The Docking Mechanism of Public and Enterprise Green Behavior in China: A Scenario Game Experiment Based on Green Product Classification

Abstract

In the international context of pursuing green development, a good connection between the production and consumption ends is a necessary condition for forming a diversified governance interaction system. The academic community has conducted in-depth discussions on green behavior among enterprises and the public. However, few studies have focused on the docking process between the subjects. Therefore, this paper takes Chinese participants as an example to design a game experiment, considering the role of government regulation. From a dynamic, interactive perspective, it explores the motivation for green behavior between the public and enterprises, as well as the effectiveness of government subsidies. The research results indicate that: (i) The public is willing to burden a premium for the green behavior of enterprises, which has significant differences in substantive and symbolic green behavior. (ii) In green consumption scenarios, public purchasing information can be transmitted to enterprises, but the reverse path has not yet been verified. (iii) Government subsidies have a strong incentive effect on public green behavior. (iv) There are significant differences in the impact of public environmental awareness on public green behavior, but there is no significant difference in the green behavior of enterprises. The research conclusion provides theoretical support for the study of public and enterprise behavioral variables and provides a practical basis for enterprise management and government governance. This will facilitate the construction of a systematic and effective environmental governance framework, thereby promoting coordinated green development among entities.

1. Introduction

Since entering the 21st century, environmental issues have increasingly become the focus of governance. Due to the diversity of stakeholders and the complexity of interests involved, it poses serious challenges to government-led environmental governance. Therefore, the country has introduced a series of policies targeting various entities to alleviate the ecological pressure brought about by economic development. It is essential to build a green action system for the common governance of the government, enterprises and the public. There are complex influence and restriction relationships among the co-governance subjects. The behavior of one party may cause attitude changes and behavioral reactions in others. This relationship ensures that Multi-Agent Coordination achieves more effective results under the guidance of common goals than the environmental governance alone [1]. Therefore, government, enterprises and the public should maintain the “cooperation and coordination” mechanism, give play to the maximum superposition effect of the implementation effect of green behavior, and establish the “restriction and balance” mechanism on this basis [2]. Under various interaction scenarios, in order to realize coordinated and linked green development [3], subjects should apply to different mechanisms and transform the conflict of interests among the parties into an important force of environmental governance.

In the process of dynamic interaction, subjects form a complex interaction relationship network. Sorting out the different action paths and mechanisms in the network is of great significance to the implementation of subject difference governance. From the perspective of the government, the government mainly uses the mandatory and authoritative norms of environmental regulation to advocate public green behaviors (PGB) and enterprise green behaviors (EGB). For example, government-led green publicity promotes the generation of EGB [4]. However, this process is not one-way, the emission reduction capacity of enterprises will also have an impact on the government’s regulatory strategy [5]. From the perspective of enterprises and the public, on the one hand, based on the virtuous cycle of the green market, enterprises provide green products that determine the public’s purchase range. After the public pays, the public’s purchase behavior will feed back the generated demand information, which will affect the enterprise’s green investment plan. In this process, the government will adopt corresponding reward, punishment, and subsidy policies to make up for the imbalance in the current market. On the other hand, public supervision can also participate in the process of environmental governance [6]. When realizing that EGB is difficult to maintain or will even damage the environment, the public will reflect it to the government to indirectly improve the environmental situation [7]. Research from the perspective of individual subjects shows that there is externality in the green behavior of enterprises and the public. It is still difficult for China to implement green behavior spontaneously at the current stage. In a market-oriented context, the interactive relationship between the supply side and the demand side can not only guide the green behavior of both enterprises and the public, but also affect the implementation effect of policies.

Therefore, this paper aims to identify the green behavior motivations of enterprises and the public, determine the causal relationship between the decision-making variables, and analyze the differences in the intervention effects of government subsidies in the docking mechanism. The research provides effective solutions for establishing a benign docking mechanism between enterprises and the public to achieve spontaneous implementation of green behavior by market entities. However, the existing literature mostly uses rational methods, such as game theory and policy simulation, or conducts research on the influencing factors of one party’s behavior. There are few studies based on the interaction scenarios between both parties. In order to conduct in-depth research on the interaction effect, this paper constructs a green behavior docking mechanism between enterprises and the public based on the interaction scenario of market transactions.

This paper provides theoretical and application insights for the discussion of the relationship between Chinese market entities. The theoretical value of this study is highlighted in two aspects: (i) From a research perspective, previous literature has mainly studied the impact mechanism of PGB or EGB from a single aspect. This paper overcomes the limitations of isolated research on different entities. By using game experiments, this study analyzes the two-way interaction relationship between the public and enterprises in different scenarios. It is convenient to obtain timely and real feedback from the dynamic adjustment of both parties that can explore the mechanism of behavioral variables between subjects. (ii) Regarding research content, existing research mostly verifies the subsidy effect of the government on overall green products. Based on the Value Belief Norm Theory (VBN), this paper further subdivides the egoistic and altruistic green attributes of green products. It explains the differences in the effectiveness of government subsidies for different green products and enriches the research on the docking mechanism between enterprises and the public based on green product classification.

The application value of this paper’s scenario game experiments to the green behavior of microentities provides reliable practical evidence for the green development of production and consumption in China. It also has important reference significance for research on the docking mechanism between enterprises and the public in other countries. In addition, mastering the practical rules is beneficial for the government to issue targeted policies, and China’s efforts to promote environmental ecology will drive the world to continue moving towards sustainable development. The arrangement of this paper is as follows: Section 2 consists of a literature review. Section 3 is composed of the theoretical basis and research hypothesis. Section 4 is the experimental design. Section 5 is the model construc-tion. Section 6 is discussion. Section 7 comprises the conclusion and research implication, and the Section 8 is limitation.

2. Literature Review

In order to grasp the current research status of the docking mechanism between enterprises and the public’s green behavior, this paper filters the latest research on the ecological environment, psychology, and behavior, further selecting the relevant studies using the green behavior of the subject as the keyword. Then, this paper categorizes the literature into three parts: the correlation mechanism, motivation of green behavior, and government subsidy. It summarizes the relevant research on the behavioral mechanisms of PGB and EGB. Finally, by comparing the research content, methods, and perspectives of existing literature, this study extracts research design methods that can be used as a reference and support further research breakthroughs.

2.1. Correlation Mechanism between PGB and EGB

Existing research, when discussing the interactive relationship between the PGB and EGB, often focuses on the perspective enterprise or the public, incorporating another subject into the analysis framework as an external influencing factor, but it lacks attention to the two-way docking mechanism. (1) The role of enterprises towards consumers is manifested as consumer response, where enterprises guide PGB by selling green products. The differences in packaging, price [8], brand [9], and other attributes of green products can lead to differences in the public’s willingness to pay. (2) The impact of the public on enterprises is manifested as consumers being the core stakeholders in direct transactions with enterprises, and the public’s willingness and beliefs influence the innovation and reform of enterprises [10]. It is obvious that the above studies are essentially exploring unidirectional mechanisms of action or response, separating the dynamic influence processes between subjects. Therefore, the formation of a docking mechanism framework between the public and enterprises is the research foundation of this study.

Notably, previous studies have shown that different EGB convey different signals to the public. For example, corporate social responsibility increases individual level green self-classification, green emotional commitment, and green self-esteem, thereby increasing environmental behavior [11]. The marketing of green products by enterprises can affect public purchasing behavior [12], and even different marketing methods (such as energy-saving information exposure [13] and environmental advertising on social media [14,15]) can have opposite effects. However, when classifying EGB, scholars often analyze it from the perspective of enterprises, lacking integration and comparison based on consumer response perspectives. For example, the entire lifecycle based on product development, production, and recycling is divided into green technology innovation, green production, and terminal processing [16]. This classification is difficult to reflect the public’s response preferences. Therefore, this paper draws inspiration from the perspective of institutional schools. EGB is divided into the following two types based on the strength of the signal EGB conveyed to the outside world: One is symbolic EGB, which is difficult to distinguish authenticity and transmit green information to the public. Another is the substantive EGB, which can fulfill commitments and improve green environmental protection standards in the production and operation activities of enterprises. Based on this, this study explores the differences in public willingness to pay for different green behaviors of enterprises.

2.2. Motivation of Green Behavior

From existing research, scholars mainly focus on the micro-level motivation of green behavior among unilateral entities and have not yet formed a complete motivation framework for the interaction mechanism. Scholars believe that (1) PGB is mainly influenced by psychological factors (including environmental attitudes [17], environmental protection values [18], altruistic values [19], etc.) and external situational factors (including environmental governance [20], and cultural atmosphere [21]). When consumers’ psychological needs are reflected in public purchasing decisions, they are more focused on measuring the level of benefits and costs manifested as individual behavioral preferences. (2) Exploring the impact of internal factors (including enterprise size, maturity, intangible assets [22], etc.) on EGB from the perspective of internal governance, or the stimulating effect of external pressures [23] (including environmental governance [24], social media, and public attention [25]). Although the above motivations play a role to some extent in the interaction between the public and enterprises, in essence, PGB places more emphasis on improving environmental conditions at the individual level [26], while enterprises aim at obtaining other benefits. The impact of management group decision making is usually greater than individual willingness. Therefore, the motivation of one party’s behavior will generate external pressure through the interactive chain, which will have an impact on the other subject [27]. Based on this, it is necessary to further explore the detailed impact of PGB and EGB motivations in the interactive process.

2.3. Government Subsidy

Considering the effectiveness of government subsidies, it is necessary to include the supply and demand sides in the analysis framework. However, current research mostly focuses on only one aspect. Studies have shown that the effect of government subsidies on the public is roughly a positive stimulus to demand [28]. At the same time, due to the strong subjective initiative of individuals, there is uncertainty in the transformation effect of public behavior, making it difficult to directly form beneficial strategies. Scholars have conducted in-depth research on the relationship between government subsidies and EGB, but the research results are inconsistent. On the one hand, scholars have found that government subsidies have an “incentive effect” on enterprises’ green performance, which is reflected in the fact that the government can provide certain financial support to carry out relevant green activities [29,30]. However, this effect should probably fall back after the decline of subsidies [31]. On the other hand, a few studies have found that government subsidy has a “crowding out effect”, manifested as government subsidies causing enterprises to transfer funds originally used for R&D activities to other projects, thereby having a negative impact on the green innovation output of enterprises [32,33]. Therefore, the effectiveness of government subsidies for enterprises in the context of interaction between the public and enterprises still needs further exploration. This paper takes the public and enterprises as different subsidy targets and tests the effectiveness of government subsidies through comparative analysis.

Although there is a small amount of literature discussing the differences in the effectiveness of subsidies for different entities, there is still a lack of detailed analysis of the interaction effects. Huang et al. [34] found that subsidies for enterprises are better than subsidies for consumers through vertical differential subsidies between green and non-green products. However, Lei and Sai [35] believe that subsidies to consumers are better than subsidies to enterprises. These achievements provide important insights for the study of this paper. However, their research paradigm is limited to multi-party games and simulation. Scholars establish a rational framework to consider government subsidies in investment decisions and obtain optimal decision making through cost and benefit trade-offs. Games have advantages similar to economic experiments, such as being able to reflect the interests of stakeholders through simulation. However, in the Perfect rationality game process, the flexibility of behavior variables is limited, which may lead to an overly ideal result. Economic experiments can, to some extent, solve this problem. Designers test the effectiveness of single or multiple factor combinations by adjusting the experimental background and process. It may also lead to unexpected results due to incomplete rational behavior of participants, providing a practical basis for in-depth research.

2.4. Summary

Scholars have made significant efforts and achieved rich results in the fields related to PGB and EGB. However, there are some gaps need to be filled. From a research perspective, current research mainly focuses on the interaction between enterprises and the public from a one-way process. They highlight the guidance of enterprises on public green consumption, focusing less on the two-way interaction mechanism in multiple scenarios. From the perspective of the research methods, theoretical games or actual data from markets are often used. The data have abstract characteristics, making it difficult to analyze the docking behavior between entities in detail. To expand the methods of data acquisition and achieve two-way communication, this paper designs a scenario game experiment, where participants play economic roles in a simulated market, experience economic rules, and make decisions with real economic consequences. During the simulation process, real-time reactions between enterprises and the public under external conditions are recorded to reduce major deviations between decision-makers’ wishes and behaviors. The dynamic characteristics of experimental data are utilized to achieve collaboration and constraints, providing reference suggestions for supply and demand balance.

3. Theoretical Basis and Research Hypothesis

3.1. Theoretical Basis

The Value Belief Norm Theory (VBN), defined as value and belief, can affect an individual’s behavior by encouraging them to make the right choices and reflects three orientations that affect decision making regarding environmental consequences, including egoism, altruism, and ecologism [36]. This is one of the basic theories for analyzing PGB [37], covering the impact of internal and external factors such as green awareness, green product prices, and green policies [38]. It is also applicable to the research process of the motivation of the subject’s green behavior. Based on the context of this paper, in the process of green consumption, on the one hand, products with different attributes will reflect the different values and purchasing needs of the public, which will also generate different willingness to pay (WTP, the degree to which consumers are willing to bear the premium for green products) [39]. On the other hand, enterprises make decisions on the level of green investment out of egoistic and altruism.

This paper combines VBN theory and refers to the practice of Sheng et al. [40], further dividing green products into altruistic and egoistic green products. Altruistic green products mean that the product’s performance is similar to that of traditional products, and consumers cannot obtain additional benefits other than environmental benefits. For example, mercury-free batteries can reduce soil pollution but consumers cannot obtain actual benefit feedback in a short time. Egoistic green products mean that, in addition to environmental benefits, consumers can obtain additional personal benefits from the use or recycling process, such as health or energy-saving benefits. For example, the formaldehyde-free characteristics of environmental protection coatings are beneficial to consumers’ health, and energy-saving appliances can reduce consumers’ electricity costs. Therefore, in this paper, the docking mechanism between the public and enterprises revolves around green products (as shown in Figure 1), and the purchase/investment situation can reflect the measurement of the subject’s altruistic value. On this basis, this paper explores the differences in the effectiveness of government subsidies for different objects.

3.2. Research Hypothesis

Signal transmission theory is used to solve the problem of information asymmetry between trading parties. In reality, it is difficult for consumers to accurately evaluate which green products or services are worth purchasing. The behavior of enterprises can convey different signals to the public. When the information contained in the signal meets the public’s fair price expectation [41], it will be more likely to form a green behavior motivation. Therefore, if a company can effectively transfer time or economic costs to consumers and generate trust consumption information for the public [42], it will be beneficial to improve its WTP. Although both substantive and symbolic green behaviors can encourage consumers to have a positive and positive connection between businesses and environmental protection, when consumers create more specific perceptions of corporate environmental protection, they are more likely to generate positive opinions. Therefore, with the enhancement of market transparency, in an environment with high participation in information search, substantive environmental behavior can bring more reputation [43]. Based on the above analysis, we propose the below hypothesis:

H1. 

The public’s WTP for substantive green behavior of enterprises is higher than that for symbolic green behavior.

3.2.1. Motivation of PGB

The public’s motivation for green behavior is a continuous process of measuring cost-effectiveness. The improvement in public environmental awareness extends the public’s interests to the comprehensive effect of economic and environmental benefits. Physiological and safety needs are fundamental components of Maslow’s hierarchy of needs. In the case of insufficient funds, in order to meet basic living needs, the majority of the public will pay attention to the functional benefits brought by green products, such as lower prices and longer service life [44]. With material satisfaction, consumers begin to shift towards spiritual needs. Green Consumer behavior is a socially responsible behavior, and consumers can gain social respect and recognition.

The theory of reasoned action (TRA) points out that the possibility of an individual implementing a specific behavior is closely related to his or her behavior intention, and the stronger the individual’s behavior intention, the more likely he or she is to implement the behavior. When the public has a high level of environmental awareness, in the same situation, they resonate with the green information conveyed by the enterprise and pay more attention to the green attributes of the product. With the improvement in environmental awareness, environmental benefits have become a driving force for stimulating green consumption [45], thereby implementing green behavior into daily life. In short, only when consumers’ basic needs are met, that is, when their own interests are maximized, the public will be attracted to sustainable benefits, pursue green attributes, and purchase green products [46]. Therefore, consumers’ green purchasing behavior is actually the result of the comprehensive effect of environmental awareness, economic benefits, and environmental benefits. Therefore, this article proposes the following assumptions:

H2a. 

Benefit-oriented motivation positively drives the public to implement green consumption behavior.

H2b. 

Environment-oriented motivation positively drives the public to implement green consumption behavior.

H2c. 

Public environmental awareness promotes the public to implement green consumption behavior.

3.2.2. Motivation of EGB

As a collection of economic activities, the EGB is driven by expected profits. However, driven by the social background, enterprises can gain unique competitive advantages by focusing on environmental protection while maximizing profits, reducing environmental pollution, protecting natural resources and the environment, and striving to achieve a natural balance in nature [47]. Enterprises that adopt green behavior are more sustainable than those that do not [48]. In addition, demand is the source of economic incentives for enterprises, and consumer preferences will enable enterprises to focus on the production of green products [49]. According to the new Industrial organization, market demand is an important force that determines the speed and direction of innovation activities. Enterprises are more likely to benefit from markets with demand or potential. Market demand is the source of sustainable products and services, and the public plays the role of a change agent in emerging markets [50]. However, the degree to which the market value of a green product or service can be realized is often uncertain. The uncertainty of market demand is transmitted to enterprises on the production and supply side, resulting in insufficient expectations of green behavior returns. Therefore, a stable public behavior motivation will motivate enterprises to make green production decisions. Based on the above analysis, this article proposes the following assumptions:

H3a. 

Benefit-oriented motivation positively drives enterprises to implement green investments.

H3b. 

Environment-oriented motivation positively drives enterprises to implement green investments.

H3c. 

Demand-oriented motivation can positively encourage enterprises to implement green investments.

3.2.3. Government Subsidy

The theory of externality points out that there is a deviation between the subject and society in the benefits and costs generated by the subject’s behavior. In the process of spontaneous market interaction, both enterprises and the public can choose to give up green behavior. On the one hand, the dual externality between environmental pollution and PGB lead to difficulties in enterprise research and development [51]. Economic costs, time costs and opportunity costs are obstacles in the process of eco-investing. At this point, the government can influence EGB through environmental regulation, environmental investment, and financial subsidies [52]. Institutional theory suggests that when the institutional foundation related to green innovation is relatively weak, government subsidies to enterprises can share some costs, reduce risks, and promote enterprise development [53]. Based on the above analysis, this article proposes the following assumptions:

H4a. 

The government’s subsidy to enterprises’ altruistic green products will promote enterprises’ investment in altruistic green products.

On the other hand, PGB has positive externality. It can directly or indirectly improve the environment and achieve the sustainable development of social ecology. However, the prices of green products and services are usually higher than those of ordinary products [54]. The public of bounded rationality will adopt different strategies according to their own needs [55]. When the public’s awareness of environmental protection is weak, they may make decisions that harm the environment due to short-term interests. Then, while guiding public green behavior, government subsidies also promote the formation of public environmental awareness, enabling them to consciously implement green behavior [56]. This paper proposes the following assumptions:

H4b. 

The government’s subsidy to the public’s altruistic green products will promote the public’s investment in altruistic green products.

All assumptions show as Figure 2.

4. Experimental Instructions

This experiment recruited a total of 192 subjects who were college students. The sample size was detected with 95% power using a one-sided t-test at α = 0.05. According to Wright [57], sample sizes larger than N = 100 can be expected to yield reasonably precise estimates within a Rasch scale calibration. The students represented a relatively homogeneous group in terms of age and education level, which enabled researchers to control the characteristics of the participants. This is conventional practice for conducting similar experiments [58].

The composition of the recruits is as follows: There are 78 male students, accounting for 40.64%, and 114 female students, accounting for 59.36%. The educational background of the subjects covered 60 undergraduate, 105 master’s, and 27 doctoral students, accounting for 34.38%, 54.69%, and 10.93% respectively. There are 60 students who are single children in their family, accounting for 31.20%, and the remaining 132 students have siblings, accounting for 68.80%. There are 66 students in the top 30% of the class, accounting for 34.38%, 84 students in the middle, accounting for 43.75%, and 42 students in the bottom 40%, accounting for 21.87%. In addition, there are 64 student cadres, accounting for 28.13%; 60 students with education loan experience, accounting for 31.25%, and 42 students who had economic experiment experience, accounting for 21.88%.

The design of this experiment is based on the public’s different purchasing behaviors due to cost-benefit considerations [59], reflecting their implementation tendencies towards green behavior through the similarities and differences in investment choices. There are three experiments: Experiment 1 measures the public’s WTP for different EGB. The second experiment is a market transaction experiment, which analyzes the investment/purchase behavior motivation of enterprises/the public. In Experiment 3, the external stimulus of government subsidy is added on the basis of Experiment 2, and the similarities and differences between PGB and EGB are explored by grouping. During the experiment, the subjects hold some experimental coins at the beginning and need to make investment choices for various activities. After each, except Experiment 1, participants can gain real monetary rewards at 5:1, according to the average value of the final account balance in each round.

4.1. Experiment 1: Public WTP for EGB

This experiment has two aims: The first is to measure the public’s WTP for different EGBs. The second is to measure differences in consumers’ WTP for the different EGBs when buying egoistic green products and altruistic green products.

4.1.1. Experimental Process

(1)

Pre-Experiment

The main task of the pre-experiment is to select appropriate green products for the scenario simulation. We prepare to select an egoistic green product from energy-saving air conditioners, energy-saving refrigerators, and green coatings. Alternative altruistic green products include fluorine-free refrigerators and mercury-free batteries. Target products should have the characteristics that college students have purchasing experience in daily life and measure its general value. After a group discussion and expert review, this paper selected the energy-saving lamp as the egoistic green product, and the mercury-free battery as the altruistic green product. In addition, according to the specific classification of Wang and Zhao [60], specific behaviors are given to them: For energy-saving lamps, the behaviors are expressed as “improve green product technology, make products more energy-saving”, “carry out green lighting energy-saving certification for your own products”, “the publicity of product energy-saving degree is consistent with the actual situation”, “use recyclable and degradable packaging during transportation”, “pay attention to the disclosure of enterprise environmental information” and “use non-toxic, harmless, environmentally friendly and energy-saving raw materials”. For the green behavior of mercury-free batteries, the expression of enterprise behavior is “actively carry out green product certification”, “improve green product technology to make products more environmentally friendly”, “the publicity of product environmental protection recycling is consistent with the actual situation”, “use recyclable and degradable packaging during transportation”, “pay attention to the disclosure of enterprise environmental information”, and “purchase healthy and environmentally friendly raw materials”.

(2)

Formal experiment

The formal experiment recruited 192 subjects divided into two groups. Two groups filled in the form under two kinds of goods, respectively. The prompt words were: if you want to buy energy-saving lamps, the average market price of energy-saving lamps is 100 experimental coins (10 experimental coins ≈ 1 RMB). Now you have 100 experimental coins in your hand. During the process of searching for product information, you learned that relevant enterprises have implemented the following green behaviors. How many experimental coins are you willing to pay for these behaviors when purchasing energy-saving lamps? In another test of WTP for mercury-free batteries, the keyword of energy-saving lamps was changed to mercury-free batteries. In the experiment, mutual communication is not allowed, and the subjects need to make decisions independently.

4.1.2. Experimental Result

The basic statistics of the experimental results are shown in

Table 1. Statistics of the public’s WTP for EGB.

EGB Classification	Energy-Saving Lamp			Mercury-Free Battery
	N	Mean	SD	N	Mean	SD
Green technologies innovation	96	50	33.27	96	56.25	24.46
Green management innovation	96	43.13	31.77	96	51.875	29.71
Green marketing	96	43.13	28.92	96	46.875	27.5
Green logistics	96	40.63	31.72	96	48.75	31.38
Environmental information disclosure	96	40	31.62	96	46.25	35
Clean production	96	50	27.81	96	55	28.05

and Figure 3a (to deduct the impact of environmental awareness, the average of each group is deducted from the investment). The public’s WTP for green technologies and clean production is higher than other behaviors, especially in the purchase of energy-saving lamps. In addition, the subjects of mercury-free battery packs have higher overall WTP than the subjects of energy-saving lamp packs, which makes it impossible to directly compare WTP for egoistic green products and altruistic green products. Table 1 also shows the descriptive statistics of WTP, which can be obtained from the average value. Under the purchase scenario of energy-saving lamps, the public’s WTP for green technology and clean production is higher. In the whole product life cycle, these two items are EGB in the production and recycling stages, indicating that compared with the symbolic green behaviors (green management innovation, environmental information disclosure, and green marketing), the information transmitted to the public by the behavior with substantive green behaviors (green technologies innovation, green logistics, and clean production) can stimulate the public’s WTP (see as Figure 3b), which supports H1.

4.2. Experiment 2: Market Transaction

Green products need to carry out some changes and innovations on the basis of traditional products. Tilikidou and Delistavrou [61] observed that these innovations are cost-oriented. However, with the improvement in public environmental awareness, consumers are willing to pay part of the cost for long-term environmental benefits. This paper considers the green investment/purchase activities of enterprises/the public as the voluntary supply of public goods. Based on R. Mark Isaac’s voluntary contribution mechanism (VCM), a scenario game experiment was designed to study the behavioral variables of enterprise investment and consumer purchase. This experiment relies on situational interaction to verify the green behavior motivation of enterprises and the public, and to explore the mechanism of action between behavior variables.

4.2.1. Experimental Process

Before the experiment, the experimental staff will introduce the characteristics of egoistic green products and altruistic green products with examples, then randomly assign subjects numbers. Subjects are divided into two groups according to odd and even numbers. Two groups conducted an investment/purchase experiment on behalf of the enterprise and the public. All subjects have 100 experimental coins at the beginning. The enterprise is faced with the choice of investing x (0~100) to invest altruistic green products and y (0~100) to egoistic green products (x + y ≤ 100); the public choose a (0~100) experimental coins to buy altruistic green products and b (0~100) experimental coins to buy egoistic green products (a + b ≤ 100). The minimum unit of investment and purchase of products is one coin. The remaining experimental coins are kept in their accounts. After the end of each round, the total income and environmental income obtained by each party will be announced to both parties. Five rounds will be repeated without changing the grouping.

The income calculation method is: If enterprises and the public invest in altruistic green products at the same time, their transaction is successful. The situation is as follows: Each unit of the experimental coin will generate 20% environmental spillover income, 1.2 times the total income, which will affect the enterprise and the public on average (both can obtain 0.6 (x + a) income). For egoistic green products, in addition to the average income, the public can obtain 10% additional energy-saving income in the process of using the products (the enterprise can obtain 0.6 (y + b) income, and the public can obtain 0.7 (y + b) income).

When the sale fails, if the enterprise invests in the product but the public does not choose to pay, the enterprise will not receive any income and the investment costs will be deducted. If the public is willing to buy but the enterprise does not invest in the product, the public fails to pay and the experimental coins remain in the personal account. The process simulates the sales failure in a real situation: If the enterprise fails to sell the investment products successfully, its investment is difficult to recover and no income is generated.

In this round, the account balance Z of the enterprise and the public in each case is expressed as:

Both products were successfully sold:

• Enterprise: ${\mathrm{Z}}_{\mathrm{e}}=100-\mathrm{x}-\mathrm{y}+0.6(\mathrm{x}+\mathrm{a})+0.6(\mathrm{y}+\mathrm{b})$
• Public: ${\mathrm{Z}}_{\mathrm{p}}=100-\mathrm{a}-\mathrm{b}+0.6(\mathrm{x}+\mathrm{a})+0.7(\mathrm{y}+\mathrm{b})$

Successful sales of altruistic green products, failed sales of egoistic green products:

• Enterprise: ${\mathrm{Z}}_{\mathrm{e}}=100-\mathrm{x}-\mathrm{y}+0.6(\mathrm{x}+\mathrm{a})$
• Public: ${\mathrm{Z}}_{\mathrm{p}}=100-\mathrm{a}+0.6(\mathrm{x}+\mathrm{a})$

Successful sales of egoistic green products and failure of altruistic green products:

• Enterprise: ${\mathrm{Z}}_{\mathrm{e}}=100-\mathrm{x}-\mathrm{y}+0.6(\mathrm{y}+\mathrm{b})$
• Public: ${\mathrm{Z}}_{\mathrm{p}}=100-\mathrm{b}+0.7(\mathrm{y}+\mathrm{b})$

4.2.2. Experimental Result

The investment trend is shown in Figure 4. The investment is divided into four parts: enterprise altruistic investment (EAI), enterprise egoistic investment (EEI), public altruistic investment (PAI), and public egoistic investment (PEI). The trend shows that the public’s WTP for egoistic green products is higher than altruistic green products, and the gap is gradually widening. For enterprises, at the beginning, in order to meet the market demand of the public, EEI is greater than EAI. However, due to their own interests and the psychology of “free riding”, the investment in green products will be gradually reduced by comparing the benefits obtained with the public in the previous rounds, but the fluctuation range is small compared with the public.

4.3. Experiment 3: Government Subsidy

Driven by the interests of egoistic green products, under the assumption of rational people, the public will choose to invest more in egoistic green products, which will gradually reduce the sales of altruistic green products. However, in order to realize the greening of products comprehensively, the government needs to provide subsidies for altruistic green products. There are two forms of government subsidies: One is to directly subsidize the public, and the public will obtain a certain proportion of discount when buying altruistic green products. The other is to subsidize enterprises and pay part of the costs for cleaner production of enterprises, such as tax reduction and exemption. This experiment explores the implementation effect of policy subsidies according to the investment changes and analyzes whether the subsidy effects are consistent under different subsidy methods.

4.3.1. Experimental Process

In this experiment, there are 96 enterprises and 96 consumers, similar to experiment 2, except they are divided into three groups, respectively, each consisting of 32 enterprises and 32 public. In the first group, they continue to invest according to the rules of experiment 2. In the second group, enterprises can receive 10% of the investment cost subsidy from the government when they successfully sell altruistic green products (the enterprises can obtain an environmental income of 0.6 (x + a) and subsidy income of 0.1 x through investing in altruistic green products). The public will not be stimulated by the subsidy. In the third group, the public can receive 10% of the investment cost subsidy from the government when they successfully purchase altruistic green products (the public can obtain an environmental income of 0.6 (x + a) and subsidy income of 0.1 a through investing in altruistic green products). The enterprise will not be stimulated by the subsidy. The initial funds and investment methods of the three groups were the same as those of experiment 2. After each round, the income and environmental benefits are still announced, at the same time, the three groups repeat five rounds without changing the grouping. The specific account balance Z calculation formula of this experiment is shown in

Table 2. Calculation formula of the account balance in experiment 3.

Transaction Successfully		Subsidy to the Enterprise	Subsidy to the Public
Both	Enterprise	${\mathrm{Z}}_{\mathrm{e}}=100-\mathrm{x}-\mathrm{y}+0.6\left(\mathrm{x}+\mathrm{a}\right)+0.6(\mathrm{y}+\mathrm{b})+0.1\mathrm{x}$	${\mathrm{Z}}_{\mathrm{e}}=100-\mathrm{x}-\mathrm{y}+0.6\left(\mathrm{x}+\mathrm{a}\right)+0.6(\mathrm{y}+\mathrm{b})$
	Public	${\mathrm{Z}}_{\mathrm{p}}=100-\mathrm{a}-\mathrm{b}+0.6\left(\mathrm{x}+\mathrm{a}\right)+0.7(\mathrm{y}+\mathrm{b})$	${\mathrm{Z}}_{\mathrm{p}}=100-\mathrm{a}-\mathrm{b}+0.6\left(\mathrm{x}+\mathrm{a}\right)+0.7(\mathrm{y}+\mathrm{b})+0.1\mathrm{a}$
Altruistic products	Enterprise	${\mathrm{Z}}_{\mathrm{e}}=100-\mathrm{x}-\mathrm{y}+0.6\left(\mathrm{x}+\mathrm{a}\right)+0.1\mathrm{x}$	${\mathrm{Z}}_{\mathrm{e}}=100-\mathrm{x}-\mathrm{y}+0.6(\mathrm{x}+\mathrm{a})$
	Public	${\mathrm{Z}}_{\mathrm{p}}=100-\mathrm{a}+0.6(\mathrm{x}+\mathrm{a})$	${\mathrm{Z}}_{\mathrm{p}}=100-\mathrm{a}+0.6(\mathrm{x}+\mathrm{a})+0.1\mathrm{a}$
Egoistic products	Enterprise	${\mathrm{Z}}_{\mathrm{e}}=100-\mathrm{x}-\mathrm{y}+0.6(\mathrm{y}+\mathrm{b})$	${\mathrm{Z}}_{\mathrm{e}}=100-\mathrm{x}-\mathrm{y}+0.6(\mathrm{y}+\mathrm{b})$
	Public	${\mathrm{Z}}_{\mathrm{p}}=100-\mathrm{b}+0.7(\mathrm{y}+\mathrm{b})$	${\mathrm{Z}}_{\mathrm{p}}=100-\mathrm{b}+0.7(\mathrm{y}+\mathrm{b})$

.

4.3.2. Experimental Result

(1)

The government subsidizes enterprises to produce altruistic green products

The overall distribution of results is shown in Figure 5. The public continues the state of the previous experiment and gives higher investment to egoistic green products to seek more benefits in the first and second rounds. However, since the third round, PAI exceed the PEI, and the gap continues to widen. In contrast, the enterprises have increased EAI at the beginning. Actually, the government’s subsidies to enterprises are not fully publicized, and the public’s understanding of the relevant policies is not high, therefore the public’s reaction is not timely when the policies are just promulgated. However, with the interaction with enterprises, the public gradually understands and fully considers the needs of both parties to make a win-win decision.

(2)

The government subsidizes the public to buy altruistic green products

It can be seen from Figure 6 that the response speed of the public is faster when the subsidy is given to the public than when the subsidy is given to the enterprise. When the government subsidizes the public, the investment/purchase willingness of enterprises and the public to altruistic green products increases. Meanwhile, the investment/purchase willingness of egoistic green products decreases, which achieves the expected purpose of the government subsidy.

5. Model Construction and Data Processing

5.1. Variables and Models

The investment situation of the public and enterprises is used to replace the demand level and the supply level. They can obtain both environmental and its own overall benefits. According to the practices of Liu et al. [62], who described environmental protection awareness with the public’s WTP for green products, this paper uses the public payment for EGB in experiment 1 as the degree of environmental protection awareness. In addition, this paper collected the individual information of the subjects with questionnaires (gender, education level, whether they are only children, performance ranking, whether they are student cadres, whether they have applied for education loans, and whether they have experimental experience). The specific variable settings are shown in

Table 3. Variable definition.

	Variable	Symbol
Enterprise Green Behavior (EGB)	Environment-oriented motivation	EOM
	Interest-oriented motivation	EIO
	Demand-oriented motivation	PGB
Public Green Behavior (PGB)	Environment-oriented motivation	EOM
	Interest-oriented motivation	PIO
	Environmental Consciousness	EC
Control variables	Gender	GEN
	Education	EDU
	Single	SIN
	Grade	GRA
	Leader	LEA
	Loan	LOA
	Experience	EXP

.

In order to avoid the problem of false regression in the model and verify the causal relationship between variables, this paper carries out the unit root test on variables. As shown in

Table 4. Results of unit root test.

	IPS	HT	ADF	Stationarity
EAI	−6.9888 ***	−6.2554 ***	−7.5041 ***	Yes
PAI	−5.6833 ***	−5.8658 ***	−3.0385 ***	Yes
EEI	−6.4573 ***	−8.1006 ***	−7.2193 ***	Yes
PEI	−5.8882 ***	−8.5413 ***	−5.6156 ***	Yes
EOM	−6.1299 ***	−7.4425 ***	−5.7837 ***	Yes
EIO	−6.7920 ***	−7.4498 ***	−6.1961 ***	Yes
PIO	−7.1855 ***	−9.1198 ***	−7.2178 ***	Yes

Note: *** means that the data are significant at the level of 0.01.

, all variables reach a stable state at the 95% significance level, meeting the hypothetical conditions of the causal test. In addition, the bootstrap program is added in the process of causal test command to reduce the interface dependence of panel data and ensure its robustness. The result of the causality test is: At a 95% significance level, enterprise investment is not the Granger cause of consumer purchase (EAI-PAI: Z = 3.3616, p = 0.5400 > 0.05; EEI-PEI: Z = 3.3327, p = 0.5100 > 0.05), and consumer purchase is the Granger cause of enterprise investment (PAI-EAI: Z = 13.5930, p = 0.0167 < 0.05; PEI-EEI: Z = 25.2793, p = 0.0033 < 0.05). Therefore, public purchase behavior is listed as one of the influencing factors of enterprise investment behavior in the model.

Based on combining the results obtained in experiments 1 and 2, this paper sets up a basic regression model for the docking of PGB and EGB to explore their impact mechanisms. Model (1) is the EGB model, and model (2) is the PGB model.

$$\mathrm{E}\mathrm{G}\mathrm{B}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1\mathrm{P}\mathrm{G}\mathrm{B}+{\mathsf{\alpha }}_2\mathrm{C}\mathrm{S}\mathrm{I}+{\mathsf{\alpha }}_3\mathrm{E}\mathrm{B}+{\mathsf{\alpha }}_4\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}+\mathsf{\epsilon }$$

(1)

$$\mathrm{P}\mathrm{G}\mathrm{B}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1\mathrm{E}\mathrm{C}+{\mathsf{\beta }}_2\mathrm{P}\mathrm{S}\mathrm{I}+{\mathsf{\beta }}_3\mathrm{E}\mathrm{B}+{\mathsf{\beta }}_4\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}+\mathsf{\epsilon }$$

(2)

where: EGB represents the green behavior of enterprises, including the investment in altruistic green products (EAI) and the investment in egoistic green products (EEI); PGB represents the public green behavior, including the purchase of altruistic green products (PAI) and the purchase of egoistic green products (PEI); EC represents the public environmental awareness, EIO represents the corporate income, PIO represents the public income, EB represents the environmental income obtained by enterprises and the public, Control stands for the control variable such as gender and education level; ε is the residual term.

In addition, in order to further analyze whether the government’s subsidy effect on altruistic green products is significant, this paper sets up a double difference model.

$$\mathrm{E}\mathrm{G}\mathrm{B}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1\mathrm{P}\mathrm{G}\mathrm{B}+{\mathsf{\alpha }}_2\mathrm{C}\mathrm{S}\mathrm{I}+{\mathsf{\alpha }}_3\mathrm{E}\mathrm{B}+{\mathsf{\alpha }}_4\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}+{\mathsf{\alpha }}_5\mathrm{D}\mathrm{I}\mathrm{D}+\mathsf{\epsilon }$$

(3)

$$\mathrm{P}\mathrm{G}\mathrm{B}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1\mathrm{E}\mathrm{C}+{\mathsf{\beta }}_2\mathrm{P}\mathrm{S}\mathrm{I}+{\mathsf{\beta }}_3\mathrm{E}\mathrm{B}+{\mathsf{\beta }}_4\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}+{\mathsf{\beta }}_5\mathrm{D}\mathrm{I}\mathrm{D}+\mathsf{\epsilon }$$

(4)

where: DID = Period × Treat, Period is the virtual variable of time, Period = 0 represents the round before the government subsidy, and Period = 1 represents the period after the government subsidy; Treat is an individual dummy variable. When the subject belongs to the control group, Treat = 0; when the subject belongs to the experimental group, Treat = 1; ε is the residual term.

5.2. Analysis of Empirical Results

5.2.1. Market Transaction Experiment

First of all, this paper conducts a regression analysis on models (1) and (2). The regression results are shown in

Table 5. Regression results of market transaction experiment.

	(1)	(2)	(3)	(4)
	EAI	EEI	PAI	PEI
PAI	0.2343 ***
	(0.0384)
PEI		0.2522 ***
		(0.0355)
EIO	−0.4870 ***	−0.6930 ***
	(0.0539)	(0.0541)
PIO			−0.3637 ***	−0.3531 ***
			(0.0616)	(0.0675)
EC			−0.0087	0.0016
			(0.0113)	(0.0110)
EOM	0.4539 ***	0.4344 ***	0.2880 ***	0.5232 ***
	(0.0405)	(0.0389)	(0.0464)	(0.0509)
GEN	2.6233	−3.3034	−6.8849	−3.7339
	(7.7731)	(6.6022)	(5.5287)	(5.3992)
EDU	2.3336	−1.6242	18.9495 *	−34.8579 ***
	(2.9220)	(2.4985)	(10.6177)	(10.3560)
SIN	−1.6571	0.7490	−0.7797	−3.2590
	(4.8413)	(4.1151)	(4.9724)	(4.8569)
GRA	−0.0875	0.5125	−1.9867	−4.1954
	(2.0699)	(1.7515)	(4.7209)	(4.6110)
LEA	3.5612	−3.3357	−7.8473	1.3780
	(5.4795)	(4.6600)	(6.7055)	(6.5987)
LOA	2.5578	−2.6240	6.4776	1.1294
	(5.2789)	(4.4719)	(5.1014)	(4.9961)
EXP	−8.0467 *	5.2421	10.3809	−16.9893
	(4.8150)	(4.1433)	(12.9512)	(12.6272)
_cons	35.4812 *	81.0081 ***	24.7169	141.2695 ***
	(18.1031)	(15.4916)	(42.5681)	(41.8159)
N	960	960	960	960

Note: * and *** means that the data are significant at the level of 0.1 and 0.01.

. In columns (1) and (2), the public’s purchase of altruistic green products will significantly promote the enterprise’s investment in altruistic green products, like egoistic green products, which verifies H3c. At the same time, environmental benefits significantly promote the green investment behavior of enterprises, so that H3b is also verified. In addition, the self-interest of the enterprise inhibits EGB. H3a has not been verified. In the results, the influence of control variables on the green behavior of enterprises is not significant, which is in line with the fact that enterprises are more in pursuit of profit maximization and are less affected by their own preferences. Columns (3) and (4) are the regression results of model (2). The public green purchase behavior is significantly affected by their own interests and environmental interests. However, the coefficient of their own interests is significantly negative. The results verify H2b but H2a is not verified. Furthermore, the impact of environmental awareness on PGB is not significant, and even the estimated coefficient in the public altruistic green investment is negative. H2c has not been verified. The occurrence of this situation may be due to the large difference in environmental awareness. This paper will conduct further heterogeneity analysis on the level of public environmental awareness later. As a result, the inhibition of self-interest in green behavior may be due to the “free riding” mentality. Enterprises and the public will reduce their investment or purchase of green products, which is also due to self-interest considerations to a certain extent.

5.2.2. Analysis of the Effect of Government Subsidy

In this experiment, the treatment of the two experimental groups is that the government gives subsidies to enterprises when they invest in altruistic green products and the public when they buy altruistic green products. The t-test results are shown in

Table 6. T-test of investment difference between enterprises and the public under different subsidy scenarios.

Groups Comparation	Investment	t-Value	p-Value	Significant Differences
Subsidy to Enterprise vs. No subsidy	EEI	−5.9034	0.0000	Yes
	EAI	5.2593	0.0000	Yes
	PEI	−4.9706	0.0000	Yes
	PAI	4.6885	0.0000	Yes
Subsidy to public vs. No subsidy	EEI	−7.5264	0.0000	Yes
	EAI	5.7632	0.0000	Yes
	PEI	−7.6450	0.0000	Yes
	PAI	8.4725	0.0000	Yes
Subsidy to Enterprise vs. Subsidy to public	EEI	1.2655	0.2076	No
	EAI	−0.4231	0.6728	No
	PEI	1.8231	0.0702	No
	PAI	−2.2994	0.0228	Yes

. At the 95% significance level, there is a significant difference between the government subsidy experimental group and the control group. The investment of enterprises and the public in altruistic green products decreased and began to shift to altruistic green products. This indicates that the policy subsidies effectively stimulate the choice of enterprises and the public, and both of them reflect a stronger WTP for altruistic green products in this round. In addition, by testing the two experimental groups, only the public investment in altruistic green products was found to have significant differences under different subsidy scenarios. Through the comparison of the coefficients, on the whole, when the government directly subsidizes the public, the market reaction is more obvious and has more of an incentive effect than those of subsidies to enterprises.

T-test verifies the implementation effect of government subsidies by testing the gap between groups. In order to explore the impact on the same subject before and after the policy implementation, this paper establishes a double difference model for further verification. Figure 7 shows the subsidy effect diagram of the parallel trend test, where the ordinate represents the investment effect coefficient after removing the mean value, and the abscissa 0 represents the subsidy effect of the sixth round. It can be seen that before the sixth round, the subjects were not stimulated by the government subsidy, and the coefficient was in a relatively stable and near zero state. In addition, there was no significant difference between the experimental group and the control group on the influence of green behavior. However, under the role of psychological games, every subject expected to receive a big return with minimal effort. Especially, the existence of a mentality of “free riding” led the overall investment situation of enterprises and the public to show a downward trend. After the implementation of policy subsidies, the coefficient suddenly changed from negative to positive, showing a large growth trend. The stimulation effect is significant and sustainable, that is, the double difference model meets the hypothesis premise of the parallel trend test. Due to space limitations, this paper only shows the trend of the investment effect of enterprises on altruistic green products when the government gives subsidies to enterprises (a) and the public (b), respectively.

The results of the double difference model are shown in

Table 7. The effect of government subsidy to enterprises.

	Subsidy to Enterprises				Subsidy to the Public
	(1)	(2)	(3)	(4)	(1)	(2)	(3)	(4)
	EAI	EEI	PAI	PEI	EAI	EEI	PAI	PEI
DID	11.5259 ***	−6.1227 *	12.4779 ***	−12.8577 ***	11.4902 ***	10.7771 ***	18.8861 ***	−17.9454 ***
	(3.5110)	(3.2349)	(3.5014)	(3.6358)	(3.6420)	(3.7662)	(3.6412)	(4.0613)
PAI	0.2240 **				0.1193
	(0.0898)				(0.0873)
PEI		0.2847 ***				0.1602
		(0.0717)				(0.0988)
EIO	−0.4939 ***	−0.7027 ***			−0.5366 ***	−0.5865 ***
	(0.0776)	(0.0656)			(0.0720)	(0.1050)
EOM	0.3741 ***	0.4742 ***	0.1693 ***	0.6744 ***	0.4321 ***	0.4947 ***	0.2376 ***	0.5741 ***
	(0.0673)	(0.0668)	(0.0617)	(0.0999)	(0.0717)	(0.0556)	(0.0597)	(0.1101)
PIO			−0.3344 ***	−0.4326 ***			−0.2970 ***	−0.4045 ***
			(0.0910)	(0.1438)			(0.0880)	(0.1323)
EC			−0.0104	0.0009			−0.0096	0.0053
			(0.0105)	(0.0101)			(0.0100)	(0.0098)
Control	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
_cons	29.3188 **	87.7819 ***	14.5099	149.4547 ***	40.8986 ***	69.8421 ***	42.5429	125.0664 ***
	(12.6692)	(10.0238)	(41.2139)	(36.8243)	(9.6687)	(10.4236)	(38.1719)	(37.6432)
N	640	640	640	640	640	640	640	640

Note: *, **, *** means that the data are significant at the level of 0.1, 0.05 and 0.01.

. The coefficients of DID are all significant, that is, when the government subsidizes the enterprises’ investment in green products, there is a significant difference between the experimental and control group. After the implementation of the policy subsidy, the investment distribution of the enterprises in green products changed greatly. The enterprises increased their investment in altruistic green products, so the effect of the policy subsidy is obvious, which supports H4a. The subsidies given by the government to the public when purchasing altruistic green products also have significant effects on public and enterprise behavior, which supports H4b.

5.3. Robustness Test

5.3.1. Policy Externality Test

In order to ensure that effective expectations have not been formed for the subjects before the implementation of policy subsidies, this paper draws on the practices of Sun and Song [63]. This paper adds DID₋₁ (dummy item before the implementation of the government subsidies multiplied by Treat) to the double difference model and obtains the regression results in

Table 8. Expected effect of government subsidy.

	Subsidy on Enterprises				Subsidy on the Public
	(1)	(2)	(3)	(4)	(1)	(2)	(3)	(4)
	EAI	EEI	PAI	PEI	EAI	EEI	PAI	PEI
DID	13.7046 ***	−8.7408 **	6.2332 *	−5.1198 *	14.6554 ***	−13.8237 ***	15.1148 ***	−14.4573 ***
	(3.6792)	(3.7115)	(3.4992)	(3.5915)	(4.5021)	(4.6451)	(5.3856)	(5.0660)
DID−1	−2.1215	3.0119	7.8408	−9.9255	−3.1425	3.2121	4.0260	−4.5417
	(2.1591)	(2.4532)	(6.1886)	(5.4598)	(3.6226)	(3.7229)	(5.2752)	(4.5430)
PAI	0.2213 **				0.1193
	(0.0937)				(0.0900)
PEI		0.3017 ***				0.1664
		(0.0740)				(0.1030)
EIO	−0.5228 ***	−0.6768 ***			−0.5532 ***	−0.5667 ***
	(0.0926)	(0.0833)			(0.0841)	(0.1119)
EOM	0.3811 ***	0.4613 ***	0.1516 **	0.6677 ***	0.4199 ***	0.5051 ***	0.2344 ***	0.5496 ***
	(0.0684)	(0.0674)	(0.0687)	(0.1105)	(0.0769)	(0.0576)	(0.0642)	(0.1204)
PIO			−0.3471 ***	−0.3749 **			−0.3316 ***	−0.3169 **
			(0.0961)	(0.1572)			(0.0949)	(0.1451)
EC			−0.0126	0.0030			−0.0096	0.0051
			(0.0107)	(0.0105)			(0.0108)	(0.0106)
Control	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	576	576	576	576	576	576	576	576

Note: *, **, *** means that the data are significant at the level of 0.1, 0.05 and 0.01.

. The regression results show that the coefficient of DID and other core explanatory variables are consistent with the coefficients in the basic regression. The DID₋₁ estimated coefficient is not significant, which indicates that there is no expected effect and the policy is not exogenous.

5.3.2. Placebo Test

In order to further test whether there are other potential factors other than government subsidies that cause changes in PGB and EGB, this paper refers to the practice of Song et al. [64] and uses the method of randomly selecting the false treatment group to conduct a placebo test. The specific operations are as follows: First, 64 subjects are randomly selected as the false treatment group, and the remaining subjects are set as the control group. Secondly, the dummy variable DID is constructed according to the number of rounds of government subsidies. Finally, the reconstructed DID is substituted into model (3) and model (4) for regression. In order to make the sampling results more reliable, 500 sampling treatments were conducted in this paper. The results are drawn in Figure 8, which shows the distribution of the estimated coefficients of the randomly set treatment group DID (due to the limitation of space, only the inspection results of the enterprise’s investment in altruistic green products when the government subsidizes the public are shown). It is not difficult to see that the estimated coefficient is distributed around 0, which is significantly different from the regression coefficient of the real treatment group composition model indicated by the vertical dotted line, thus supporting the effectiveness of the dual difference model.

5.4. Heterogeneity Analysis

The level of public awareness of environmental protection will affect the effect of government subsidies [65], and will also affect the technological innovation of enterprises [66]. In order to explore the impact of environmental protection awareness on the green behavior of enterprises, the subjects are divided into two groups. According to the median of public WTP in experiment 1 is 460 experimental coins, the public with an investment of more than 460 experimental coins are classified as high environmental protection awareness public. The group regression results are shown in

Table 9. Influence of environmental protection consciousness on enterprise investment behavior.

	EAI		EEI
	Low Environmental Awareness	High Environmental Awareness	Low Environmental Awareness	High Environmental Awareness
PAI	0.2606 ***	0.2557 **
	(0.0719)	(0.1207)
PEI			0.3561 ***	0.2272 *
			(0.0700)	(0.1281)
EIO	−0.2981 ***	−0.7057 ***	−0.9953 ***	−0.4794 ***
	(0.0776)	(0.1334)	(0.0899)	(0.1294)
EB	0.3776 ***	0.4344 ***	0.4992 ***	0.4444 ***
	(0.1020)	(0.0872)	(0.0901)	(0.0859)
Control	Yes	Yes	Yes	Yes
N	480	480	480	480

Note: *, **, *** means that the data are significant at the level of 0.1, 0.05 and 0.01.

and

Table 10. Influence of environmental awareness on public purchasing behavior.

	PAI		PEI
	Low Environmental Awareness	High Environmental Awareness	Low Environmental Awareness	High Environmental Awareness
PIO	−0.3161 ***	−0.3613 **	−0.2600	−0.4837 **
	(0.0967)	(0.1621)	(0.1825)	(0.1933)
EB	0.4308 ***	0.1775 **	0.2644	0.7374 ***
	(0.1335)	(0.0764)	(0.1834)	(0.1522)
EC	−0.0215	0.2625 **	−0.0061	−0.0081
	(0.0417)	(0.1227)	(0.0426)	(0.1296)
Control	Yes	Yes	Yes	Yes
N	480	480	480	480

Note: ** and *** means that the data are significant at the level of 0.05 and 0.01.

. Under different environmental protection awareness, the sign direction and significance of the regression coefficient of the variables of enterprise investment behavior have hardly changed. However, compared with low environmental protection awareness, the coefficient of public purchase behavior on enterprise investment behavior under high awareness is significantly reduced. This may be due to the fact that the public with high environmental protection awareness pays more attention to the environmental protection benefits when consuming products, resulting in a mismatch between the interests of enterprises and the public. Furthermore, as shown in Table 10, in the influence mechanism of public purchasing behavior, the coefficients of various variables in the model have great differences. First of all, the self-interest of the public with high environmental protection awareness significantly inhibits the purchase of egoistic green products. This means that the public is willing to buy altruistic green products rather than pursuing the benefits brought by egoistic green products, and the impact of environmental benefits on their behavior is more significant. In life, such people are more willing to respond to the government’s ecological policies, abandon part of their self-interest, and lead green consumption.

6. Discussion

Based on the current investment/purchase situation of scenario game experiments and the double difference results of government subsidy incentives, this paper discusses the conclusions drawn from the following three aspects.

(1)

Docking mechanism

The results of this paper indicate that the investment behavior of enterprises is significantly influenced by public purchasing behavior, but the impact of enterprises on the public is not significant. The public’s role in promoting green behavior in enterprises has been confirmed [49,50]. The study of Li et al. [67] reached similar conclusions, particularly that public attention to environmental pollution and green behavior has a significant impact on enterprise environmental investment. As for the opposite path, Yang and Chai [68] believe that substantial green behavior by enterprises can, to some extent, enhance the public’s willingness to pay. However, at the micro individual level, compared to enterprises, the public’s choices are more random and subjective. There is room for comparative selection. However, the level of demand will directly affect the performance of enterprises [49], resulting in differences in the implementation of EGB.

In addition, the paper divides green products into egoistic and altruistic green products, and there are differences in the investment situation in the docking mechanism. According to the research of Lavuri et al. [69], egoism values and altruism values drive the public in China and India to conduct green behavior, which leads to the difference between the two products. From a public perspective, different behaviors of enterprises have a substantial impact on the environment. From the perspective of enterprises, EGB plays an important role in guiding and encouraging consumers’ green behavior.

(2)

Motivation of green behavior

The research results indicate that in market trading scenarios, both enterprises and the public will implement green behavior driven by their own interests and environmental interests. However, companies are negatively affected by their own interests, which is inconsistent with many previous studies. Li et al. [23] found from the perspective of intrinsic motivation that overall benefits can promote the willingness of enterprises to implement green and low-carbon models, and exhibit the strongest effect among multiple influencing factors. In behavioral motivation testing, due to the existence of a “free riding” mentality, dynamic constraints between both parties prevent them from making the most reasonable investment choices for the maximization of their own interests. The process results in negative effects of their own interests on behavior. Unlike individual factors of the public, the implementation of EGB is closely related to its performance. EGB will increase their own costs in the short term, becoming a burden on the company [70]. However, as time goes by, enterprises will achieve green technology upgrades and their environmental investment will, to some extent, promote their economic performance [71], presenting a “U-shaped” relationship.

The implementation of PGB in this paper is influenced by public environmental awareness, although this impact was not apparent before the heterogeneity analysis. This is consistent with the research results of Kadic Maglajlic et al. [72]. Determined by self-identity and consumer value, young people’s environmental protection and social consumer participation are expected to enhance their environmental protection. This phenomenon can also be explained from the perspective of another study that suggests public recognition can reduce price sensitivity to green products [73]. Therefore, the public with high environmental awareness is more willing to give up some of their own interests, leading to green consumption.

(3)

Government subsidy

The conclusion of this article is that the government has a significant positive incentive effect on both enterprises and the public when providing subsidies. This is consistent with the results of most studies. Lian et al. [74] confirmed the incentive effect of government subsidies on the green behavior of enterprises through Panel data of Chinese-listed enterprises. Zakari et al. [75] proposed studying the effects of government subsidies in China and Japan by investigating the role of green finance in environmental performance, and suggested promoting the green energy system through subsidies. In addition, for European countries, the proposed subsidy can even benefit from the perspective of consumers [76]. In fact, policy subsidies and price promotions have the same effect. When the premium of green products is partially compensated, the public’s WTP will increase, which is consistent with the results of field experiments [77]. In the current state, the docking mechanism between enterprises and the public is actually a process of balancing the interests of both parties, especially reflected in the enterprise’s pricing and the public’s fair expectations of prices [41]. Due to the significant impact of supply on demand levels, the investment behavior of enterprises will also be subject to significant fluctuations when the government provides subsidies to the public in the future.

7. Conclusions and Implications

7.1. Conclusions

This paper studies the interaction between PGB and EGB by simulating transactions in the Chinese market, and explores the docking mechanism between enterprises and the public. This paper places them in a practical consumption environment to stimulate their environmental awareness. Then, this study examined the effectiveness of government subsidies. The conclusion is as follows:

Firstly, EGB can compensate for a portion of low-level investment brought about by green product premiums. The public’s WTP for substantive green behavior is higher than that for the symbolic green behavior of enterprises, which wins more reputation for enterprises in this process [43]. In the two scenarios of pursuing egoistic and altruistic green products, the public’s WTP for green technology innovation and cleaner production of enterprises is higher, while the WTP for environmental information disclosure is the lowest. In addition, influenced by self-interest and altruistic values [78], the public will have a higher WTP in altruistic green products compared to altruistic green products [79].

Secondly, there are differences in the two-way mechanisms between enterprises and the public. In the investment/purchase scenario game experiment, public green purchasing significantly promoted the investment of enterprises [10,49], and the reverse path was not verified. Indicating that market demand can significantly drive the green behavior of enterprises. Both the public and businesses are significantly influenced by self-interest and environmental interests [80].

Thirdly, government subsidies play an important role in stimulating green behavior among businesses and the public. Research has found that the government plays a balancing role in the interaction between enterprises and the public, using subsidy mechanisms to compensate for differences in the WTP [74].

Fourthly, environmental awareness significantly affects public behavior. The public with high environmental awareness is more likely to implement green behaviors [72]. They are more willing to buy altruistic green products than to pursue the interests of egoism. In addition, environmental interests have a greater impact on their behavior.

7.2. Implications

Based on the research on the public’s WTP for EGB, this paper puts forward practical management suggestions for enterprises. As the main body of the green economy, enterprises should take on corresponding environmental responsibilities to help China achieve its “double carbon” goal. Enterprises shoulder the responsibility of promoting the green industry and should implement the green concept in all stages of products, including technology, manufacturing, and logistics. With the improvement in public environmental awareness, enterprises should appropriately increase their investment in environmental protection to improve their image and reputation, gain public recognition, and reduce additional losses caused by public reports. In this process, enterprises should implement green behaviors in substantive or symbolic ways or focus on publicity activities for high green recognition behaviors. This is a direct way for consumers to understand the concept and operation mode of their green products, which will help enhance consumers’ trust in green products and thus improve the public’s willingness to pay.

In addition, based on the important role of the government in the docking mechanism between enterprises and the public, this paper puts forward the following policy suggestions: First, it makes rational use of the behavior variables of enterprise decision makers and the public. Investment and purchase decision making between enterprises and the public is a dynamic decision-making process and a process of two-way selection. The production of green products by enterprises can guide the public to purchase green products. The behavior of enterprises is greatly affected by public demand. Therefore, while promoting structural reform on the supply side, control on the consumption side cannot be relaxed. Second, establish and improve a reasonable financial subsidy and incentive system for green products. At the initial stage of the implementation of the environmental control policy, it is difficult for enterprises to balance the cost distribution of green behavior and technology research and development. The implementation of green behavior becomes the burden of technological management innovation in the production and operation of enterprises. If the government subsidy can bear part of the cost of green transformation for enterprises, it will improve their enthusiasm for technological innovation in pollutant discharge. Third, strengthen public awareness of environmental protection. Consciousness leads to different behavior decisions to a certain extent, and policy-making needs to pay attention to the differences in microsubject consciousness. When the public and enterprises are not sufficiently aware of environmental protection, both parties may give priority to their own interests and neglect the value pf environmental protection, resulting in a failure to implement the green consumption policy smoothly. In the publicity process, using multiple channels and methods to popularize green behavior among different people will make its implementation more sustainable.

8. Limitation

This paper has innovative research perspectives and methods, but there are also two limitations. On the one hand, most of the experimental subjects in this paper come from China, which is a reflection of consumer awareness and behavior in the context of traditional Chinese culture. The interaction process between entities is influenced by the environment, and different regional cultures can lead to differences in consumer demand. The situation leads to changes in the business strategy of enterprises in the market, which may result in different experimental conclusions. On the other hand, this paper selects behavioral motivation variables based on current relevant research, which is limited by experimental duration and mechanism control. The path of research needs to be addressed and upgraded. Currently, research on the consumption motivation of microentities is ongoing, and there is still great room for development in the future. Despite the development of the times and the improvement of consumer attitudes, the importance of behavioral motivation will also be emphasized. Exploring substantive and universal motivations at each stage is of great significance for the sustainability of human activities.

Therefore, future research can be expanded in the following aspects: (1) Future research can select experimental subjects in more regions and countries and expand the application of research conclusions based on the cultural background of the region. This will also facilitate an in-depth comparison. (2) Future research can utilize the characteristics of experimental data to incorporate more behavioral variables into the docking mechanism and enrich research results. (3) The research results can be verified by a variety of methods, such as using big data to investigate consumers’ willingness to consume and actual behavior, so as to realize the connection between production and consumption.