The Relationship between Subjective Status and Corporate Environmental Governance: Evidence from Private Firms in China

Abstract

Private enterprise governance is influenced by how status is perceived by business owners. In this essay, the role of business owners’ subjective statuses on their decisions on pollution control is discussed, along with social-psychology-based justifications. This study distinguishes front-end pollution treatment from back-end pollution treatment and finds the following: (1) A higher subjective status increases the front-end pollution control input but has no discernible impact on the back-end pollution control input. (2) Both the front-end and back-end control inputs are significantly influenced positively by a subjective economic status (ability motive). (3) A subjective social status only increases the front-end pollution control input (altruistic motive). (4) A subjective political standing (prestige motivation) significantly lowers environmental violations of businesses more than it does front-end or back-end inputs. (5) A subjective status can also boost owner expectations and business confidence, as well as help businesses to make the best pollution control measures. The optimum of private firms’ pollution control strategies depends heavily on their subjective status when the government’s pollution regulations are lax. The importance of subjective status is also influenced by how well capitalized private businesses are. This paper offers a novel perspective on the value of social governance from the viewpoint of maximizing the pollution management of private firms. It argues that raising the subjective status of private entrepreneurs has value.

1. Introduction

The primary component of the environmental governance system is an enterprise’s environmental governance behavior, which can be separated into front-end and back-end treatments. While the latter refer to environmental governance costs paid based on the volume of pollutants discharged to the government or third-party pollution treatment organizations, the former refer to equipment procurement, product green innovation, and process transformation to reduce pollution emissions [1,2]. The two pollution control strategies can be similar in theory. The ideal pollutant measurement system and market trading mechanism are crucial here, though. In fact, according to the study, businesses can easily avoid their environmental protection obligations by paying environmental governance fees because of the insufficient pollutant charge system, lax pricing standards, outdated pollution monitoring equipment, and other flaws [3,4]. Simply said, the front-end treatment results in a direct decrease in the amount of pollution that firms discharge, whereas the back-end treatment does not do so but instead transfers responsibility for pollution control to a third party by paying an environmental governance fee. The expense of environmental treatment is not entirely paid by businesses when the cost of pollution treatment is insufficient; instead, it is passed on to society. From this vantage point, the enterprise’s social duty for environmental protection is not adequately reflected by the back-end treatment. As a result, switching from a back-end to a front-end treatment more accurately captures the optimal private environmental governance mode in a given context.

Small- and medium-sized private businesses are more numerous and dispersed in China than state-owned or large businesses, making them more difficult for environmental regulators to control. As a result, these private businesses are more driven to reduce the cost of pollution management by paying pollution control fees. In China, there were more than 3000 small- and medium-sized private businesses by 2020. The front-end pollution control investment of these enterprises often only followed industry standards and was far below social expectations [5]. A challenging issue for the government to solve in order to protect the environment has always been pollution control for small- and medium-sized private businesses. Despite the modest size of each individual pollution emission, there are a lot of private businesses; therefore, the total amount of pollution cannot be understated. Additionally, the end-treatment mechanism of paying environmental governance fees has an undesirable impact on environmental governance [3,6]. To promote their environmental governance from end-treatment fees to front-end environmental governance investment, private businesses should optimize their environmental governance model.

Small- and medium-sized private businesses in China depend on their owners for decision-making. Particularly, a sizable part of private businesses are family businesses, where the founder or an heir is in charge of making choices. Compared to major firms or state-owned businesses (including publicly traded organizations), this is considerably different. Large or state-owned businesses frequently have good governance, and neither the chairman nor the CEO have complete authority over the business. Investigating how private businesses might improve their pollution control methods based on their owners’ personalities may be crucial. Rising income (i.e., the constant improvement in objective status) does not coincide with the concurrent rise in subjective status perception for private business owners who are benefiting from China’s reform and opening up. The gap between subjective and objective status seems to be strangely off in this case. This is known as Gatsby’s paradox in some investigations [7]. Following the 18th Congress of the Chinese Communist Party, this phenomenon changed. In other words, there appeared to be some convergence between a subjective and objective status [8]. The shift in how the Chinese political party and government view the value of the private economy may be responsible for the change in how privately owned business owners subjectively perceive their standing. However, it is unclear what other economic or social effects this move will have.

The upper echelons theory points out that managers’ social psychology has an important impact on enterprises’ strategic decision-making. In particular, an enterprise’s behavioral decisions, with social responsibility, made by managers are closely related to their social psychology. Will Chinese private business owners lead private enterprises to engage in more socially responsible behavior due to systematic changes in their subjective status perception (especially their subjective status promotion)? If this is the case, an improvement in subjective status will lead private enterprises to invest more in front-end pollution control and fulfill their corporate social responsibilities. This study examines how variations in private business owners’ subjective status perceptions affect the optimization of firms’ pollution control models using data from the Chinese Private Enterprise Survey (CPES). Private businesses in China may demonstrate their social responsibility and improve the mode of pollution management if they invest more in front-end pollution control. In order to provide empirical evidence that improving the subjective status of private business owners can optimize the pollution treatment mode of these private enterprises, this paper primarily examines the heterogeneous effects of the subjective status of private business owners on front-end and end-pollution treatment inputs.

The multiple regression and instrumental variable regression analysis built in this paper, based on the CPES (2006–2014), reveals that improving subjective status significantly prompts private enterprises to invest in front-end environmental governance but does not prompt them to invest in back-end environmental governance. The primary drivers for prompting private businesses to actively carry out front-end environmental governance are their subjective economic standing and subjective social status, which represent their ability motivation and prosocial incentive, respectively. However, subjective political status, which reflects the incentive for status among business owners, can deter pollution infractions but has little bearing on front-end and back-end environmental governance. Additionally, as subjective status improves, business expectations and confidence also rise, which further supports the fundamental finding. When formal institutions are weak, the promotion of subjective status might be crucial, but it depends on how well businesses operate. The empirical study in this paper provides a preliminary investigation and response to the theoretical claim made above. The research conclusion is not only conducive to expanding the thinking of the environmental governance of private enterprises, but also provides empirical evidence for comprehensively understanding the social governance value of affirming the status of the private economy.

2. The Literature

2.1. Research on Enterprise Environmental Protection Behavior

There are three perspectives to study the influencing factors of enterprises’ environmental behavior. Firstly, there is the research based on institutional theory. Legal rules, behavioral norms, and social expectations influence organizational behavior [9]. In order to survive, enterprises must abide by the rules of the environment and will be forced to carry out environmental management activities under the pressure of the external environmental system. From this perspective, most empirical studies on China focus on the impact of government environmental legislation, the reform of environmental legal institutions, and administrative orders of environmental protection departments on enterprises’ environmental behavior [10,11,12]. Undeniably, even with institutional constraints, enterprise environmental violations or illegal behavior still occur frequently. As part of the introduction, in the context of China, the pollution control fee system is actually a failure and becomes a channel for enterprises to evade their environmental social responsibilities [3,4].

The second type of research is based on the perspective of resource constraints. Such studies generally agree that environmental investment squeezes out the available resources of enterprises and increases their costs [13,14]. Therefore, enterprises are not willing to invest resources in environmental protection when there are fewer available resources. This kind of empirical research is mainly concerned with the vicarious impact between environmental investment and other investment opportunities [15]. On the contrary, abundant resources (such as profitability) or access to government subsidies (especially for enterprises’ environmental behavior) can relieve the pressure on resources and promote enterprises’ environmental behavior [16,17]. Of course, there are other views that environmental protection behavior increases the resource acquisition capacity of enterprises via innovation and reduces uncertainty [18].

The third type of research is based on the stakeholder perspective, especially the interests of enterprises’ decision-makers and environmental behavior. Berrone and Gomez-Mejia [19] found that long-term incentives for management contribute to companies’ environmental behavior. This may suggest that the benefits of environmental protection take time and are at odds with the short-term gains for management. In fact, when considering the characteristics of decision-makers, it is not only the interests, but also other characteristics of decision-makers themselves that can influence the behavior of enterprises in terms of social responsibility, including environmental social responsibility [20].

2.2. Decision-Maker Psychology and Enterprises’ Behavior

The psychological traits of decision-makers may be an important influencing factor for CSR behavior. Studies explaining enterprises’ social responsibility behavior (CSR) from the perspective of managers’ psychological characteristics or values (primarily using samples of senior executives of listed companies) frequently rely on indirect tests of managers’ special experience or demographic traits. This is because these studies are based on the upper echelons theory. Benmelech and Frydman [21] found that business managers with military experience are more likely to make corporate decisions with social responsibility [12]. Similar findings may hold true in the Chinese context. A study by Xu and Li [20] found that executives who had experienced poverty were more likely to make decisions that were in line with social expectations, such as donating more [21]. However, this kind of research on the relationship between decision-makers’ psychological factors and CSR is mostly based on the special experiences of decision-makers. In terms of the logic of the argument, there is a lack of correlation, namely whether these particular experiences will cause some particular psychological occurrence of decision-makers.

However, there are limited empirical data to explicitly examine psychological variables [22,23]. Another body of literature focuses on the relationship between decision-makers’ perceived status (i.e., subjective status) and decision-making. This may be the study that more directly examines the psychological factors of decision-makers. Status, especially perceived status, has a substantial impact on people’s behavior and decision-making, according to research in the fields of psychology and sociology [24]. The status of a person is determined by their level of restriction, which has both subjective and objective characteristics. Objective status is a categorical indicator that is related to a person’s money, education, professional identity, and other factors. Subjective status is a person’s assessment of themselves in relation to other social participants in their surroundings, a subjective psychological representation of their own identity, and a relative indication [25].

The Gatsby paradox states that although objective status serves as the material foundation for subjective status, the former does not determine the latter, particularly in the case of Chinese private entrepreneurs, whose subjective status has lagged behind their objective status as determined by economic power [7]. The class to which a person belongs can be determined by their subjective status, which is more crucial for understanding how private businesses make decisions [25]. According to some studies, the “theory of original sin” (the denial of private property under class struggle thinking) has been limiting the release of the entrepreneurial spirit of private individuals because of “non-standard” initial wealth accumulation (such as the loss of state-owned assets during the process of turning state-owned enterprises into private enterprises). A status perception that does not match one’s income level significantly inhibits their risk taking, long-term investment, and even their fulfillment of social responsibility [25,26,27].

2.3. Subjective Status and Behavioral Motivation

Subjective economic, social, and political elements were employed by the CPES to generate a 1–10 self-rating scale using the widely used MacArthur scale [22]. Using the valid sample used in this work as an example, the correlation coefficients for economy–society, economy–politics, and society–politics were 0.80, 0.65, and 0.76, respectively. This demonstrates that subjective status may contain a range of social psychology. Studies in sociology and psychology offer a wealth of information that may be used to comprehend the social and psychological traits of private entrepreneurs that are represented in their subjective status.

According to Haushofer [23], the psychological identification of poverty will cause people to make decisions that are primarily concerned with the short-term, ignore the long-term, and lack preparation. Additionally, failure in conduct and ambition are both consequences of poverty psychology [24]. This psychological justification shows that, on the flip side, a person’s social psychology will be strengthened by riches, making them more capable of achieving great things and dealing with the long-term. As a result, ability motivation is the social psychology that subjective economic status represents. (2) An important theory to explain corporate social responsibility behavior is prosocial motivation [28,29]. According to Klein [30], wealthy people have a propensity for making negative comparisons, developing stronger empathy, and exhibiting prosocial drive. The concept of class here is similar to the subjective social status question in the CPES. Lu [31] used the MacArthur scale to measure subjective social status and confirmed that people with high subjective status have more prosocial motivation. They also excluded the utilitarian purpose of obtaining expected returns, that is, people with a high subjective status show more altruism. Furthermore, Xie [32] showed that those with a high subjective social status value interpersonal ties and have an altruistic mindset toward those who are socially isolated. (3) According to Wu and Jin [33], there is a considerable positive association between the appeal of one’s own status, with prestige motive playing a large role, and the development of one’s political identity. In other words, a subjective political standing could be a reflection of company owners’ need for prestige. In Max Weber’s works, reputation not only enhances the legitimacy of individual domination of others or resources, but also produces a sense of glory, and then restrains individual behavior [34,35].

When front-end and back-end environmental governance are compared, it becomes clear that back-end control targets present pollution whereas front-end control involves long-term, significant expenditure and requires the ability and motivation of decision-makers. Additionally, front-end investment can lessen or completely remove root cause interference, which requires the backing of decision-makers’ prosocial or altruistic motivation. As a result, it may be said that an increase in subjective social status encourages private companies to assume a leadership role in environmental governance. For private businesses, upholding their social duty for environmental protection is in accordance with what the public and the government expect. On the other hand, if they fail to uphold their societal obligation to safeguard the environment, it could harm their reputation. Therefore, this article cannot conclusively conclude that subjective political status will increase the front-end investment of private firms in environmental regulation based on the social psychology of prestige motivation.

There are four parts in the literature that could benefit from the contributions of this paper. First, there are studies on corporate environmental governance that have mostly focused on internal governance structures [5,11] or external institutional constraints [9,10]. These have missed the opportunity to examine the traits of business owners and to evaluate the unique environmental governance of private firms. Therefore, in this study, the importance of informal factors and subjective position in the environmental governance of private firms is emphasized. Second, studies based on high-level theory, which aim to explain corporate social responsibility (CSR) behavior, often rely on demographic factors or indirect tests that draw on the unique experiences of managers [12]. The direct measure of the perceived status of the entrepreneur’s psychology is examined in this research as a response to the literature in this area. Third, some domestic literature [8,9,29] examines how private enterprises make decisions on R&D, donations, and staff development. These studies provide an overview of trust or altruistic motive; however, they lack significant empirical support. This study expands the description of the mechanism and its impact on CSR decision-making based on the three categories of subjective status. Fourth, private enterprise social responsibility behavior will be seen as utilitarian [36,37,38], that is, in exchange for financial or political resources [8,39,40,41,42], but the importance of individual decision-makers’ personal values cannot be disregarded. On the basis of controlling political connection and performance level, this paper confirms that non-utilitarian motivation will lead private enterprises to be more active in front-end environmental governance, which supports the personal value theory.

3. Methods and Data

3.1. Econometric Model

The theoretical analysis gives the judgment that the enhanced subjective status of business owners optimizes the pollution management pattern of private firms. Around this proposition, our baseline regression is specified as

$$\mathit{ln}{y}_{ijct}={\beta }_0+{\beta }_{1}st{a}_{it}+Control{s}_{it}+{\delta }_j+{\delta }_c+{\delta }_t+{\delta }_{j\times c}+{\delta }_{j\times t}+{\delta }_{c\times t}+{\epsilon }_{ijct}$$

(1)

The subscripts $i$, $j$, $c$, and $t$ represent the enterprise, industry, region, and year. There are two explained variables ($lny$): (1) the front-end environmental governance inputs (i.e., the environmental protection investment incurred by enterprises in equipment purchase, process transformation, and green product research and development for emission reduction); (2) the back-end environmental governance inputs, that is, the environmental governance fee paid by enterprises for pollution discharge. If the promotion of the subjective status of the entrepreneur can produce a change in the environmental governance mode of the enterprise as described in the theoretical analysis, it should be observed via regression analysis that the subjective status ($sta$) variable has a significant positive impact on the front-end environmental governance inputs, while the subjective status has no significant impact on the back-end inputs, or a less significant impact than that on the front-end inputs.

Controlling the interference of confounding elements is important in order to obtain the causal influence of subjective status on the environmental governance method of firms. Intuitively, the external environment (business climate, institutional restrictions, etc.) and company characteristics (size, development stage, product type, etc.) are the main factors influencing the environmental governance decision-making of firms. Regarding private businesses, it is impossible to disregard the causal identification interference brought on by entrepreneur traits. Although complex, the external environment’s involvement is based on mixed cross-section data from sizable samples of private businesses. It can be fully controlled by setting the dummy variables of the region, industry, and year to which the enterprises belong and their interactions. In particular, the introduction of the interaction terms of three groups of dummy variables basically realizes the full control of the external environment missing variable bias that varies with region–industry, region–year, and industry–year. The number of control variables is about 1000 after introducing two-by-two interaction terms for the three types of fixed effects.

The issue of endogeneity of subjective status variables in Equation (1) is mostly brought on by the absence of firm-level or owner-level variables. For instance, the owner’s underlying personality qualities or certain company-specific characteristics may influence both the owner’s subjective status judgments and pollution management choices, altering the ability to identify causal impacts. Such interference cannot be reduced in mixed cross-sectional data by adding firm or individual fixed effects. Additionally, there is no particular generative mechanism that the change in subjective status follows, and endogeneity cannot be reduced by conventional breakpoint design techniques. Based on Ma [25], who used business owners’ former employment history as an instrumental variable of subjective status, this research searched for instrumental variables of subjective status. Zhou [18] discovered that the decision-making psychology of business owners is more original-sin-prone in private firms that were acquired through the restructuring of state-owned or communal enterprises. In order to create 0–1-type variables for subjective status in Equation (1), this paper queries the enterprise manager or person in control, and queries whether the enterprise was bought by state-owned or collective enterprises, respectively. Equation (1) makes use of the 0–1 variables as instrumental variables to reduce the influence of endogeneity problems.

3.2. Data

3.2.1. Data Sources

Data from the Chinese Private Enterprise Survey (CPES), 2006–2014, were used in the empirical study. One of the longest running comprehensive national sample surveys in China is the CPES. The large-scale social surveys in China began much later, and many of them were short-lived. The CPES was founded in 1992, shortly after the Communist Party of China’s 14th National Congress suggested creating a socialist market economy. It has been maintained to the present day and is continually evolving and getting better. Numerous parties have acknowledged it as the definitive source for private enterprise research in China. Researchers from the All-China Federation of Industry and Commerce, the State Administration for Market Regulation, the Chinese Academy of Social Sciences, and the China Society of Private Economic Studies made up the Research Group on Private Enterprises, which carried out the CPES. A meeting has been held 13 times between 1993 and 2018 and occurs every two years. However, at the time of writing, the researchers could only access data up until 2016.

In order to conduct multi-stage sampling, the poll was based on a specific percentage (approximately 0.05%; each time, the percentage is slightly changed) across the nation. Firstly, they determined the total number to be sampled and the number of sampled enterprises in the provinces, municipalities (directly under the central government), and autonomous regions. Secondly, in the provinces, municipalities, and autonomous regions, they selected separately planned cities (the economic development of China’s planned cities is relatively high and can reach or even surpass the level of provincial capitals; moreover, the official rank of the party secretary or mayor of the planned city is higher than that of the ordinary prefecture-level city) or provincial capitals, prefecture-level cities and county-level cities (1 for each level of cities), and high, medium, and low levels of economic development of the counties (1 for each level of counties). That was a total of 6 cities and counties. Third, the sample size of businesses in the two locations was determined in accordance with the distribution proportion of businesses between urban and rural areas. Fourth, based on the distribution of each industry in urban and rural areas, the number of businesses polled in each industry was determined. Fifth, they selected the investigated enterprises according to the principle of equal distance.

Private businesses of various sizes and sectors are included in the CPES samples for 31 provinces (including municipalities and autonomous regions). Only the survey samples from 2006, 2008, 2010, 2012, and 2014 included the information on businesses’ investments in pollution management among the CPES data that were made publicly available for each year. Therefore, the primary data sources for this study were the CPES data from 2006, 2008, 2010, 2012, and 2014, with corresponding sample sizes of 3837, 4098, 4614, 5073, and 6144. The CPES contains a wealth of variable information that can be used to measure each variable in Equation (1), including information on business owners (including information on subjective characteristics and subjective attitude) and enterprises (including financial information, input factor information, pollution control information, industry, and regional information).

3.2.2. Data Cleaning

Particularly, during the observation years used in this paper, the CPES reported the subjective status scoring information of private entrepreneur owners, including three subjective status evaluations of economic status, social status, and political status. Over the years, the order of question items in the questionnaire has changed, but the method and measurement have been consistent. This measurement is an internationally accepted MacArthur scale of subjective status using a question. The question is “Where do you think you are in the following three social ladders (economic, political and social) compared with other members of the society around you”, and it is measured “On a scale of 1 to 10, 1 is the highest and 10 is the lowest”. The database reported the measurement results of three types of subjective status, which had good variability, and it was helpful in identifying the impact of subjective status on the environmental protection behavior of enterprises.

In addition, according to the definition of the main variables in Equation (1), it was necessary to clean the initial data. The cleaning principle is that the key indicators of the construction variables should not be missing or have logical errors; otherwise, they should be deleted. To be specific: (1) the front-end and end-input of pollution control should not be missing or negative; (2) three types of subjective status score values should not be missing, be less than 1, or greater than 10; (3) the information used to construct the 0–1-type control variables should not be missing, such as gender, whether they are a party member, whether they are the NPC deputy, or whether they are a CPPCC member; (4) enterprise capital data should not be missing or less than 0; (5) there should be no shortage of or less than 2 employees; (6) enterprise sales shall not be missing or less than 0; and (7) the educational information of business owners should not be missing. After data cleaning, the number of effective samples was more than 11,000 (there were slight differences in the number of valid samples for different variables).

3.3. Variables

3.3.1. Explained Variables

There are two explained variables in Equation (1): The first variable is the front-end pollution control input (Front-end input), namely the investment in emission reduction equipment, production process transformation, and green product research and development carried out by enterprises to reduce pollution emissions. The CPES database reports the amount of such investment of the enterprise. The amount is added by 1, and the natural logarithm is taken as the measure of this variable.

The second variable is the end pollution treatment input (Back-end input), that is, the fee of pollution treatment paid by enterprises to the government or third-party pollution treatment agencies. The amount of the fee for the pollution control of the enterprise is also reported in the CPES database. The amount adds 1, and takes the natural logarithm as the measure of this variable.

3.3.2. Core Explanatory Variables

Based on the MacArthur scale, business owners rank their perceived subjective status in the CPES database. The highest score is one, while the lowest is ten. An evaluation of the perceived economic, social, and political position of business owners was requested. This study computes the subjective status variable measurement in two steps using the three ratings that each business owner provided to the database. First, we deducted 11 from the score. The perceived status increased as the value increased. Then, principal component analysis was utilized to provide a comprehensive subjective status score (Subjective status) based on the three different types of derived subjective status data. The weighted average of the three different types of subjective state was, of course, the comprehensive subjective status. Taking 2014 as an example, the weights of economic, social, and political subjective status were 0.36, 0.39, and 0.35, respectively.

Principal component analysis’s composite index permits weights that do not add up to 1. As a result,

Table 1. Descriptive statistics of t variables.

	N	Mean	Std. Error	Min.	Max.
Front-end input	11,753	2.179	3.791	0	13.998
Back-end input	11,537	2.059	3.267	0	10.914
Subjective status	11,395	5.851	1.952	1.08	11.1
Firm-level Control Variables:
Sales	11,753	6.707	2.301	0	11.775
Employment size	11,753	3.758	1.485	0.693	7.601
Union	11,753	0.489	0.499	0	1
Party	11,753	0.334	0.471	0	1
Enterprise age	11,427	8.935	5.248	0	36
Government shareholding	11,735	0.301	0.497	0	1
Person-level Control Variables:
Gender	11,718	0.148	0.355	0	1
Age	11,678	45.759	8.614	16	70
Political connection	11,753	0.411	0.492	0	1
Party member	11,753	0.361	0.480	0	1
Years of edu.	11,612	14.131	2.973	6	19

Notes: Based on CPES; calculated by the authors after performing sample cleaning.

’s subjective status field includes a maximum value of greater than 10. This work also attempts to investigate the interference of weight difference on the research conclusion in the robustness analysis by using the simple average of three different subjective status types as a measure. Additionally, objective status factors such as income, education, political identity, age, and gender have an impact on subjective status. This research attempts to employ complete subjective status variables to regress the aforementioned objective status elements in the robustness test section and proposes the residual as the measure of subjective status for robustness analysis.

3.3.3. Control Variables

The control variables in Equation (1) include the key characteristics of business owners and enterprises. Among them, the characteristics of enterprises include the following: (1) Enterprise sales income (Sales), which describes the financial strength of enterprises. We took the natural logarithm of the sales value as the measure. (2) Number of employees (Employment Size), reflecting the size of the enterprise (in fact, fixed assets reflect the size of the enterprise more, but these data are not reported in the CPES). We took the natural logarithm of the number of employees to measure it. (3) Union organization (Union) reflects whether private enterprise governance conforms to the modern enterprise system. Variables of 0–1 were constructed according to whether a trade union had been established. (4) Whether to establish a primary Party organization of the Communist Party of China (Party), reflecting whether enterprises are subject to more non-institutional constraints from the government. The 0–1 variable measure was constructed based on whether the enterprise had set up a primary organization of the CPC. (5) Enterprise age (Enterprise age), which describes the degree of enterprise growth. The measurement was taken by subtracting the observation-year from the established-years. (6) Government shareholding (Government shareholding), used to capture the government’s influence on corporate governance decisions. According to whether the registered capital of the enterprise had government funds, state-owned enterprises, or collective enterprise funds, 0–1 variables were constructed.

The characteristics of business owners included the following: (1) The gender of the business owner (Gender). Variables of 0–1 were constructed according to gender information, with 1 for female and 0 for male. (2) The age of the business owner (Age), reflecting the business experience and social cognition of the business owner. We subtracted the birth-year from the observation-year. (3) The relationship between business owners and the government (Political connection). Variables of 0–1 were constructed. If the business owner had the status of NPC (National Peoples’ Congress) deputy or CPPCC (Chinese People’s Political Consultative Conference) member, it was 1; otherwise, it was 0. (4) Whether the owner is a member of the Communist Party of China (Party member), which reflects whether the decision of the owner is influenced by political propaganda. (5) The education level of the business owner (Years of edu.), which reflects the cognition or learning ability of the business owner. It was converted to a specific number of years based on the educational information of the business owner (such as middle school, high school, college, etc.). Primary school and below, junior high school, senior high school, junior college, and undergraduate and graduate students were defined as 5 years, 8 years, 11 years, 15 years, and 18 years of education, respectively.

In addition, three groups of dummy variables, including industry, region, and year, were set according to the industry, regional postcode (the postcode corresponded to the postal area, and the prefecture-level city could be identified together with the information of the province), and observation year of the enterprise, so as to control the influence of the external business environment. The descriptive statistics of the above main variables are shown in Table 1.

4. Results

4.1. Baseline Regression

Based on the cleaned, valid sample data, regression analysis of Equation (1) was conducted. The results are presented in

Table 2. Baseline regression results.

Explained Variable	Front-End Input		Back-End Input
	(1)	(2)	(3)	(4)
Explanatory variable	0.043 **	0.044 **	0.020	0.023
Subjective status	(0.018)	(0.019)	(0.017)	(0.017)
Firm-level variables
Sales	0.102 ***	0.121 ***	0.062 ***	0.082 ***
	(0.018)	(0.019)	(0.017)	(0.018)
Employment size	0.474 ***	0.447 ***	0.412 ***	0.379 ***
	(0.034)	(0.035)	(0.032)	(0.033)
Union	0.238 ***	0.213 ***	0.374 ***	0.322 ***
	(0.077)	(0.080)	(0.072)	(0.074)
Party	0.471 ***	0.517 ***	0.252 ***	0.331 ***
	(0.089)	(0.091)	(0.080)	(0.081)
Enterprise age	−0.014 **	−0.013 *	−0.019 ***	−0.017 ***
	(0.006)	(0.006)	(0.006)	(0.006)
Government shareholding	−0.199	−0.177	−0.076	0.019
	(0.179)	(0.190)	(0.168)	(0.173)
Business owner variables
Female	−0.149 **	−0.169 **	0.004	−0.012
	(0.076)	(0.079)	(0.072)	(0.074)
Age	−0.003	−0.003	0.001	0.001
	(0.004)	(0.004)	(0.004)	(0.004)
Political connection	0.085	0.053	0.092	0.050
	(0.077)	(0.078)	(0.071)	(0.071)
Party member	0.034	0.025	0.082	0.053
	(0.070)	(0.072)	(0.064)	(0.065)
Years of edu.	−0.011	−0.022 *	−0.006	−0.014
	(0.011)	(0.012)	(0.010)	(0.011)
Municipality FE	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
Municipality*Industry FE	No	Yes	No	Yes
Municipality*Year FE	No	Yes	No	Yes
Industry*Year FE	No	Yes	No	Yes
No. of obs.	10,880	10,880	10,683	10,683
R-squared	0.223	0.360	0.316	0.415

Standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01.

, where columns 1 and 2 assess the impact of subjective status on the front-end inputs for pollution control and columns 3 and 4 examine the impact on the back-end inputs for pollution control. A 1-point improvement in subjective status perception was used to boost front-end environmental governance inputs by roughly 4%. It appears that the subjective status of the business owner has a considerable beneficial effect on the positive environmental behavior of private firms. However, the significance test could not be passed by the coefficient of the effect of subjective status on enterprises’ back-end environmental governance inputs. It is particularly noteworthy that, compared to columns 1 and 3, columns 2 and 4 report the regression results after adding the two-by-two interaction fixed effects of industry, year, and region, and it can be seen that the optimization effect of subjective status on the enterprise environmental governance mode still performs robustly, and the coefficients and significance of the effects on front-end inputs and back-end inputs do not change significantly. In other words, external environmental factors do not seriously interfere in identifying the causal effect of subjective status on pollution treatment patterns. However, the goodness of fit of the regression analysis was significantly improved by controlling for the effects of external environmental factors for industry–region, industry–year, and region–year variants, with columns 2 and 4 improving the goodness of fit by about 13% and 10%, respectively, relative to columns 1 and 3. This indicates that external environmental factors are indeed an important aspect that affects private firms’ pollution management decisions, and that controlling for them is necessary.

The characteristics of firms have a strong ability to explain how they behave in terms of environmental protection. The front-end and back-end investments that businesses make in environmental governance will increase significantly as a result of higher success, as measured via sales volume, and this will have a greater impact on front-end investment relative to back-end expenditure. Although there is essentially no difference between the front-end and back-end impacts, the scale index is also favorably connected with environmental governance. The end governance appears to be more constrained by the trade union index, which indicates the uniformity of governance. However, front-end environmental governance is less allowed by the indications of party organizations that show control and restriction. The negative effect of enterprise age on environmental governance investment may be due to the fact that the mature or declining enterprises have more sound pollution discharge technology and lower environmental governance investment. However, there is no significant relationship between government ownership and private enterprises’ environmental governance decision-making, which proves that the decision-making power of private enterprises is mainly concentrated in the intuitive judgment of business owners. The regression results of the characteristics of entrepreneurs show that female owners are less likely than male owners to invest in front-end environmental governance, and politically connected business owners seem to be more active in environmental governance.

4.2. Endogeneity

4.2.1. Instrumental Variable Approach

Referring to the ideas of Ma [25], the 0–1 instrumental variables (IV1 and IV2) were constructed with the information of “whether the enterprise is restructured and acquired” and “whether it is the management or the person in charge of the enterprise before the establishment of the enterprise”. The regression results are summarized in

Table 3. Instrumental variable regression results.

Explained Variable	Front-End Input			Back-End Input
	(1)	(2)	(3)	(4)	(5)	(6)
	IV1	IV2	IV1 + IV2	IV1	IV2	IV1 + IV2
Subjective status	0.034 **	0.029 *	0.031 **	0.019	0.015	0.018
	(0.015)	(0.017)	(0.014)	(0.023)	(0.023)	(0.023)
Controls	Yes	Yes	Yes	Yes	Yes	Yes
FE	Yes	Yes	Yes	Yes	Yes	Yes
Hansen J (P score)	-	-	2.764 * (0.096)	-	-	3.204 * (0.073)
N	10,880	10,880	10,880	10,683	10,683	10,683
R-squared	0.142	0.201	0.253	0.127	0.121	0.112
First-stage results	Subjective status
	(7)		(8)		(9)
Restructured enterprise	−0.111 **				−0.141 **
	(0.055)				(0.056)
Pre-establishment experience			0.169 ***		0.181 ***
			(0.038)		(0.038)
F-value	15.48 ***		25.50 ***		25.51 ***

Standard errors in parentheses. * $p<0.10$, ** p < 0.05, *** $p<0.01$. Note: the control variables in the first-stage regression are consistent with the benchmark regression and the second-stage regression.

. The positive impact of subjective status on front-end input was still significant, and the impact on the back-end input did not pass the significance test. Compared with the OLS estimation, the impact coefficient of subjective status under IV estimation is smaller, which proves that the missing variables at the individual or enterprise level may overestimate the role of OLS on subjective status. The results of the first-stage estimation showed that both IV1 and IV2 were significantly correlated with subjective status. The F statistic values were more than 10, and there was no problem of a weak correlation. Moreover, the subjective status of the owners of private enterprises acquired from restructuring declined significantly, which is consistent with the original sin theory [26]. The conclusion that the management experience before founding an enterprise is positively correlated with subjective status is also consistent with Ma [25]. However, the test for exclusivity of these two instrumental variables (i.e., the Hansen J statistic under heteroscedasticity) is weak, and the null hypothesis of exclusivity can only be accepted at a significance level of 5% or higher.

4.2.2. Sensitivity Check for Omitted Variable Bias

Because of the weak exclusivity of the instrumental variables, there is reason to worry that restructured enterprises and pre-establishment experience may be confounding factors that need to be controlled. So, the concern that the disturbance term of Equation (1) is related to subjective status and thus leads to endogeneity still exists. Based on the method of Oster [43], the above two IVs were added to the control variables to discuss whether the possible missing variables at the individual or enterprise level would affect the estimation of the subjective status variable coefficient in Equation (1). Following the proof of Oster [43], the unbiased estimate of the subjective status variable coefficient (${\beta }^{*}$) in Equation (1) (the ideal state where all missing variables are controlled) satisfies the following:

$${\beta }^{*}\approx \tilde{\beta }-\delta \left[\dot{\beta }-\tilde{\beta }\right]\frac{R_{max}-\tilde{R}}{\tilde{R}-\dot{R}}$$

(2)

($\dot{\beta },\dot{R}$) and ($\tilde{\beta },\tilde{R}$) are, respectively, the goodness of fit between the regression coefficient of subjective status and the model in the case of Equation (1) regression, further adding restructured enterprise and pre-establishment experience as control variables. $R_{max}$ is the goodness of fit of the regression in the ideal state where all missing variables are controlled. $\delta$ characterizes the relative importance of the variables that are still missing and the control variables that have been added. If $0<\delta <1$, the missing variables can better explain the explained variables. Oster [43] believed that from the perspective of empirical research, the most important factors are often selected as explanatory variables, so the index is unlikely to be greater than 1. Taking the research of this paper as an example, what affects the investment in enterprise pollution control is often the industry and its own scale and external institutional environment constraints, which are controlled in the control variables. Both $R_{max}$ and $\delta$ are not available. If the missing variable is completely unimportant, the unbiased estimate of the subjective status coefficient is the regression result after adding two control variables, as shown in column 1 of

Table 4. Missing variable sensitivity (based on beta parameter test).

Test Equations and Parameters	$\mathit{\delta }=0,{\mathit{\beta }}^{*}=\tilde{\mathit{\beta }}$	$\mathit{\delta }=1,{\mathit{\beta }}^{*}\ne \tilde{\mathit{\beta }}$
		${\mathit{R}}_{\mathit{m}\mathit{a}\mathit{x}}=$ 1.3 $\tilde{\mathit{R}}$	${\mathit{R}}_{\mathit{m}\mathit{a}\mathit{x}}=$ 2.2 $\tilde{\mathit{R}}$	${\mathit{R}}_{\mathit{m}\mathit{a}\mathit{x}}$
Front-end inputs Subjective status factor	0.032 **	0.033 **	0.033 **	0.050
	(0.015)	(0.014)	(0.015)	(0.159)
End-end inputs Subjective status factor	0.020	0.023	0.026	0.114
	(0.017)	(0.021)	(0.021)	(0.122)

Standard errors in parentheses. ** p < 0.05. Note: standard deviation in parentheses, obtained via Bootstrap 100 times.

, which is not significantly different from the instrumental variable regression results (columns 3 and 6 of Table 3). To see whether the missing variable is significant, referring to Bryan [44], this paper set the significance parameter to 1, assigned $R_{max}$ as 1.3 times and 2.2 times that of the value of $\tilde{R}$, and then observed whether the sign and significance of the regression coefficients of the subjective status variables remained consistent with the baseline regression. Oster gives two important critical values [43] and reports that, generally, $R_{max}$ will only fall between 1.3 $\tilde{R}$ and 2.2 $\tilde{R}$. As shown in columns 2 and 3 of Table 4, the coefficient on subjective status remains robust and differs little from the results in column 1. The significance of the effect of subjective status on front-end inputs disappears only when $R_{max}$ is assigned a value of 1, which is almost impossible. Therefore, combining the results of Table 3 and Table 4, it can be concluded that even if the exclusivity performance of the two instrumental variables is not perfect, it should be sufficient to identify the causal effect between subjective status and the private enterprise environmental governance mode. From the combination of Table 3 and Table 4, whether it is ‘restructured enterprise’ and ‘pre-establishment experience’ as IV or control variables, the correction effect on the coefficient estimation of the subjective status variable seems to be replaceable.

4.3. Robustness Check

In the baseline regression, there is a zero-value problem for both front-end and back-end environmental governance, and further evidence of robustness is needed to obtain a comprehensive measure of subjective status using principal component analysis. The results in

Table 5. Regression results of the robustness check.

Robustness Check	Tobit	Tobit + IV	Change Measure	Change Measure	Change Measure
Explained Variable	Front-End	Front-End	Front-End	Front-End	Front-End Proportion
Subjective status	0.084 **	0.095 ***			0.641 **
	(0.041)	(0.026)			(0.313)
Subjective status 1			0.051 ***
			(0.020)
Subjective status 2				0.032 **
				(0.014)
Robustness check	Tobit	Tobit + IV	Change measure	Change measure
Explained variable	Back-end	Back-end	Back-end	Back-end
Subjective status	0.040	0.082
	(0.041)	(0.166)
Subjective status 1			0.025
			(0.018)
Subjective status 2				0.020
				(0.017)

Standard errors in parentheses. ** $p<0.05$, *** $p<0.01$. Notes: Subjective status 1 is the average of the three types of subjective status. Subjective status 2 are the residuals extracted from the regression of the indicators obtained from the principal component on the dummy variables of age, education, personal income, political status and age of the business owner. The MLE estimation in Tobit + IV analysis encountered non-convergence problems, and the results obtained from the two-stage estimation method are presented in the table. The control variables are consistent with the baseline regression.

show that the positive effect of subjective status on front-end inputs is robust, but the effect on end inputs is still insignificant. To address the robustness of the subjective status measure, two alternative measures were considered: (1) The simple average of the three types of subjective status value was used as the composite subjective status measure. Since the comprehensive subjective status measure obtained via the principal component depends on the weight of the three types of subjective status distribution, it is feared that the weight may affect the empirical results. Consider using a simple average measure of composite status to verify that empirical conclusions may in fact have nothing to do with weight distribution. (2) The composite subjective status obtained from the principal component analysis was regressed on the variables of personal income, political affiliation, education, age, and gender of the business owners, and the residuals were extracted as the composite subjective status measure. Although empirical studies have confirmed that the correlation between subjective status and objective status is only in the middle and lower levels, many sociologists and psychologists still believe that objective factors such as personal wealth, knowledge, and position will affect subjective status perception. These factors may constitute confounding factors affecting the causal identification of Equation (1). This paper uses the income, education, political identity, age, and gender of business owners to extract residuals from the regression of subjective status in order to eliminate the interference of objective status factors.

As shown in Table 5, the effects (sign and significance) of subjective status on front-end inputs and end inputs under the two measures are consistent with the baseline regression. In addition, this paper also constructs the front-end share variable (that is, front-end input/(front-end input + end input) and multiply by 100) as the explanatory variable and repeats the instrumental variable regression under Equation (1), and the results show that the conclusion that subjective status contributes to the front-end of environmental governance in private firms is still robust at the firm level.

4.4. Explanations from a Psychosocial Perspective

4.4.1. Different Effects of Three Subjective Statuses

Subjective status is the status perception of business owners by comparing with the surrounding members of society, which reflects the social and psychological characteristics of business owners. Obviously, the three different status perceptions reflect the different social and psychological characteristics of entrepreneurs. The most intuitive method to understand the mechanism of the impact of comprehensive subjective status on the heterogeneity of front-end and back-end environmental governance inputs of private enterprises is to use three types of subjective status to conduct regression analysis on the environmental governance inputs of enterprises. Based on the regression results shown in

Table 6. The different effects of three subjective statuses.

Explained Variable	Front-End (1)	Front-End (2)	Front-End (3)	Front-End (4)
Economic status	0.054 ***			0.101 ***
	(0.014)			(0.029)
Social status		0.031 **		0.015 **
		(0.013)		(0.007)
Political status			0.015	−0.026
			(0.012)	(0.022)
Explained variable	Back-end (5)	Back-end (6)	Back-end (7)	Back-end (8)
Economic status	0.037 **			0.033 *
	(0.018)			(0.019)
Social status		0.026		0.025
		(0.017)		(0.030)
Political status			0.006	−0.025
			(0.015)	(0.021)

Standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01. Notes: Due to space limitations, only the regression coefficients for the three statuses are reported. The control variables are fully consistent with the benchmark regression. Since there are only two instrumental variables, columns 4 and 8 report the results of OLS regressions with Restructured enterprise and Pre-foundation experience as control variables. The rest of the regression results are for the instrumental variables.

, it is found that the positive impact of subjective status on the front-end investment of enterprises mainly comes from the role of subjective economic status and social status, and political status has no significant impact. Although subjective social status and political status have no significant impact on the end investment, the improvement in subjective economic status will still promote enterprises to pay environmental governance fees.

Obviously, the regression results in Table 6 indirectly provide evidence for the socio-psychological interpretation of the influence of subjective status on the heterogeneity of corporate environmental governance behavior: Subjective economic status reflects the confidence of entrepreneurs in their own wealth resources. The money management subjective economic status reflects the ability motivation of entrepreneurs. Under the same constraint conditions, the ability motivation will strengthen the decision-making of entrepreneurs. This is also reflected in the significant impact of subjective economic status of enterprises on the front-end and back-end investment. Subjective social status reflects the importance or class cognition of business owners to their own society. The so-called “the higher the status, the greater the responsibility,” subjective social status reflects the prosocial motivation or altruistic motivation of business owners [24]. The front-end environmental governance investment is more beneficial to the society, and the subjective social status will strengthen the front-end environmental governance investment of enterprises. Subjective political status reflects the judgment of business owners on the recognition and legitimacy of their own political system environment. The higher the status of the supervisor, the stronger the prestige motivation. Clearly, corporate environmental responsibility does not strengthen the reputation of business owners, but once violations occur, the blow to the reputation of business owners is huge. The results of Table 6 are basically consistent with the theoretical judgment based on social psychological explanation, but it is still necessary to supplement the test of social psychological explanation.

The regression results shown in Table 6 are well confirmed by the social psychological explanation of subjective status affecting environmental governance decision-making in theoretical analysis and provide a mechanism basis for understanding the heterogeneous influence shown in Table 2. However, whether the psychosocial explanation of the three subjective positions is rigorous or not needs to be supported by empirical evidence to enhance the robustness of the analysis conclusions.

4.4.2. Ability Motivation and Altruistic Motivation

The interpretation of altruistic motivation of subjective social status is easy to test. Medical insurance coverage rate and pension coverage rate [45,46,47] indicators within enterprises are selected to depict the welfare care given to employees by private enterprises. If subjective social status reflects the altruistic motivation of business owners, it will significantly improve the level of coverage of the two types. Corporate altruistic behavior towards society is mainly reflected in charitable donations, and it should be observed that subjective social status has a significant positive impact on it. The regression results in

Table 7. Ability Motivation and Altruistic Motivation.

Explained Variable	Medical Insurance Coverage		Pension Coverage		Donation of Enterprises
	(1)	(2)	(3)	(4)	(5)	(6)
Subjective status	0.013 **		0.012 *		0.049 ***
	(0.006)		(0.007)		(0.005)
Economic status		0.017 ***		0.012 **		0.016 **
		(0.006)		(0.006)		(0.007)
Social status		0.011 *		0.010 **		0.034 ***
		(0.006)		(0.005)		(0.008)
Political status		0.008		0.008		0.004
		(0.005)		(0.006)		(0.006)

Standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01. Notes: The control variables are identical to the baseline regressions and are omitted. Columns 1, 3, and 5 are regressions on instrumental variables, while columns 2, 4, and 6 include restructured business and pre-foundation experience as control variables in the OLS regressions.

verify the rationality of interpreting altruistic motivation of subjective social status.

There is no particularly suitable research design for the test of ability motivation interpretation of subjective economic status. This is because the perception of wealth resource status will be reflected in the business decisions of many business owners, which can also be reflected in Table 7. Subjective economic status also has a significant impact on the medical insurance coverage, pension coverage and donation of enterprises. Moreover, under the internal indicators of enterprises, the impact of subjective economic status is higher. Because making decisions to improve the coverage of medical insurance and pension for employees and increasing the cost burden of enterprises regularly need the support of the ability and motivation of business owners. On the contrary, the donation behavior of enterprises is often accidental. In addition to the support of the ability and motivation of business owners, the improvement in donation depends more on the altruistic psychology of business owners. Column 6 of Table 7 can verify this judgment.

4.4.3. Reputation Motivation

Under the aforementioned indicators, the subjective political status of business owners does not play a role, which is supposed to be related to the social psychology of business owners characterized by this status. This paper believes that subjective political status reflects the prestige motive of business owners. If this explanation holds, then the higher the subjective political status is, the more business owners are willing to do things that will put them in a good light and the less they are willing to do things that might discredit them. This paper does not have suitable data to evaluate the former, however, in the context of analyzing environmental behavior, the environmental violations of firms fit well with the latter. As shown in

Table 8. Reputation Motivation.

Explained Variable	Environmental Fines
	(1)	(2)	(3)	(4)	(5)
Subjective status	−0.013
	(0.009)
Economic status		−0.004			0.004
		(0.008)			(0.008)
Social status			−0.009		0.003
			(0.008)		(0.007)
Political status				−0.013 *	−0.013 **
				(0.007)	(0.005)

Standard errors in parentheses. * p < 0.10, ** p < 0.05. Notes: The control variables are identical to the baseline regressions and are omitted. Columns 1, 3, and 5 are regressions on instrumental variables, while columns 2, 4, and 5 include restructured businesses and pre-foundation experience as control variables in the OLS regressions.

, subjective political status significantly inhibits environmental violations by firms, while the regression coefficients for the remaining subjective status indicators are negative but insignificant.

4.4.4. Confidence and Expectations

The established literature studying the subjective status of private business owners generally suggests that the status reflects the optimistic and confident psychological state of business owners. However, studies generally lack empirical evidence. Such a psychological state may influence business expectations and business confidence, and promote activities with high investment and long payoff cycles, such as R&D innovation. This psychological mechanism explanation is also applicable for understanding the heterogeneous influence of subjective status on pollution management behavior. Only the 2016 private enterprise database reports information on the measure of business owners’ expectations of economic development in the next five years, as a measure of business owners’ business confidence, whereby the indicator is a scale of 1–5, where higher values indicate that business owners have better expectations for the economic situation in the next five years. The results of regressions on subjective status are shown in

Table 9. Confidence and expectations.

Explained Variable	Expectations for Economic Development in the Next 5 Years
	(1)	(2)	(3)	(4)	(5)
Subjective status	0.039 ***
	(0.010)
Economic status		0.038 ***			0.035 **
		(0.009)			(0.016)
Social status			0.029 ***		−0.021
			(0.009)		(0.018)
Political status				0.034 ***	0.028 **
				(0.008)	(0.012)

Standard errors in parentheses. ** p < 0.05, *** p < 0.01. Notes: data from the CPES (2016) are used here, and the control variables are consistent with the baseline regressions (without year variables), from those omitted for the ready claim.

. An increase in subjective status does strengthen business owners’ expectations of future operations, and the effect mainly originates from a subjective economic and political status.

4.5. Heterogeneity Analysis

The aforementioned study confirms that the subjective elevation of private entrepreneurs’ status can produce windfalls in terms of improved corporate pollution management patterns. However, this subjective influence should be constrained or moderated via objective conditions. There are at least two objective conditions that need to be clarified. The objective environmental governance system, and the realistic ability of private enterprises to conduct environmental behavior. Intuitively, corporate pollution management behavior will be subject to external institutional hard constraints, whereby the institutional constraints will be enhanced to replace the endogenous drive of business owners; conversely, when the formal system of environmental protection is insufficient, the psychological decision-making business owners offer a useful complement. The subjective status acting on the environmental protection behavior of private enterprises must be supported by the objective performance capabilities of the enterprise, that is, only for the enterprise with good performance is there material support for them to convert the subjective factors of the enterprise owner into front-end pollution control investment.

Table 10. Further discussion.

Explained Variable	Front-End Input
	(1) OLS	(2) IV	(3) OLS	(4) IV
Subjective status	0.223 ***	0.207 **	−0.190	0.196
	(0.049)	(0.087)	(0.147)	(0.723)
Subjective status*environmental courts	−0.214 ***	−0.183 **
	(0.059)	(0.071)
Subjective status*sales			0.035 ***	0.098 ***
			(0.007)	(0.011)
Environmental courts	1.243 ***	4.865 *
	(0.406)	(2.837)
Sales			0.064 **	0.129 **
			(0.030)	(0.055)

Standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01. Notes: firm performance uses the sales, and the controls are consistent with the baseline regression.

reports the results of assessing the role of the two moderating variables mentioned above. Referring to Fan [10], the impact of setting up environmental protection courts (collected manually by the authors) in each prefecture-level city is used as a measure of the rule of law in environmental protection. It can be seen that the impact of the environmental rule of law on enhancing firms’ front-end environmental governance inputs is highly significant. Moreover, the promotion effect of the subjective status of firm owners on front-end environmental governance is weakened in the case of the environmental rule of law, as the coefficient of the cross-product term is significantly less than 0. The substitution between subjective status and the impact of the environmental rule of law on firms’ front-end environmental governance is robust to both ordinary least squares regression and instrumental variable estimation methods. Measuring the level of firm performance in terms of sales, the regression results show that subjective status affects firms’ active front-end environmental governance and is very dependent on whether the firm is financially strong. For firms with very weak performance, even a high subjective status of the owner does not affect firms’ environmental governance inputs, as the coefficients of the primary term of subjective status in columns 3 and 4 are not significant. The subjective status effect only becomes apparent as the firm’s performance strengthens.

4.6. Discussion and Limitations

Existing studies have pointed out that under the government’s environmental regulations (including legal or administrative constraints), enterprises’ behavior of evading environmental responsibility still occurs frequently [3,4]. Further study is required, from the perspective of resource constraints, to explore the constraints of enterprise environmental behavior, such as profit level or government subsidies [16,17]. However, this still cannot constitute a comprehensive cognition of enterprise environmental protection behavior. Part of the literature focuses on the influence of decision-maker factors on enterprises’ environmental decision-making. For example, the decision-maker’s incentive problem [19], the decision-maker’s social psychology [22], and so on. The research in this paper was carried out from the perspective of decision-makers’ psychological factors. In particular, this paper focuses on the pollution control of private enterprises in China. This kind of enterprise is special, the number is large, and the distribution is discrete. The government has consistently challenged such enterprises to deal with pollution.

This paper analyzes the importance of informal factors in the pollution control of private enterprises from the perspective of subjective status. Studies on the explanation of CSR behaviors from the perspective of managers’ psychological traits or values based on the high-level ladder theory often resort to demographic characteristics, resulting in a “black box” of causal chain, or indirect tests with the help of managers’ special experiences [21]. This paper uses the direct measure of the psychological perception status of business owners to conduct tests, and makes an extension of the literature. Subjective status and decision-making in private firms in the Chinese context have been studied, such as R&D, donation, and employee development [25]. Based on the three categories of subjective status, this paper improves the mechanism explanation of its influence on CSR decision-making. The social responsibility behavior of private enterprises will be interpreted as utilitarian [36], that is, to exchange economic or political resources [8,40], but the role of personal value factors of decision-makers should not be ignored [22]. This paper finds that non-utilitarian motives will lead private enterprises to take more active front-end pollution control measures, which supports the importance of decision-makers’ personal values on CSR behaviors.

A significant share of China’s total number of enterprises is private businesses. However, encouraging private businesses to fulfill their social obligation is quite challenging. Private businesses are less financially powerful than state-owned or big businesses, but more significantly, it is harder to observe their activities. As a result, institutional restraints on private firms will be less effective. The study presented in this paper offers some ideas for potential future research projects: (1) It is crucial to comprehend private enterprise decision-making from the standpoint of firm owners. The traits of the proprietors may provide an explanation for several odd phenomena seen in private businesses. (2) The research in this work solely focuses on the investment in pollution control because it is impossible to observe the precise pollution emission information of private firms. Future research content may be further refined if enterprise pollutant emission data are collected. (3) In addition to environmental protection, various private sector social responsibility practices can be examined from the standpoint of subjective status.

5. Conclusions

5.1. Research Conclusions

The subjective position of private business owners has greatly increased as a result of the Chinese government’s stated affirmation and support for the private economy and the improvement in the business climate. Because of this occurrence, academics are now more interested in the subjective status of business owners and how this position affects corporate governance and societal value. This study extends the study to examine whether and why an enhanced subjective status of business owners can improve pollution governance in private firms. The Chinese Private Enterprise Survey (CPES) provides detailed measures of subjective economic, social, and political status, providing data to support an in-depth understanding of the association between subjective status and private firms’ environmental decisions. Contemporary theories suggest that subjective status encompasses multiple social psychologies and thus has multiple possibilities for acting on corporate environmental behavior. The empirical analysis based on the firm sample data shows that subjective status has a significant positive effect on private firms’ positive front-end environmental governance inputs; however, it does not promote firms’ negative back-end environmental governance inputs. The findings remain robust under instrumental variables, substitution measures, and substitution regression models.

A subjective economic status, which characterizes ability motivation, has a positive effect on both front-end and back-end environmental governance investment. Subjective social status, which characterizes prosocial or altruistic motivation, has a positive effect on front-end environmental governance. A subjective political status, which expresses prestige motivation, has no significant effect on either front-end or back-end environmental governance, but can discourage environmental violations that damage corporate or personal reputation. In addition, a subjective status also boosts business confidence and the expectations of business owners, and is primarily an effect of a subjective economic and political status, which may also have explanatory power for the antecedents of environmental governance.

5.2. Policy Recommendations

There are boundary constraints on the role of subjective status in promoting the front-loading of environmental governance by private firms. When the external binding regime is insufficient, the function of subjective status becomes more prominent. In other words, if formal institutions are legally enforceable, disparities in business owners’ subjective statuses do not result in variations in the inputs used for front-end environmental governance. The social psychology of business owners is brought to light via subjective status, which has an impact on the front-end environmental governance investment choices that are reliant on the performance strength of private enterprises and can only function when firms have higher performance strengths.

From the perspective of legitimacy and economic contribution, giving private business owners their due status, protection, and respect has been an important part of the government’s commitment to improving the business environment. According to the study presented in this paper, appreciating business owners’ contributions and elevating their standing also benefits the quality of environmental governance practiced by private enterprises. In reality, non-institutional elements, particularly the views of decision-makers and psychological factors that can play a significant complementary role, are frequently crucial when external environmental constraints are insufficient. The article comes to the conclusion that strengthening private entrepreneur relationships and recognizing their value elevates corporate social responsibility governance.