China’s agricultural and rural modernization and to implement the rural revitalization strategy, agricultural green development holds great significance for ensuring global food security, resource security, and ecological security.

In February 2023, the Ministry of Agriculture and Rural Affairs of China issued the “Implementation Opinions on Implementing the Key Work Deployments of the Party Central Committee and the State Council on Comprehensively Promoting Rural Revitalization in 2023.” This document emphasized the necessity to “strengthen the protection of agricultural resources and environmental governance while promoting the comprehensive green transformation of agriculture.” Agricultural enterprises are encouraged to undertake large-scale green and low-carbon science and technology projects at the national level, facilitate the green science and technology revolution, and enhance corporate environmental governance practices to improve environmental performance [8]. Additionally, these enterprises are urged to promote green economic development and enhance environmental protection efficiency through innovations in green technology [9]. Consequently, it is essential for agricultural enterprises to bolster their capabilities in green technology innovation. Therefore, it is of great significance for agricultural enterprises to enhance their ability of green technology innovation. In recent years, numerous scholars have examined the relevant factors influencing the development of green technology innovation in enterprises from both internal and external aspects. The political relationship [10], leadership style, ecological cognition, opportunity awareness, and R&D investment of enterprise management can enhance the efficiency of green innovation [11,12]. Additionally, the identification and integration capabilities of enterprise resources such as information and technology can facilitate the green technology innovation of enterprises by transforming the internal innovation outcomes of enterprises [13,14]. National strategy, relevant government policies [15,16], media attention, market consumption demand, competitive environment, and other external factors [17,18], as well as enterprises’ initiative to undertake social responsibility, all exert a driving force on enterprises’ green technology innovation [19]. When green technology innovation is incorporated into the production process, it can not only reduce the waste of resources [20], but also establish an ecological recycling system and minimize the damage to the environment [21,22,23]. Furthermore, with the utilization of green technology in enterprise operation and environmental improvement, it also boosts the profitability and corporate value of enterprises and generates green performance [24,25]. It achieves the unity of comprehensive economic, social, and ecological benefits and promotes the healthy development of enterprises [26].

In September 2022, the Implementation Plan for Building a National Agricultural Green Development Pioneer Zone and Promoting the Comprehensive Green Transformation of Agricultural Modernization Demonstration Zones’ issued by the Ministry of Agriculture and Rural Affairs of China emphasized the ‘integrated promotion of agricultural green technologies’ as a key task. It called for enhanced collaboration among various government departments and units, including agriculture and rural areas, development and reform, ecological environment, the People’s Bank of China, and supply and marketing cooperatives, to establish a framework of government guidance, market leadership, and social participation. This highlights the importance of green technology innovation in driving the change and sustainability of the agricultural economic development model [27,28], necessitating coordinated efforts between the government and various agricultural management entities. Therefore, investigating the influence of the symbiotic relationship on the green technology innovation of agricultural enterprises has emerged as a new research foothold. In order to promote green technology innovation in agricultural enterprises, it is essential for agricultural leading enterprises to establish connections with upstream and downstream farmers, herdsmen, and new agricultural business entities in terms of interests, emotions, and meaning construction. Strengthening cooperation and communication among symbiotic relationship subjects serves as the foundation for organizational integration. The intricate nature of these mutualistic relationships highlights the pivotal role of agricultural leading enterprises in driving agricultural green transformation and fostering the development of innovative green technologies [29]. While scholarly research on the relationship between agricultural enterprises’ symbiotic relationships and green innovation is limited, existing studies on the impact of relationships with customers, suppliers, and other stakeholders on enterprise innovation can offer valuable theoretical insights for further research [30]. As ecological and environmental protection have gained prominence in global economic development, corporate stakeholders are increasingly demanding higher social and environmental standards [23]. Stakeholders advocate for enterprises to transition towards green technology development in order to mitigate environmental pollution resulting from corporate operations and production, achieved through the establishment and execution of corporate green innovation strategies [31]. Research indicates that community stakeholders and regulatory bodies may not significantly influence corporate green technology innovation, whereas consumers can impact corporate green innovation intentions. A more stable symbiotic relationship between customers and suppliers leads to greater knowledge spillover, increased communication frequency, higher levels of trust, and ultimately enhanced collaboration in innovation [32,33].

The significance of symbiotic relationships on the innovation of green technology in agricultural enterprises is crucial and should not be overlooked. While the studies mentioned offer a theoretical foundation for this paper, there are some notable limitations that need to be addressed. Firstly, there is a gap in quantitative research focusing on the symbiotic relationship between agricultural enterprises and symbiotic units. Secondly, there is a lack of relevant research exploring the effects of symbiotic relationships at the enterprise level on innovation performance. Therefore, it is essential to explore the role and mechanism of symbiotic relationships in driving green technology innovation within agricultural enterprises. Building upon theoretical foundations, this study utilizes data from listed agricultural enterprises in China spanning from 2011 to 2020. Employing balanced panel data and a moderated mediation effect model, the research empirically evaluates the impact of symbiosis on green technology innovation in agricultural enterprises and its underlying mechanisms. This paper aims to contribute in several ways: Firstly, it empirically validates the role of symbiosis in fostering green technology innovation in agricultural enterprises, marking a preliminary foray into the quantitative application of symbiosis theory in management. Secondly, by utilizing a moderated mediation effect model, the study delves into the mechanisms through which symbiotic relationships influence organizational green cognition via environmental regulation, subsequently impacting green technology innovation, as well as the pathways through which symbiotic relationships affect green technology innovation through dynamic capabilities. This perspective enhances the existing driving mechanism theory of green technology innovation. Lastly, the research conducts a comparative analysis of the heterogeneous impact of symbiotic relationships on different green technology innovation models, offering theoretical insights for practical application in corporate settings.

2. Theoretical Analysis and Research Hypothesis

2.1. Symbiotic Relationship and Green Technology Innovation

Green technological innovation refers to the technological advancements aimed at promoting sustainable development of the ecological environment through environmental protection measures. It seeks to utilize scientific environmental knowledge and technology to achieve a balanced growth of the economy and the environment during the production process [34,35]. Symbiotic relationships play a crucial role in driving green technology innovation in three main aspects. The first mechanism is the expression of suggestions within a symbiotic relationship. Symbiotic relationships facilitate effective communication among symbiotic units, allowing stakeholders such as farmers, herdsmen, new agricultural entities, or cooperatives to voice their environmental concerns and needs through collaboration with agricultural enterprises [36,37]. Certain scholars employed an evolutionary game approach to demonstrate that collaboration among government entities [38], recycling processing manufacturers, and producers is beneficial for driving innovation in corporate green technology. The second aspect involves the selection and withdrawal mechanism of symbiotic units. One of the primary motivations for the development of green technology in agricultural businesses is to establish organizational legitimacy. Symbiotic relationships play a crucial role as the stakeholders most closely associated with agricultural enterprises. The decisions regarding selection and disengagement of symbiotic partners can have a significant impact on the adoption of green practices within agricultural businesses, prompting them to invest in green technology innovation [39,40]. The third is the platform mechanism of the symbiotic interface. The symbiotic interface facilitates the connection of interests, emotions, and meaning construction, creating a community of interests, cause, and destiny with agricultural enterprises. Serving as a platform for information, capital, relationships, technology, and other resources, the symbiotic interface enhances relationships that enable agricultural enterprises to acquire these resources. By improving symbiotic relationships, agricultural enterprises can effectively capture market signals, make forward-looking predictions, and gain a first-mover advantage. This can lead to a shorter green technology innovation process, reduced investment in green innovation, or increased efficiency and success rate of green technology innovation [41,42]. Therefore, the following hypothesis is proposed:

H1. 

Symbiotic relationships have a positive impact on promoting the green technology innovation of agricultural enterprises.

2.2. Symbiotic Relationship, Organizational Green Cognition and Green Technology Innovation

Organizational green cognition is the perception, knowledge and psychological experience formed by organizations on environmental issues. Symbiotic relationships play a crucial role in enhancing the environmental awareness of agricultural enterprises. These relationships not only improve the sense of social responsibility within the enterprises but also foster a stronger connection with their symbiotic partners. This connection motivates enterprises to actively fulfill their corporate citizenship duties, leading to mutual recognition. By consistently practicing social responsibility, agricultural enterprises can expedite the development of their environmental awareness. Additionally, symbiotic relationships offer a range of potential opportunities for enterprises. Close partnerships enable enterprises to promptly and accurately access environmental information from their symbiotic partners. By receiving feedback and recommendations from these partners, agricultural enterprises are better positioned to identify green business opportunities embedded in this information [43,44]. Consequently, the enterprises can shift their perception of environmental issues from being perceived as threats to being viewed as opportunities.

Organizational green cognition plays a crucial role in motivating agricultural enterprises to engage in environmentally friendly practices. When an organization perceives environmental issues as threats, it may view fulfilling environmental responsibilities as detrimental to economic performance. However, external pressures from society, government, and the market often compel enterprises to meet green obligations. Conversely, when an organization sees environmental challenges as opportunities, it recognizes the potential for significant business growth and innovation. In this scenario, the enterprise may proactively pursue green technology innovation to gain a competitive edge. This perspective enables enterprises to achieve economic gains while minimizing negative environmental impact [45]. Based on the analysis above, the following hypotheses are proposed:

H2. 

Symbiotic relationships have a positive impact on promoting the green cognition of agricultural enterprises.

H3. 

Organizational green cognition positively influences the green technology innovation of agricultural enterprises.

H4. 

Organizational green cognition acts as a mediator in the relationship between symbiotic relationships and green technology innovation.

In addition, the symbiotic relationship plays a crucial role in enhancing the environmental awareness of agricultural enterprises by providing suggestions and transmitting information. This information exchange is influenced by environmental regulations. In situations where environmental regulations are weak, agricultural enterprises struggle to interpret green signals effectively, making symbiosis a key channel for managing changes. In contrast, when environmental regulations are strong, agricultural enterprises can readily access green signals externally, leading to an overlap between signals from symbiotic relationships and direct sources. In such cases, environmental regulations tend to overshadow the impact of symbiotic relationships on green technology innovation within agricultural enterprises. Therefore, it is hypothesized that environmental regulations negatively moderate the connection between symbiotic relationships and organizational green awareness. Therefore, the following hypothesis is proposed:

H5. 

Environmental regulation negatively moderates the relationship between symbiotic relationships and organizational green cognition.

2.3. Symbiotic Relationship, Dynamic Capability and Green Technology Innovation

The dynamic ability of agricultural enterprises is the ability of enterprises to integrate internal and external resources according to situational changes. The positive impact of symbiotic relationships on dynamic capabilities can be seen in three main ways. Firstly, it expands the resource acquisition channels for enterprises. Symbiotic relationships, as a community of interests for agricultural enterprises, serve as a crucial avenue for obtaining resources. Moreover, these relationships facilitate access to additional resources through information and connections. Secondly, symbiotic relationships enhance the efficiency of resource integration for agricultural enterprises. Strong symbiotic relationships enable better dissemination of policies, market trends, and other relevant information, leading to more focused resource integration. Additionally, symbiotic relationships directly provide specific resources to agricultural enterprises, further enhancing their resource integration efficiency. According to some scholars, enterprises can develop high-quality products and services while minimizing natural resource consumption through government support for technological innovation [46]. Lastly, close symbiotic relationships grant enterprises greater flexibility in allocating resources at the symbiotic interface, fostering the exchange of interests, knowledge, and skills with symbiotic partners. Over time, as these relationships deepen, the mutual understanding and exchange between partners improve, laying the groundwork for the qualitative enhancement of enterprises’ dynamic capabilities.

The promotion effect of dynamic capability on green technology innovation is reflected in the following aspects: first, enterprises with strong dynamic capability can obtain and process green opportunity signals in the external environment through diversified channels, and correctly link them with internal resource conditions, so as to improve the success rate of green technology innovation. Secondly, enterprises with strong absorptive capacity have accumulated a lot of knowledge in the past development, expanded the breadth and depth of the knowledge source of enterprises, and can better deal with emergencies in the process of green technology innovation, and improve the success rate and efficiency of enterprises’ green technology innovation. Third, enterprises with strong resource integration ability can timely and quickly coordinate relevant resources after making relevant decisions on green technology innovation, ensure continuous and stable resource input in the process of innovation, and accelerate the implementation of green technology innovation. Compared with traditional technology innovation, green technology innovation has the characteristics of large capital investment, long profit cycle and unpredictable risk, so to solve the market failure problems such as environmental externalities, path dependence and imperfect capital market, enterprises are required to invest more capital in green innovation [47]. Fourthly, enterprises with strong restructuring ability can better overcome organizational rigidity and have strong green technology innovation motivation after discovering green opportunities.

Based on the above analysis, the following hypotheses are put forward:

H6. 

Symbiotic relationship plays a positive role in promoting the dynamic capability of agricultural enterprises.

H7. 

Dynamic capability plays a positive role in promoting green technology innovation of agricultural enterprises.

H8. 

Dynamic capabilities act as a mediator in the relationship between symbiosis and green technology innovation.

Symbiotic relationships not only enhance dynamic capabilities to promote green technology innovation, but also regulate the strength of this promotion. A stronger symbiotic relationship results in a more effective increase in dynamic capabilities for promoting green technology innovation. In terms of dynamic capability allocation, enterprises tend to allocate more resources to economic innovation due to the higher risks and spillover associated with green technology innovation. However, a good symbiotic relationship allows for the transmission of green demand and opportunity information, leading to a shift in dynamic capability allocation towards green technology innovation. Therefore, the following hypothesis is proposed:

H9. 

Symbiotic relationship has a positive moderating effect on the relationship between dynamic capabilities and green technology innovation.

To sum up, the moderated parallel mediating effect model constructed in this paper is shown in Figure 1.

3. Research Design

3.1. Model Setting

Based on the following model, this paper empirically tests the total effect of symbiotic relationship on the green technology innovation of agricultural enterprises:

$$G{I}_{it}={\beta }_0+cS{ym}_{it}+\alpha X_{it}+YearFE+RegFE+InduFE+{\epsilon }_{it}$$

(1)

Among them, $GI$ represents the explained variable green technology innovation, which is further subdivided into breakthrough green technology innovation ($GI\_ori$), progressive green technology innovation ($GI\_sec$), independent green technology innovation ($GI\_ind$) and joint green technology innovation ($GI\_uni$) as the explanatory variables. $Sym$ represents the symbiosis of core explanatory variables, $X$ represents the control variable, $YearFE$ is the time fixed effect, $RegFE$ is the regional fixed effect, $InduFE$ is the industry fixed effect, ${\epsilon }_{it}$ is the random disturbance term, subindex $i$ is the sample enterprise, and $t$ is the observation year. The control variables include enterprise nature ($Own$), establishment years ($Age$), capital density ($Den$), market environment ($Mar$) and employee incentive ($Eso$).

In addition, as the theoretical analysis mentioned above, the research of this paper involves different forms of moderated mediating effect tests. With reference to the research method of (Hayes and Scharkow, 2013) [48], the following model is established:

$$Co{g}_{it}=d_1+a_{0}Sy{m}_{it}+X_{it}{\lambda }_0+YearFE+RegFE+InduFE+{\epsilon }_{it}$$

(2)

$$G{I}_{it}=d_2+c_0’{Sym}_{it}+b_{0}Co{g}_{it}+X_{it}{\delta }_0+YearFE+RegFE+InduFE+{\epsilon }_{it}$$

(3)

$$\begin{array}{c}G{I}_{it}=d_3+c’Sy{m}_{it}+b’Cog+eSy{m}_{it}\times En{v}_{it}+X_{it}{\gamma }_0\\ +YearFE+RegFE+InduFE+{\epsilon }_{it}\end{array}$$

(4)

$$Dy{c}_{it}=g_1+a_{1}Sy{m}_{it}+X_{it}{\lambda }_1+YearFE+RegFE+InduFE+{\epsilon }_{it}$$

(5)

$$G{I}_{it}=g_2+c_1’{Sym}_{it}+b_{1}Dy{c}_{it}+X_{it}{\delta }_1+YearFE+RegFE+InduFE+{\epsilon }_{it}$$

(6)

$$\begin{array}{c}G{I}_{it}=g_3+c_2’Sy{m}_{it}+b_{2}Dy{c}_{it}+e_{1}Dy{c}_{it}\times Sy{m}_{it}+X_{it}{\gamma }_1\\ +YearFE+RegFE+InduFE+{\epsilon }_{it}\end{array}$$

(7)

Among them, $Cog$ is the mediating variable of organizational green cognition, $Dyc$ is the mediating variable of dynamic ability, and $Env$ is the moderating variable of environmental regulation. Equation (2) tests the effect of symbiotic relationship on organizational green cognition; Equation (3) tests the influence of organizational green cognition on green technology innovation and the direct effect of symbiotic relationship on green technology innovation; Equations (1)–(3) jointly test the mediating effect of organizational green cognition between symbiotic relationship and green technology innovation; Equation (4) tests the moderating effect of environmental regulation on the mediating effect of organizational green cognition between symbiotic relationship and green technology innovation in the first paragraph; Equation (5) tests the impact of symbiotic relationship on dynamic capability; Equation (6) tests the impact of dynamic capability on green technology innovation and the direct effect of symbiotic relationship on green technology innovation; Equations (1), (5) and (6) jointly test the mediating effect of dynamic capability between symbiotic relationship and green technology innovation; and Equation (7) tests the moderating effect of organizational cognition on the second part of the mediating effect of dynamic capabilities between symbiotic relationship and green technology innovation.

3.2. Variable Definitions

Explained variable: green technology innovation. Green technology innovation is the green innovation in the field of technology. Green technology innovation includes the reduction of resource input, the development or improvement of environmentally affected products, and the reengineering of production processes and the reconstruction of production systems to produce products that meet the requirements of environmental protection [49,50]. According to previous studies, this paper uses the sum of green invention patents and green utility model patent applications plus 1 to take the value of natural logarithm to measure the green technology innovation of enterprises. And it uses the sum of green invention patents and green utility model patents plus 1 to take the value of natural logarithm as the substitute index of green technology innovation to conduct robustness test [51,52]. In addition, this paper uses the number of green invention patent applications plus 1 to take the value of natural logarithm, the number of green utility model patent applications plus 1 to take the value of natural logarithm, the sum of green invention patent and green utility model patent independent applications plus 1 to take the value of natural logarithm, and the sum of green invention patent and green utility model patent joint applications plus 1 to take the value of natural logarithm to measure four distinct green technology innovation modes: breakthrough, progressive, independent, and joint green technology innovation. Green patent data were obtained from China Research Data Service Platform.

Core explanatory variable: symbiotic relationship. Scholars predominantly focus on qualitative research when studying symbiotic relationships, with limited references to quantitative research. A symbiotic relationship, as per symbiosis theory, denotes the interaction or fusion of symbiotic units, showcasing the manner, intensity, and material exchange within these units. In the context of agricultural enterprises, symbiotic relationships primarily encompass the connections between agricultural enterprises and upstream/downstream farmers and herdsmen, as well as with agricultural product consumers. The Corporate Social Responsibility Report’s discussion on safeguarding suppliers’ and consumers’ rights effectively illustrates this symbiotic relationship. To quantify this symbiotic relationship, the paper calculates the average of whether a listed company discloses protection measures for suppliers (disclosure = 1, non-disclosure = 0) and consumers (disclosure = 1, non-disclosure = 0). Furthermore, the paper utilizes the ‘supplier and consumer responsibility’ score from Hexun’s social responsibility report as an alternative index for robustness testing.

Mediating variables: organizational green cognition and dynamic ability. Following Ref. [53], this study assesses the organizational green cognition of enterprises by examining green responsibility fulfillment and green investment intention. The willingness of enterprises to fulfill green responsibility is measured using the environmental responsibility score from the social responsibility report, while green investment willingness is measured by the sum of the green investment amount plus 1 to take natural logarithm. After standardizing the indicators in these two dimensions, the mean value is calculated as the organization’s green cognition index. Drawing inspiration from previous research [54], this study evaluates the dynamic capabilities of enterprises across three dimensions: perception ability, reconstruction ability, and orchestration ability. Perception ability is assessed using the sum of research and development investment plus 1 to take natural logarithm; reconstruction ability is evaluated based on the proportion of employees with a bachelor’s degree or higher in the total workforce; and orchestration ability is measured by the ratio of net profit to average total assets. After standardizing the indicators in these three dimensions, the average results are considered as the evaluation index of dynamic capability.

Moderating variable: environmental regulation. Under stringent environmental standards, companies will enhance their research and development efforts and adopt pro-environmental technologies to effectively manage pollutant emissions [55]. Previous studies have commonly utilized indicators such as government subsidies, public opinion pressure, or industry competition to assess environmental regulation [56,57]. Building on this existing literature, this study measures enterprise environmental regulation across three key dimensions: government, society, and market. Specifically, the environmental regulation at the government level is measured by taking the natural logarithm of the total number of government subsidies and tax breaks received by agricultural enterprises plus 1; the public opinion pressure at the social level is measured by taking the natural logarithm of the total number of search word frequencies of environmental protection related words in the locations where agricultural enterprises are situated; and the inverse of the Herfindahl index is utilized to measure market-level competition intensity. The indicators of the above three dimensions are standardized, and the result after taking the average value is taken as the comprehensive assessment result of environmental regulation [58,59]. Data for this analysis were sourced from the CSMAR database, with word frequency data on environmental protection manually collected from the Baidu search website.

Control variables: This paper utilizes enterprise nature, establishment years, capital density, market environment, and employee incentive as control variables. Enterprise nature is represented by a dummy variable indicating the holding situation, with a value of 1 for state-owned holding and 0 for non-state-owned holding. The durations of the business’s establishment are calculated as the difference between the observation year and the enterprise’s establishment year. Capital density is defined as the natural logarithm of the ratio of net fixed assets to the total number of employees in the firm’s financial statements. Market environment is measured as the natural logarithm of the total index of consumer trust. Employee incentive is captured through a dummy variable for employee stock ownership, with a value of 1 for employee stock ownership and 0 for non-employee stock ownership.

3.3. Sample Selection

This study focuses on A-share listed agricultural enterprises in the Shanghai and Shenzhen stock markets from 2011 to 2020. Initially, A-share enterprises in these markets were chosen as the starting point. Subsequently, following the criteria outlined in the 2021 China Agriculture-related Enterprise Innovation Report, 204 non-ST (no significant risks or problems) and non-PT (no risk of suspension) agricultural enterprise samples were selected based on having agriculture-related main business income greater than 50% or diversified operation with agriculture-related main business income accounting for more than 30%. Enterprises with more than two years of missing observations and those listed after 2011 were excluded, resulting in a final panel data set of 1380 observations from 138 A-share agricultural listed companies. Missing values in the data were handled by filling GTI data with 0 and utilizing the ‘multiple imputation’ method for other missing data.

4. Empirical Results and Explanations

4.1. Benchmark Regression

Pearson correlation analysis, ADF stationarity test, and Kao test were conducted on the dataset in this study. The findings indicated that the correlation among the key variables aligned with the anticipated patterns, and there were no issues of multicollinearity or non-stationarity in the data. Furthermore, to mitigate the impact of outliers, a 1% winsorization was applied to the data. For ease of comparison, the variables related to symbiotic relationship, organizational green cognition, and dynamic capability were standardized.

Table 1. Benchmark regression results for symbiosis studies.

Variable	(1)	(2)
Symbiotic relationship (Sym)	0.233 ***	0.192 ***
	(0.054)	(0.049)
Environmental regulation (Env)		2.999 **
		(1.121)
Nature of business (Own)		0.060
		(0.075)
Age of establishment (Age)		0.027
		(0.011)
Capital density (Den)		0.089 **
		(0.036)
Market Environment (Mar)		0.308
		(0.058)
Employee motivation (Eso)		0.222 ***
		(0.065)
YearFE	Yes	Yes
RegFE	Yes	Yes
InduFE	Yes	Yes
N	1380	1380
R2	0.500	0.509

Note: Standard errors in parentheses (** p < 0.05, *** p < 0.01).

presents the regression results of a multi-dimensional fixed effects model with robust standard errors. In Column (1), the benchmark regression results are shown without including control variables. The regression coefficient for the symbiotic relationship is 0.233, and the null hypothesis is rejected at the 1% significance level. This indicates that the symbiotic relationship significantly enhances the green technology innovation of agricultural enterprises. Moving to Column (2), the regression coefficient for symbiotic relationship is 0.192, still rejecting the null hypothesis at a significance level of 1%. This reaffirms that symbiotic relationships continue to have a significant positive impact on the green technology innovation of agricultural enterprises even when control variables are considered. Control variables such as establishment years and market environment exhibit negative effects on green technology innovation, indicating that enterprises with strong organizational practices and better product sales prioritize resource allocation towards product production over green technology innovation. Conversely, capital density and employee incentives show positive effects on green technology innovation, suggesting that enterprises with robust capital strength and employee stock ownership plans tend to have higher green technology innovation output. The comparison between Columns (1) and (2) reveals a decrease of 0.041 in the estimated coefficient for symbiotic relationship, indicating that differences in establishment years, capital density, and market environment among enterprises explain 17.60% of the marginal contribution of symbiotic relationship to green technology innovation. The significance level remains at 1%, confirming the substantial promotion effect of symbiotic relationships on the green technology innovation of agricultural enterprises both with and without control variables, thus verifying Hypothesis 1.

4.2. Robustness Testing and Endogeneity Problem Handling

Given the potential issues of omitted variables and reverse causality, it is essential to test the robustness of the model. This paper conducts robustness testing by incorporating control variables, substituting dependent variables, altering independent variables, and modifying the model structure.

Table 2. Robustness test results of symbiosis research.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	Benchmarks Regression	Adding Control Variables	Substitution Dependent Variable	Replacement Independent Variables	Replacement Model	Lag One Period
Symbiotic relationship (Sym)	0.192 ***	0.211 ***	0.148 ***	0.306 ***	0.192 ***
	(0.049)	(0.048)	(0.044)	(0.092)	(0.065)
Symbiotic relationship lag (l.Sym)						0.169 **
						(0.051)
Environmental regulation (Env)	2.999 **	2.644 *	2.898 **	2.861 **	2.861 ***	3.951 ***
	(1.121)	(1.194)	(1.045)	(1.162)	(0.547)	(1.119)
Nature of business (Own)	0.060	0.009	0.060	0.032	0.032	0.057
	(0.075)	(0.077)	(0.065)	(0.084)	(0.077)	(0.076)
Age of establishment (Age)	0.027	0.030	0.028	0.027	0.027	0.026
	(0.011)	(0.012)	(0.006)	(0.010)	(0.007)	(0.012)
Capital density (Den)	0.089 **	0.076 **	0.120 ***	0.092 **	0.092 **	0.100 **
	(0.036)	(0.033)	(0.032)	(0.036)	(0.036)	(0.043)
Market environment (Mar)	0.308	0.276	0.379	0.307	3.525 ***	0.331
	(0.058)	(0.079)	(0.069)	(0.060)	(1.012)	(0.045)
Employee motivation (Eso)	0.222 ***	0.199 ***	0.177 **	0.212 **	0.212 *	0.199 **
	(0.065)	(0.060)	(0.058)	(0.069)	(0.122)	(0.071)
Financing constraints (Cap)		0.218
		(0.071)
External financing (Efin)		0.676 ***
		(0.192)
Peer competition (Comp)		0.970
		(0.573)
YearFE	Yes	Yes	Yes	Yes	Yes	Yes
RegFE	Yes	Yes	Yes	Yes	Yes	Yes
InduFE	Yes	Yes	Yes	Yes	Yes	Yes
N	1380	1380	1380	1380	1380	1242
R2	0.519	0.531	0.540	0.520	-	0.529

Note: Standard errors in parentheses (* p < 0.1, ** p < 0.05, *** p < 0.01).

presents the outcomes of the robustness assessment. Column (1) displays the benchmark regression results for comparative purposes. Column (2) exhibits the regression outcomes with additional control variables including financial constraints ($Cap$), external financing ($Efin$), and peer competition ($Comp$). Column (3) showcases the regression results with the dependent variable replaced by the sum of the total green invention patents acquired and the total green utility model patents acquired plus 1 to take natural logarithm. Column (4) demonstrates the regression outcomes after substituting the primary independent variable with the ratings of ‘supplier and consumer responsibility’ as reported in Hexun’s CSR document. Column (5) demonstrates the regression using a panel Tobit model. In summary, the coefficients representing the symbiotic relationship remain consistent across all tests, and each passes the significance test at the 1% level. This consistency indicates that the conclusion regarding the positive impact of symbiotic relationships on green technology innovation in agricultural enterprises is highly robust.

Considering that the company’s green technology innovation process is lengthy, the impact of symbiotic relationships on green technology innovation may exhibit a time lag. Furthermore, the interplay between symbiotic relationships and the outcomes of green technology innovation could introduce endogeneity issues due to reverse causality. To address this, a one-lag regression on the symbiotic relationship of the core independent variables is employed to assess the impact of this endogeneity problem on the robustness of the model. The test results indicate that the sign of the regression coefficient for the symbiotic relationship lagged by one period remains fundamentally unchanged, and it has passed the 5% significance level test. This suggests that the issue of reverse causation has not significantly altered the main conclusion of this section, thereby demonstrating that the model possesses enhanced robustness.

4.3. Mediating Effect Test of Organizational Green Cognition Under the Regulation of Environmental Regulation

In order to examine the mediating role of organizational green cognition between symbiotic relationship and green innovation intention, as well as the moderating effect of environmental regulation on this mediation, this study conducted regression analysis as presented in

Table 3. The mediation test results of organizational green cognition regulated by environmental regulation.

Variable	(1)	(2)	(3)	(4)
	GI	Cog	GI	Cog
Symbiosis (Sym)	0.192 ***	0.126 **	0.136 **	0.127 ***
	(0.049)	(0.041)	(0.048)	(0.031)
Organizing green cognition (Cog)			0.446 ***
			(0.084)
Symbiotic relationship × Environmental regulation (Sym) × (Env)				0.434
				(0.172)
Environmental regulation (Env)	2.999 **	0.346 ***	2.845 **	0.301 ***
	(1.121)	(0.079)	(1.118)	(0.074)
Nature of business (Own)	0.060	0.086 ***	0.022	0.085 ***
	(0.075)	(0.019)	(0.081)	(0.020)
Age of establishment (Age)	0.027	0.000	0.027	0.000
	(0.011)	(0.001)	(0.011)	(0.000)
Capital density (Den)	0.089 **	0.016	0.082 **	0.014
	(0.036)	(0.009)	(0.035)	(0.008)
Market Environment (Mar)	0.308	0.034	0.293	0.000
	(0.058)	(0.012)	(0.059)	(0.013)
Employee motivation (Eso)	0.222 ***	0.003	0.224 ***	0.004
	(0.065)	(0.011)	(0.066)	(0.011)
YearFE	Yes	Yes	Yes	Yes
RegFE	Yes	Yes	Yes	Yes
InduFE	Yes	Yes	Yes	Yes
N	1380	1380	1380	1380
R2	0.519	0.375	0.522	0.381

Note: Standard errors in parentheses (** p < 0.05, *** p < 0.01).

. Column (1) represents the baseline regression testing the overall impact of symbiotic relationship on green technology innovation. The coefficient for symbiotic relationship is 0.192, and the null hypothesis is rejected at the 1% significance level, indicating a significant influence of symbiotic relationship on green technology innovation. The total effect $c=0.192$. Moving to Column (2), a regression was performed to assess the influence of symbiotic relationship on organizational green cognition. The estimated coefficient for symbiotic relationship is 0.126, with the null hypothesis being rejected at the 5% significance level. This suggests that closer symbiotic relationships lead to a significant improvement in agricultural enterprises’ organizational green cognition. The first stage effect $a=0.126$, confirming Hypothesis 2. Column (3) presents a regression analysis examining the influence of organizational green cognition on green technology innovation, as well as the direct impact of symbiotic relationships on green technology innovation. The estimated coefficient for the symbiotic relationship is 0.136, with the null hypothesis being rejected at the 5% significance level, indicating a direct effect of $c’=0.136$. Similarly, the estimated coefficient for organizational green cognition is 0.446, with the null hypothesis being rejected at the 1% significance level, suggesting that enhancing organizational green cognition can significantly boost green technology innovation. The first stage effect $b=0.446$, confirming Hypothesis 3. Combining the results from Columns (1), (2), and (3), in the regression model with organizational green cognition as the mediating variable, the total effect of symbiosis on green technology innovation is 0.192, with a direct effect of 0.136 and an indirect effect of 0.056. The mediation effect represents 29.17% of the total effect, supporting Hypothesis 4. Column (4) presents the regression result incorporating the regulating variable environmental regulation, specifically by adding the cross-term between environmental regulation and symbiosis. The coefficient of the cross-term is −0.434, indicating a negative impact of environmental regulation on the relationship between symbiosis and organizational green cognition. This suggests a moderating effect, where for each one standard deviation increase in environmental regulation, the slope of the effect of symbiosis on organizational green cognition decreases by 0.434, confirming Hypothesis 5. Comparing the outcomes in Columns (1), (3), and (4), it is observed that under the influence of environmental regulations, the direct effect becomes $a’=0.127-0.434Env$, and the indirect effect becomes $a’\times b=(0.127-0.434Env)\times 0.446$. This implies that the indirect effect is contingent on environmental regulations, illustrating that the mediating effect is influenced by the regulatory environment. The findings indicate that organizational green cognition acts as a mediating factor negatively regulated by environmental regulation between symbiotic relationships and green cognition, establishing a model of moderated mediating effect. Environmental regulation notably diminishes the impact of symbiotic relationships on green technology innovation through green cognition, reflecting the intended function of the Chinese government. Stringent environmental regulations empower agricultural management symbiotic entities to enhance organizational green cognitive capabilities within the confines of government policy. This outcome aligns with prior research, demonstrating a positive influence of environmental regulations on corporate green technology innovation [60].

4.4. Mediating Effect Test of Dynamic Ability Moderated by Symbiotic Relationship

As illustrated in

Table 4. Mediation test results of dynamic capability regulated by symbiosis.

Variable	(1)	(2)	(3)	(4)
	GI	Dyc	GI	GI
Symbiotic relationship (Sym)	0.192 ***	0.019 ***	0.152 **	0.161 ***
	(0.049)	(0.004)	(0.048)	(0.049)
Dynamic capability (Dyc)			2.094 **	2.120 ***
			(0.705)	(0.650)
Symbiosis × dynamic ability (Sym) × (Dyc)				0.922 **
				(0.456)
Environmental regulation (Env)	2.999 **	0.048	2.899 **	2.867 **
	(1.121)	(0.102)	(0.918)	(0.927)
Nature of enterprise (Own)	0.060	0.008	0.077	0.069
	(0.075)	(0.014)	(0.064)	(0.060)
Age of establishment (Age)	0.027	0.002	0.022	0.022
	(0.011)	(0.000)	(0.011)	(0.011)
Capital density (Den)	0.089 **	0.007	0.103 **	0.104 **
	(0.036)	(0.005)	(0.037)	(0.037)
Market Environment (Mar)	0.308	0.058 ***	0.430	0.299
	(0.058)	(0.012)	(0.088)	(0.056)
Employee motivation (Eso)	0.222 ***	0.018	0.185 **	0.179 **
	(0.065)	(0.010)	(0.068)	(0.064)
YearFE	Yes	Yes	Yes	Yes
RegFE	Yes	Yes	Yes	Yes
InduFE	Yes	Yes	Yes	Yes
N	1380	1380	1380	1380
R2	0.519	0.501	0.533	0.533

Note: Standard errors in parentheses (** p < 0.05, *** p < 0.01).

, the first column represents a baseline regression examining the overall impact of symbiotic relationships on green technology innovation. The coefficient for symbiotic relationships is 0.192, with the null hypothesis being rejected at the 1% significance level, suggesting that symbiotic relationships have a total effect of $c=0.192$ on green technology innovation. Moving on to Column (2), a regression is conducted to assess the influence of symbiotic relationships on dynamic capabilities. Here, the estimated coefficient for symbiotic relationships is 0.019, and the null hypothesis is rejected at the 1% significance level, indicating a significant increase in the dynamic capabilities of agricultural enterprises due to symbiotic relationships. This verifies Hypothesis 6 with the first stage effect of $a=0.019$. In Column (3), a regression is performed to analyze the impact of dynamic capabilities on green technology innovation, along with the direct effect of symbiotic relationships on green technology innovation. The estimated coefficient for symbiotic relationships is 0.152, with the null hypothesis being rejected at the 5% significance level, showing a direct effect of $c’=0.152$ on green technology innovation. Additionally, the estimated coefficient for dynamic capabilities is 2.094, with the null hypothesis being rejected at the 5% significance level, indicating a significant promotion of green technology innovation through the enhancement of dynamic capabilities. This validates Hypothesis 7 with the second stage effect of $b=2.094$. Considering the results from Columns (1), (2), and (3), it can be inferred that in the regression with dynamic capabilities as the mediating variable, the total effect of symbiosis on green technology innovation is 0.192, the direct effect is 0.152, and the indirect effect is 0.040, with the mediation effect constituting 20.83%. This confirms Hypothesis 8. Column (4) presents the regression result including the modulating variable symbiotic relationship, specifically the addition of the cross-term between dynamic capabilities and symbiotic relationship. The coefficient for the cross-term is 0.922, indicating a positive regulatory impact of symbiotic relationship on the association between dynamic capabilities and green technology innovation. This suggests that for each standard deviation increase in symbiotic relationship, the effect of dynamic capabilities on promoting green technology innovation will increase by 0.922, thus confirming Hypothesis 9. When comparing the results across Columns (1), (3), and (4), it is evident that under the influence of symbiotic relationship regulation, the direct effect transitions to ‘$b’=2.120+0.922Sym$’ and the indirect effect to ‘$a\times b’=0.019\times (2.120+0.922Sym)$’. This highlights that the mediating effect is contingent on the level of symbiotic relationship. The analysis demonstrates that dynamic capabilities act as an intermediate variable, positively influenced by the symbiotic relationship, in the relationship between symbiotic relationship and green technology innovation. The mediated effect model is established. Research has demonstrated that a robust symbiotic relationship within agricultural enterprises enhances their dynamic capabilities. Furthermore, the development of these dynamic capabilities can significantly foster green technology innovation in such enterprises. Additionally, the symbiotic relationship plays a regulatory role, influencing how dynamic capabilities impact green technology innovation. Specifically, when enterprises effectively communicate green demand and opportunity information through these symbiotic relationships, they are more likely to allocate greater dynamic capabilities towards green technology innovation.

5. Heterogeneity Analysis

To further explore the diverse impact of symbiotic relationships on various models of green technology innovation, we considered green technology innovation, breakthrough green technology innovation, incremental green technology innovation, independent green technology innovation, and joint green technology innovation as the explanatory variables. Core independent variables included symbiotic relationship, organizational green cognition, and dynamic capabilities. These core independent variables were standardized, allowing for comparative analysis based on their coefficients.

In the regression analysis presented in Column (1) of

Table 5. Heterogeneity analysis results of symbiosis research.

Variable	(1)	(2)	(3)	(4)	(5)
	GI	GI_ori	GI_sec	GI_ind	Gi_uni
Symbiosis (Sym)	0.100 *	0.048	0.136 ***	0.070	0.099 **
	(0.047)	(0.038)	(0.038)	(0.047)	(0.034)
Organizing green cognition (Cog)	0.417 ***	0.384 ***	0.251 ***	0.320 **	0.278 ***
	(0.094)	(0.096)	(0.062)	(0.107)	(0.064)
Dynamic capability (Dyc)	2.066 **	1.828 ***	1.366 **	1.800 **	1.114 **
	(0.691)	(0.530)	(0.557)	(0.695)	(0.408)
Nature of enterprise (Own)	0.041	0.034	0.096 *	0.025	0.060
	(0.064)	(0.046)	(0.048)	(0.052)	(0.035)
Age of establishment (Age)	0.022	0.021	0.014	0.013	0.026
	(0.012)	(0.008)	(0.007)	(0.010)	(0.004)
Capital density (Den)	0.096 **	0.003	0.142 ***	0.094 **	0.002
	(0.037)	(0.025)	(0.041)	(0.035)	(0.026)
Market competition (Mar)	0.414	0.227	0.479	0.425	0.113
	(0.090)	(0.054)	(0.103)	(0.088)	(0.055)
Employee motivation (Eso)	0.186 **	0.106 *	0.135	0.247 **	0.087
	(0.068)	(0.055)	(0.089)	(0.078)	(0.037)
Environmental regulation (Env)	2.756 **	2.207 ***	2.346 **	3.017 ***	0.829
	(0.922)	(0.672)	(0.825)	(0.835)	(0.849)
YearFE	Yes	Yes	Yes	Yes	Yes
RegFE	Yes	Yes	Yes	Yes	Yes
InduFE	Yes	Yes	Yes	Yes	Yes
N	1380	1380	1380	1380	1380
R2	0.535	0.482	0.545	0.531	0.350

Note: Standard errors in parentheses (* p < 0.1, ** p < 0.05, *** p < 0.01).

, where the total amount of green technology innovation was the dependent variable, the coefficient for symbiotic relationships was found to be 0.100, failing to reach significance at the 5% level. On the other hand, the coefficients for organizational green cognition and dynamic capabilities were 4 times and 20 times larger than that of symbiosis, respectively. In comparison to the baseline regression, the coefficient for symbiotic relationships decreased by 47.92% and lost significance, aligning with previous findings on the mediating role of organizational green cognition and dynamic capabilities. All the above indicate that the direct promoting effect of symbiotic relationship on green technology innovation is relatively weak, and it is more through the intermediary effect of organizational green cognition and dynamic ability. Moving on to Column (2), which focuses on breakthrough green technology innovation, the analysis reveals a non-significant coefficient of 0.048 for symbiotic relationships. In contrast, organizational green cognition and dynamic capabilities showed significant coefficients of 0.384 and 1.828, respectively, at the 1% significance level. This indicates that high originality in breakthrough green technology innovation is primarily driven by the enterprise’s dynamic capabilities and organizational green cognition, with the quality of symbiotic relationships playing a lesser role in influencing the output of breakthrough green technology innovation. Column (3) examines the impact of progressive green technology innovation, showing a symbiotic relationship coefficient of 0.136 and an organizational green cognition coefficient of 0.251, both significant at the 1% level. The coefficient for dynamic capabilities is 1.366 with a 5% significance level, indicating that while enterprise dynamic capabilities remain important for incremental green technology innovation, the requirements are lower compared to breakthrough innovation. Enterprises with strong symbiotic relationships can quickly adjust and improve technology through shared information, enhancing progressive green technology innovation. Columns (4) and (5) conduct regressions with independent green technology innovation and joint green technology innovation as explanatory variables, respectively. The coefficients for symbiotic relationships are 0.070 and 0.099, with the former not passing the significance test and the latter significant at the 5% level. This indicates that a good symbiotic relationship can offer enterprises more information and resources for joint innovation, potentially even participating as joint innovators, thus promoting green technology innovation. However, the impact on independent green technology innovation is not as pronounced. Overall, symbiotic relationships have a weak direct effect on green technology innovation, primarily influencing it indirectly. Organizational green cognition and dynamic capabilities play a crucial role in all types of green technology innovation, while symbiotic relationships primarily foster incremental and collaborative green technology innovation.

6. Conclusions and Insights

A key distinction between agricultural enterprises and industrial enterprises lies in the intricate relationships and emotional connections that leading agricultural enterprises maintain with upstream and downstream farmers, herdsmen, new agricultural business entities, and cooperatives. These connections are vital for rural revitalization, agricultural industrialization development, and green transformation. To examine the role and mechanism of symbiotic relationship in green technology innovation within agricultural enterprises, this study utilizes balanced panel data from listed agricultural enterprises in China spanning from 2011 to 2020. It constructs a moderated intermediary effect model based on literature review and theoretical analysis, and conducts empirical tests to draw the following conclusions: The baseline regression results indicated that symbiotic relationships positively contribute to promoting green technology innovation in agricultural enterprises. Further regression analysis revealed that organizational green cognition within the symbiotic relationship plays a partial mediating role in green technology innovation, with this effect being negatively influenced by environmental regulations. Additionally, dynamic capabilities were found to partially mediate the relationship between symbiotic relationships and green technology innovation, with this mediation being positively influenced by the strength of symbiotic relationships. Heterogeneity analysis demonstrated that while symbiotic relationships have a relatively weak direct impact on green technology innovation, they primarily affect innovation through indirect pathways. Specifically, symbiotic relationships were found to significantly promote gradual green technology innovation and joint green technology innovation, but had less of a direct impact on breakthrough and independent green technology innovation.

The indirect driving effect of symbiosis on green technology innovation in agricultural enterprises is often overlooked by many. Empirical studies have shown that symbiotic relationships can actually boost green technology innovation by increasing organizational green cognition and enhancing dynamic capability. This highlights the added value that symbiotic relationships bring, which cannot be achieved when entities operate independently. Therefore, the significance of symbiotic relationships in driving green technology innovation should not be underestimated. The advancement of low-carbon agriculture and green technology innovation in agricultural enterprises necessitates the integration of green development principles across various stages such as production, processing, and sales, along with the collaboration of symbiotic units like agricultural leaders, cooperatives, farmers, and herdsmen. By fostering a two-way interaction between leading agricultural enterprises and other symbiotic units involved in agricultural production and operations, there is a mutually beneficial relationship that enhances the incentive for green technology innovation within leading agricultural enterprises. On the one hand, the incentive effect of symbiotic units on green technology innovation in leading agricultural enterprises should be enhanced. Some agricultural production and operation entities within symbiotic units have demands for ecological environments such as soil and water conservation and afforestation in agricultural production. This stimulates agricultural enterprises to intensify research and development in green innovation of agricultural chemicals like pesticides and fertilizers, ultimately reducing environmental pollution. On the other hand, enhancing the scientific and technological driving effect of leading agricultural enterprises on symbiotic units should promote the adoption of advanced agricultural production technology and equipment by cooperatives, farmers, and herdsmen. For instance, precision fertilization and irrigation can improve agricultural productivity, while green waste treatment methods can convert waste from agricultural production into biomass energy and organic fertilizer, thereby reducing greenhouse gas emissions and preventing unnecessary waste of resources and environmental pollution. In the practice of green technology innovation, agricultural enterprises should focus on improving the stability, depth, and breadth of symbiotic relationships, as well as on enhancing the construction and management of such relationships. Leveraging symbiotic relationships can help agricultural enterprises secure more market share, policy information, social connections, and financial and technical support. Government policies should be implemented to support the establishment of agricultural symbiosis networks, foster strong relationships and emotional bonds between leading agricultural enterprises and symbiosis units, and enhance the coordinated development of the agricultural industry chain.

The study’s findings are limited by the use of data from Chinese agricultural listed companies, potentially affecting the generalizability of the results to developed countries and other developing nations. Furthermore, differences in agricultural entities across countries could influence the relevance of the findings. Future research should include data from developed countries and other developing nations to confirm and expand upon the assumptions presented in this article.