Do state-owned enterprises really have better environmental performance in China? Environmental regulation and corporate environmental strategies

doi:10.21203/rs.3.rs-1658200/v1

Although state-owned enterprises (SOEs) have long been criticized for being inefficient compared to their counterparts, they play a vital role in realizing social welfare including the improvement of environmental quality. With governments as their controlling shareholders, the motivations and strategies of SOEs to improve their corporate environmental performance can be very distinct from their privately-owned counterparts. In this article, we propose that a unique strategy that SOEs take in improving their environmental performance is high responsiveness to the government’s goal. Using a unique and detailed firm-level dataset, we find that SOEs perform better for the pollutant assessed by government targets while they perform similarly for the unregulated pollutant compared with their privately-owned counterparts, which indicates that they are more responsive and accountable to achieve policy requirements. Using privatization as an external shock, we find that SOEs worsen their environmental performance with mandatory targets after the privatization, which is consistent with the main results. With the privatization of SOEs worldwide, determining how to avoid the negative effects of privatization as well as to motivate for-profit enterprises to improve corporate environmental performance is the key to achieving a win-win.

corporate environmental performance

state-owned enterprises

responsiveness

SOEs have long been criticized for being inefficient compared to their counterparts due to misallocation of resources (Berkowitz et al. 2017; Brandt et al. 2020; Chen et al. 2021; Hsieh and Klenow 2009; Sukhtankar 2016) and poor management (Kang and Kim 2012; Lu et al. 2010), which makes the privatization of SOEs a common choice for countries to gain economic prosperity and market vitality, such as Russia (Sprenger 2011), Ukraine (Mykhayliv and Zauner 2013), Hungary, Romania (Brown et al. 2006), Egypt (Omran 2009) and Spain (Albalate et al. 2020). However, it has widely been recognized that SOEs play roles as multitask takers (Bai et al. 2006) rather than profit maximizers (Jensen 2001). For example, SOEs have long contributed to public services such as providing employment and maintaining social stability (Hsieh and Klenow 2009), which are also the main goals of most governments.

Perceived as the major contributor of pollution (Dahlmann et al. 2019), enterprises are under increasing pressure from regulators, environmental organizations, and the public to undertake more environmental responsibilities and improve their environmental behavior (Berry and Rondinelli 1998; Hart 1995; Walker and Wan 2012). These pressures lead to the popularity among enterprises to improve their corporate environmental performance to avoid market risks (Muhammad et al. 2015), save energy and cost (Chen and Chang 2013; Porter and van der Linde 1995; Sánchez-Medina et al. 2015), gain competitive advantage (Chen 2008; Chuang and Huang 2015; Miles and Covin 2000; Shrivastava 1995), legitimacy (Berrone et al. 2017; Hart 1995) and reputation (Dögl and Holtbrügge 2014; Miles and Covin 2000). Prior studies suggest that the level of environmental performance can be affected by not only internal factors, including corporate size (Bundy et al. 2013; Patten 2002), industry type (Bundy et al. 2013; Patten 2002), availability of resources and capital (Li and Chan 2016; Zhang 2017), CEO hubris (Arena et al. 2018; Zhang et al. 2020), and diversification (Dooley and Fryxell 1999; Kang 2013), but also external factors, including stakeholders’ attitudes (Abreu 2009; Kassinis and Vafeas 2006), pressure from NGOs (Lyon and Maxwell 2008), regulations (Lyon and Maxwell 2008; Yao and Yang 2017), and competition (Duanmu et al. 2018).

Also, different ownership may vary in their environmental performance because of their diverse interests and targets. Although scholars have made a great contribution in using the political connection (Maung et al. 2016), the resource-based view (RBV) (Zhang 2017), and multitask theory (Bai et al. 2006) to explain the difference between SOEs and privately-owned enterprises (POEs), their results are mixed. Some hold the opinion that SOEs outperform in environmental performance (Bai et al. 2006; Fryxell and Lo 2001; Liang and Langbein 2021; Zeng et al. 2012; Zhang 2017) while others argue that SOEs have worse environmental performance (Chun 2009; Jiang et al. 2014; Maung et al. 2016; Rowe and Guthrie 2010; Talukdar and Meisner 2001; Wang et al. 2003). Considering the particularity of state ownership, we believe that one possible reason for the mixed evidence of SOEs’ environmental performance is that the existing literature ignored the leading role of governments in driving the environmental performance of SOEs. With governments as the controlling shareholders, SOEs are subject to stricter supervision and given more priority, which makes the government’s impact on SOEs is much greater and makes SOEs perform better in environmental performance regulated by governments. To better understand the differentiated corporate environmental performance of different ownership and fill the gap between theories and empirical evidence, this article proposes a new interpretation of the SOEs’ corporate environmental performance from the perspective of the responsiveness to the government, that is, they only outperform in regulated aspects.

As a country where SOEs play a dominant role that accounts for approximately forty percent of the national GDP¹, China is gradually transforming SOEs into POEs to realize their promotion in productivity and innovation (Bai et al. 2009; Jefferson and Su 2006) since 1997. Therefore, we take advantage of the SOE reform in the late 20th century as an opportunity to use the difference-in-differences (DID) method to not only compare the difference between SOEs and POEs but also mitigate endogeneity problems in ownership and explore the changes in SOEs’ environmental performance after privatization. In addition, China has paid increasing attention to environmental pollution and has enacted various policies to regulate the emission of major pollutants such as sulfur dioxide (SO₂). This allows us to examine how SOEs’ environmental performance strategies are affected by the government’s targets and policy goals by comparing the controlled pollutant SO₂ and noncontrolled pollutant nitrogen oxides (NO_x).

This study explores the differences between SOEs and POEs in their environmental performance using two firm-level databases from 1998 to 2007. Furthermore, we use the DID method to explore how the environmental performance of SOEs changes after privatization. The results are as follows. First, by comparing the results of different pollutants, we find that SOEs perform better for the controlled pollutant SO₂, while this difference disappears for NO_x, which indicates that SOEs take environmental responsiveness strategies in assuming environmental responsibilities. SOEs produce less, remove more and thus emit less SO₂ than do POEs by 5.04%, 18.53%, and 7.13%, respectively. Second, privatized SOEs significantly increased their SO₂ production and emissions after privatization by 7.61% and 8.18%, respectively. These results further imply that SOEs in China perform better in regulated environmental aspects and that their corporate environmental performance declines after privatization.

This article contributes to the literature on corporate ownership and environmental performance strategies in five ways. First, different from the existing literature that ignores the effect of differentiated environmental performance strategies, we propose that SOEs take responsiveness strategies to meet the requirements of the government, which highlights the role of various environmental performance strategies. This is of importance to both theory development and empirical research. Second, most existing studies take business development and profitability as the main motivation for enterprises to improve corporate environmental performance, while we focus on special ownership of SOEs and find their differentiated motivation in improving environmental performance is to meet the requirements of the government. Third, most existing studies comparing differences in the environmental performance of different ownership use regional or cross-sectional data, which leads to endogeneity problems due to the existence of missing variables. To alleviate endogeneity problems when comparing the environmental performance of different ownership, we use the exogenous shock of the privatization of Chinese SOEs and the DID approach to compare the differences between reformed and non-reformed enterprises before and after the restructuring. We use enterprise-level panel data covering all industries across the country to mitigate endogeneity problems by controlling firm fixed effects and provide more robust empirical evidence for the literature. Fourth, the use of listed companies, which are among the best companies in the country, as research objects in the existing literature lacks the representativeness of the full sample. Our data, which covers most enterprises across the country, can help us provide representative results to reflect the overall situation in China. Fifth, most existing studies of environmental performance differences of different ownership use comprehensive indicators, including the Kinder, Lydenberg, Domini (KLD) Social Ratings database, or disclosure of CSR reports, which cannot reflect the specific environmental behavior of corporations. We use pollutant emissions, production, and removal as indicators to measure corporate environmental performance. These indicators can reflect the entire process from pollutant generation to emission directly, produce more empirical evidence, and enrich measurements of the corporate environmental performance for the literature.

The remainder of this article is organized as follows. In Section 2, we propose our hypothesis. Section 3 and Section 4 describe the data and econometric models we use to estimate the differences in environmental performance between SOEs and POEs and the impact of privatization on corporate environmental performance, respectively. In section 5, we present our empirical results. Finally, section 6 concludes the article.

There are mixed opinions toward whether SOEs perform better in environmental performance. From the theoretical perspective, the political connection, resource-based theory, and multitask theory are generally used to explain the difference in environmental performance between SOEs and POEs. As one of the measurements of the companies’ political connections, SOEs are considered to be privileged and have greater bargaining power with local governments to negotiate the implementation of environmental policies and the fulfillment of environmental responsibilities (Wang et al. 2003). The existence of political connections leads to lower incentives and awareness for SOEs to improve their environmental performance. From the resource-based perspective, the corporate environmental performance of SOEs can be unclear. On one hand, governments provide SOEs with more resources including subsidies and funds, which SOEs can invest in the technology innovation and the establishment of abatement equipment to realize the improvement of environmental performance (Zhang 2017). However, on the other hand, governments also provide more and cheaper production materials for SOEs, which makes SOEs subject to fewer resource constraints and have lower motivation for efficiency improvement and energy-saving (Hart 1995; Talukdar and Meisner 2001). In the multitask theory, SOEs are viewed as multitask takers rather than profit maximizers, that is, SOEs perform better in social responsibility including environmental protection because they aim to provide social stability and achieve social welfare (Bai et al. 2006).

Corresponding to various theories, empirical research has not reached a consistent result. Jiang et al. (2014) argue that SOEs have the poorest environmental performance compared to foreign-owned and domestic publicly listed firms. Chun (2009) conducts a questionnaire survey in Shanxi Province, China, and find that the environmentalism of SOEs’ employees is lower than that of POEs. Rowe and Guthrie (2010) argue that SOEs’ enforcement of environmental regulation is insufficient, and Maung et al. (2016) find that firms with state ownership pay lower environmental levies due to the political connection. However, other scholars hold the opposite opinion that SOEs outperform in environmental performance compared to their counterparts of other ownership types. Fryxell and Lo (2001) find that SOEs’ managers self-report the strongest environmental ethics value, while managers of POEs self-report the lowest one. Zeng et al. (2012) show that SOEs are more likely to disclose environmental information and Liang and Langbein (2021) argue that provinces with greater relative SOE have better pollution control.

Although existing studies make a great contribution to understanding the difference in environmental performance between SOEs and POEs, they ignore the role of differentiated motivations and strategies of SOEs which may make the environmental performance different even within a narrow measurement and provide a reasonable explanation for the mixed results. To better understand differentiated strategies, we examine the role of governments’ mandatory targets in constructing SOEs’ responsive environmental strategies. SOEs, which are controlled and more tightly supervised by governments, are expected to address more social problems including the environment to meet their shareholders’ interests (Bai et al. 2006). In addition, governments have the authority to decide or largely influence the appointment and retention of SOEs’ executives especially in developing countries (Wong 2004), which produces more pressure on enterprises’ managers to achieve political goals (Liang and Langbein 2021; Liang and Ma 2020). Although the government has placed great emphasis on the improvement of environmental quality, considering that the environment is a comprehensive and complex system containing many aspects and measurements, it is hard for the government to cover everything when enacting policies and regulations. Therefore, policies and regulations can only emphasize the protection and assessment of certain aspects while ignoring others, which generates disproportionate pressure on SOEs to improve certain environmental performance. Therefore, SOEs tend to outperform in environmental performance with mandatory targets.

As for the environmental aspects without mandatory targets, SOEs do not receive disproportionate pressure from governments to improve this unregulated environmental performance, which makes them perform similarly to their counterparts of other ownership types. In addition, SOEs can utilize their political connections with governments to evade their environmental responsibilities, which can even worsen their corporate environmental performance without governments’ mandatory targets. If this is the case, SOEs may perform better only in regulated aspects but not as well in unregulated ones, which means that SOEs take responsive environmental strategies driven mainly by government concerns (Tihanyi et al. 2019). Therefore, we propose the following two hypotheses:

Hypothesis 1a: SOEs perform better in environmental performance with governments’ mandatory targets compared with POEs.

Hypothesis 1b: SOEs perform similarly or even worse in environmental performance without governments’ mandatory targets with POEs.

Although the dominance of SOEs in the Chinese economy, China is undergoing a gradual process of transforming SOEs into POEs to enhance their profitability and competitiveness, which makes privatization has received extensive attention. Most of the existing literature discusses and examines changes in profitability and productivity after the privatization (Berkowitz et al. 2017; Chen et al. 2021; Hsieh and Song 2015), however, discussions on changes in environmental performance are limited. According to Hypothesis 1a and 1b, we can speculate that the environmental strategies of SOEs may change after the privatization due to the change of their motivation and perceived government pressure. Without governments as their controlling shareholders, privatized SOEs no longer have the motivation to better cater to governments’ interests, which may undermine the performance of privatized SOEs in regulated aspects. Therefore, we propose our second hypothesis:

Hypothesis 2: Privatization makes SOEs no longer active in improving government-targeted environmental performance.

3.1 Variables and Measures

Corporate Environmental Performance. Currently, many studies have used resource utilization (Kikeri et al. 1992; Schmid and Robin 1995), sewage charges and illegal penalties (Becker et al. 2013; Pargal and Wheeler 1996; Wang 2000) and the cost of pollution control facilities (Wang 2000) as proxy variables of environmental performance. However, as one of the most direct ways for enterprises to influence the environment, research on the pollutant emissions of enterprises is still limited. SO₂ and NO_x are among the main pollutants emitted by enterprises into the atmosphere and cause serious harm to residents’ health and society. In China, the central government formulates Five-Year Plans (FYP) every five years to clarify the work targets during these years. Since the Ninth FYP (1996–2000) the central government began to limit the total emissions amount of pollutants including SO₂, followed by NO_x which started to be restricted from the Twelfth FYP (2011–2015). Specifically, the central government sets national emission targets and further distributes them to provinces, cities and counties.² This study characterizes the environmental performance of enterprises by the emissions, production, and removal of controlled pollutant SO₂ and noncontrolled pollutant NO_x in 1998–2007 to examine whether SOEs follow environmental responsiveness strategies when undertaking environmental responsibilities. Emissions characterize the final environmental impact of an enterprise, production characterizes the production technology of the enterprise, and removal is used to characterize the “end-of-pipe” of the enterprise.

The Definition of Ownership. Determining how to identify the ownership of an enterprise is another critical issue. At present, there are two main methods for identifying ownership. The first is to use the enterprise registration code to identify ownership according to the firm’s legal registration. Despite the merits of clarity and transparency, this method suffers from some serious shortcomings. First, it cannot effectively identify the ownership of all kinds of enterprises, such as limited liability companies. Second, the ownership of some enterprises may be inconsistent with the actual situation due to the deviated definitions of registration types. In addition, it is difficult to use firms’ legal registrations to identify the ownership of enterprises that are superficially privatized but factually remain under the control of governments (Brandt et al. 2012; Chen et al. 2021; Hsieh and Song 2015; Yu 2015). Therefore, we adopt the second method of Hsieh and Song (2015) to identify SOEs by the proportion of state capital in an enterprise’s registered capital. Specifically, we identify an enterprise as an SOE if the state is reported to be its controlling shareholder or state capital accounts for no less than 50% of the registered capital. Among the remaining companies, when personal capital accounts for no less than 50% of a company’s registered capital, we consider the company to be a POE. Furthermore, we conduct a robustness test by comparing the results obtained by using the legal registration code to identify ownership.

The Definition of Privatization. Referring to the research of Hsieh and Song (2015) and Harrison et al. (2019), we consider an enterprise to be an SOE when the state is its controlling shareholder or the proportion of state capital in the registered capital exceeds 50%. If the controlling shareholder or the proportion of state capital changed over time which leads to an inability to further recognize the enterprise as an SOE, we believe that the company was privatized.

3.2 Data Sources

The data used in the model are mainly from the China Environmental Statistics Database (CESD) and the China Industrial Enterprise Database (CIED). The environmental performance-related variables, i.e., the amounts of SO₂ and NO_x emissions, production, and removal, are derived from the 1998–2007 CESD. The CESD is the most extensive national environmental dataset of China, collected and verified by the Ministry of Environmental Protection. The database contains the annually statistical variables of environmental protection in China since 1985, involving 31 provinces, 113 cities, and 42 industrial sectors.

Information including the legal registration code, state and private capital, registered capital, gross value of output, total assets, and the number of employees is obtained from the 1998–2007 CIED. The CIED is provided by the National Bureau of Statistics, including all state-owned industrial enterprises and private industrial enterprises with annual revenue above 5 million yuan. As of 2011, the database has collected more than 580,000 industrial enterprises, accounting for approximately 95% of China’s total industrial output and covering more than 40 major categories of Chinese industry.

We collected regional characteristic variables, including GDP per capita and the proportions of secondary industries, from the 1999–2008 urban statistical yearbooks published by the National Bureau of Statistics. These statistical yearbooks contain 656 cities nationwide, recording statistics on social and economic development in the previous year. To supplement the missing values in the urban statistical yearbooks, we further searched for them in the provincial statistical yearbooks.

3.3 Summary Statistics

First, we explore the differences in environmental performance between SOEs and POEs. After eliminating the enterprises whose ownership cannot be identified, the sample contained 191,890 observations, with SOEs accounting for 52.10% and POEs accounting for 47.90%. The changes in the numbers of SOEs and POEs year by year are shown in Fig. 1. We can see that the number of SOEs showed a significant downward trend while the number of private enterprises showed a clear upward trend between 1998 and 2007. In 1998, the number of POEs was only approximately one-sixth that of SOEs. In 2003, the number of POEs exceeded the number of SOEs. By 2007, the number of SOEs was less than half that of POEs. These results are in line with the trends mentioned earlier on the reform and reduction of SOEs and the growth of POEs.

Second, we explore how the privatization of SOEs affects the environmental performance of enterprises and SOEs’ environmental performance strategies. After retaining the enterprises that are identified as SOEs initially and exist for more than one year, there are 19,202 enterprises (97,617 observations) in our sample. Of those, 29.77% have been privatized, while the remaining 70.23% have not. The number of SOEs privatized each year is shown in Fig. 2. We can see that the process of privatization of SOEs exhibited a trend of first rising and then falling. The number of privatized SOEs reached a peak in 2004 when more than 1,000 SOEs were privatized. By 2007, the number of SOEs privatized that year was less than 400.

Table 1 presents the descriptive statistics of the variables of SOEs and POEs. As shown in the table, the production situations and environmental performance of SOEs and POEs are quite different. Specifically, SOEs far exceeded POEs in total output, total capital, and the number of employees. As for environmental performance, SOEs produced and emitted more SO₂ and NO_x than do POEs. However, SOEs also removed more SO₂ and NO_x than do POEs.

Table 1

Descriptive Statistics
Data Source	Variable	Mean Value			Observations
Data Source	Variable	SOEs	POEs	Total	Observations
CIED	Output (10 million yuan)	28.99	9.07	19.45	191890
	Output (10 million yuan)	(177.61)	(35.08)	(130.86)
	Total Capital (10 million yuan)	42.56	7.03	25.54	191890
	Total Capital (10 million yuan)	(211.49)	(27.25)	(154.84)
	Employees (person)	1176.87	315.23	764.18	191890
	Employees (person)	(3722.69)	(773.22)	(2773.51)
CESD	SO₂ Emissions (kiloton)	0.68	0.11	0.40	191890
	SO₂ Emissions (kiloton)	(8.09)	(0.51)	(5.86)
	SO₂ Production (kiloton)	1.08	0.15	0.63	191890
	SO₂ Production (kiloton)	(12.85)	(0.97)	(9.31)
	SO₂ Removal (kiloton)	0.40	0.05	0.23	191890
		(8.24)	(0.76)	(5.98)
	NO_x Emissions (kiloton)	0.22	0.06	0.11	39310
		(1.45)	(0.30)	(0.88)
	NO_x Production (kiloton)	0.24	0.06	0.12	39310
		(1.58)	(0.31)	(0.96)
	NO_x Removal (kiloton)	0.02	0.002	0.01	39310
		(0.46)	(0.06)	(0.27)
Note: This table presents descriptive statistics of the production-related variables from the CIED and environment-related variables from the CESD of SOEs and POEs. The ownership type is identified using the proportion of registered capital. We eliminate the enterprises whose ownership cannot be identified. Because the CESD only recorded NO_x-related variables in 2006 and 2007, the number of observations decreases with regard to NO_x. The data are averaged by category, with standard deviations given in parentheses.

4.1 Differences Between Ownership Types

In this section, we construct empirical models to analyze whether there are significant differences in the environmental performance characterized by SO₂ and NO_x between enterprises with different ownership types and whether the privatization of SOEs leads to changes in the environmental strategies and performance of enterprises. First, to explore the existence of environmental responsiveness strategies, we compare the environmental differences between SOEs and POEs in terms of SO₂ and NO_x. The empirical model is as follows:

$${Eviper}_{ict}={\beta }_{1}{SOE}_{ict}+{\beta }_{2}{Produ}_{ict}+{\beta }_{3}{Region}_{ct}+{{\theta }_{t}+{\theta }_{j}+{\theta }_{p}+\theta }_{tj}+{\theta }_{tp}+{\epsilon }_{ict}$$

1

where Eviper_ict is the environmental performance variable of enterprise i in city c and year t, measured by the amounts of SO₂ and NO_x emissions, production, and removal. SOE_ict is the dummy variable to denote whether enterprise i is an SOE in city c and year t.³ Additionally, different ownership enterprises may have differences in production and operating status, size, and capital and labor inputs. These factors impact the environmental performance of enterprises. Therefore, it is necessary to control the production-related variables of enterprises. Produ_ict refers to the production-related variables for enterprise i in city c and year t, including the gross value of output, total assets, and the number of employees, to characterize the size, capital and labor level of the enterprise. The stringency of regional environmental regulation also impacts the environmental performance of the enterprise, which varies with the regional economic development level and economic structure, making it necessary to further control the regional characteristics. Region_ct is the regional characteristic variable of city c in year t, including GDP per capita and the proportion of secondary industry in GDP. Considering that there may be significant differences in the environmental performance of enterprises among different years and industries, these factors need to be further controlled. θ_t represents the year fixed effects, and θ_j represents the industry fixed effects. In addition, many policies, such as emission standards and sewage charges, are formulated at the provincial level, making it important to control the province fixed effects, which are represented by the variable θ_p. θ_tj represents the year-industry fixed effects, θ_tp represents the year-province fixed effects, and ${\epsilon }_{ict}$ is the error term. The logarithms of all variables are used in the regression except for the dummy variable. For variables with zero values, we uniformly add 1 to their original values.

4.2 Impact of Privatization

The above analysis does not control the inherent characteristics of enterprises, which leads to endogeneity problems. By adopting Chen et al.’s (2021) strategy, we compare differences in environmental performance by using privatization as a quasi-natural experiment. Specifically, we use the DID method to compare the differences between privatized SOEs and non-privatized SOEs before and after the privatization. In this way, we can further verify the existence of environmental performance differences between SOEs and POEs. The empirical model is as follows:

$${Eviper}_{ict}={\beta }_{1}{Privatize}_{ict}+{\beta }_{2}{Produ}_{ict}+{\beta }_{3}{Region}_{ct}+{{\theta }_{t}+\theta }_{tj}+{\theta }_{tp}+{\theta }_{i}+{\epsilon }_{ict}$$

2

In formula (2), Privatize_ict refers to whether SOE i has been privatized into a POE in city c and year t. The variable is equal to 1 when the privatized SOEs have already been privatized; otherwise, it is equal to 0 for non-privatized SOEs and privatized SOEs before privatization. In addition, θ_i represents the firm fixed effects. The meanings of other variables in formula (2) are the same as those in formula (1)

5.1 The Existence of SOEs’ Environmental Responsiveness Strategies

As mentioned earlier, SOEs may follow environmental responsiveness strategies that make them perform better in aspects regulated by governments. However, the differences may also be caused by factors such as capital and subsidies, in which cases the difference between SOEs and POEs will exist among all pollutants. To examine whether the environmental responsiveness strategies of SOEs exist, we compare the results between the controlled pollutant SO₂ and noncontrolled pollutant NO_x. Since the emissions of pollutants characterize the direct impact of enterprises on the environment, we first conduct an ordinary least square (OLS) regression with the SO₂ and NO_x emissions as the dependent variables. Columns (1) and (2) in Table 2 are dependent on SO₂ emissions, while columns (3) and (4) are dependent on NO_x. Columns (1) and (3) control the enterprise-related variables and fixed effects. Columns (2) and (4) further control the city-related variables based on columns (1) and (3). According to the results, the coefficient of SO₂ is negative and is significant at the 1% level while the coefficient of NO_x is positive and not significant. To eliminate interference from other factors as much as possible, we include all control variables and fixed effects in the subsequent regression. Based on the results in column (2), there are significant differences in SO₂ emissions between SOEs and POEs. In order to reflect the degree of influence of the coefficients more accurately, we perform a semi-logarithmic transformation on all the interpretations of the coefficients, i.e., ${\beta }_{accurate}={e}^{\beta }-1$. SOEs emit less SO₂ than POEs by 7.13%, which indicates that they have better environmental performance in regulated aspects. However, there are no significant differences in noncontrolled NO_x between SOEs and POEs in 2006 and 2007, which is in stark contrast with the results for SO₂. This indicates that SOEs are more incentivized to meet policy targets, i.e., the environmental responsiveness strategies of SOEs do exist. SOEs can perform quite well and be leaders in mandated environmental performance which is in line with our Hypothesis 1a. However, they may not necessarily do better and may even perform worse in unregulated aspects, which is consistent with our Hypothesis 1b.

Table 2

Differences in SO₂ and NO_x Emissions Between SOEs and POEs
	(1)	(2)	(3)	(4)
	SO₂	SO₂	NO_x	NO_x
SOEs	-0.0756^***	-0.0740^***	0.00707	0.0234
	(0.0287)	(0.0282)	(0.0333)	(0.0327)
Output	0.244^***	0.248^***	0.129^***	0.130^***
	(0.0139)	(0.0140)	(0.0190)	(0.0186)
Labor	0.392^***	0.382^***	0.344^***	0.338^***
	(0.0202)	(0.0197)	(0.0260)	(0.0255)
Total capital	0.125^***	0.134^***	0.251^***	0.254^***
	(0.0155)	(0.0155)	(0.0201)	(0.0202)
GDP per capita	-	-0.162^***	-	-0.147
	-	(0.0496)	-	(0.0910)
Secondary industry ratio	-	0.0579	-	0.442^**
	-	(0.0659)	-	(0.204)
Year FE	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes
Province FE	Yes	Yes	Yes	Yes
Year-industry FE	Yes	Yes	Yes	Yes
Year-province FE	Yes	Yes	Yes	Yes
Observations	191908	191890	39315	39310
adj. R²	0.412	0.413	0.342	0.342
Note: Columns (1) and (2) are dependent on SO₂ emissions, while columns (3) and (4) are dependent on NO_x emissions. Columns (1) and (3) control the enterprise-related variables, year fixed effects, province fixed effects, industry fixed effects, year-province fixed effects, and year-industry fixed effects. Columns (2) and (4) further control the city-related variables. The logarithms of all variables are used in the regression except for the dummy variable. Standard errors are clustered at the city level and given in parentheses; ^* p < 0.1, ^ p < 0.05, ^* p < 0.01.

Since the pollutant emissions of an enterprise are closely related to the production technology level and the “end-of-pipe” treatment level of the enterprise, we further investigate the reason for the differences in SO₂ emissions between different ownership types from the two aspects of the production level and the “end-of-pipe” treatment level. We characterize the production technology level and the “end-of-pipe” treatment level of an enterprise by the amount of SO₂ production and the removal of SO₂, respectively. In addition, we also take the amount of NO_x production and removal as independent variables to examine whether the production and removal of NO_x are the same between different ownership types. The results are shown in Table 3. The results show that SOEs produce less SO₂ than POEs by 5.04% and remove more SO₂ than POEs by 18.53%, caeteris paribus, indicating that there are significant differences in the production technology level and “end-of-pipe” treatment level between SOEs and POEs. The differences in SO₂ emissions between SOEs and POEs are mainly due to the combined effects of the differences in the production technology level and “end-of-pipe” treatment level between them. The results also indicate that SOEs have made more efforts and a greater contribution than POEs to the reduction of pollutants represented by SO₂. As for NO_x, there are no significant differences between SOEs and POEs in either NO_x production or removal. Although not statistically significant, the coefficient’s sign for NO_x production is positive while is negative for NO_x removal, which further confirms that SOEs do not perform better and may even perform worse in unregulated behavior and performance.

Table 3

Differences in SO₂ and NO_x Production and Removal Between SOEs and POEs
	(1) SO₂ Production	(2) SO₂ Removal	(3) NO_x Production	(4) NO_x Removal
SOEs	-0.0517^*	0.170^***	0.0146	-0.0649
	(0.0292)	(0.0595)	(0.0312)	(0.0401)
Output	0.255^***	0.147^***	0.129^***	-0.0219
	(0.0139)	(0.0277)	(0.0184)	(0.0201)
Labor	0.399^***	0.486^***	0.338^***	0.0665^***
	(0.0199)	(0.0414)	(0.0263)	(0.0244)
Total capital	0.159^***	0.347^***	0.263^***	0.0778^***
	(0.0159)	(0.0341)	(0.0200)	(0.0217)
GDP per capita	-0.127^**	0.453^***	-0.142	0.0317
	(0.0522)	(0.140)	(0.0900)	(0.0785)
Secondary industry ratio	0.0501	-0.233	0.438^**	0.383
	(0.0661)	(0.194)	(0.196)	(0.268)
Year FE	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes
Province FE	Yes	Yes	Yes	Yes
Year-industry FE	Yes	Yes	Yes	Yes
Year-province FE	Yes	Yes	Yes	Yes
Observations	191890	191890	39310	39310
adj. R²	0.429	0.167	0.358	0.028
Note: The ownership type in the table is identified using the proportion of registered capital. The logarithms of all variables are used in the regression except for the dummy variable. Standard errors are clustered at the city level and given in parentheses; ^* p < 0.1, ^ p < 0.05, ^* p < 0.01.

The method we use to identify the ownership types may affect the differences we observed in environmental performance between SOEs and POEs. Therefore, we use the legal registration code to identify ownership to test the robustness of the differences in SO₂ and NO_x performance between SOEs and POEs. The results are shown in Table 4. We can see that the number of SOEs and POEs identified by the legal registration is much less than that by the proportion of registered capital, which further proves that the problem in which only a limited number of enterprises can be recognized by legal registration does exist. The results in Table 4 are quite similar to those in Table 2 and Table 3. SOEs produce less, remove more, and thus emit less SO₂ than do POEs, ceteris paribus. The differences in SO₂ emissions between SOEs and POEs are caused by the differences between the level of production technology and the level of “end-of-pipe” treatment. As for unregulated NOx, there are no significant differences between SOEs and POEs in emissions, production, or removal. The results are robust when using different ownership identification, that is, SOEs outperform only in environmental performance with mandatory targets compared with POEs.

Table 4

Regression with Legal Registration as the Identification Method
	(1) SO₂ Emissions	(2) SO₂ Production	(3) SO₂ Removal	(4) NO_x Emissions	(5) NO_x Production	(6) NO_x Removal
SOEs	-0.176^***	-0.144^***	0.245^***	0.0115	0.0179	0.0182
	(0.0438)	(0.0458)	(0.0865)	(0.0664)	(0.0652)	(0.0692)
Output	0.240^***	0.245^***	0.193^***	0.0969^***	0.100^***	-0.00675
	(0.0148)	(0.0148)	(0.0313)	(0.0232)	(0.0230)	(0.0261)
Labor	0.379^***	0.394^***	0.407^***	0.345^***	0.335^***	0.0159
	(0.0238)	(0.0244)	(0.0468)	(0.0333)	(0.0335)	(0.0370)
Total capital	0.144^***	0.169^***	0.378^***	0.250^***	0.258^***	0.0821^***
	(0.0166)	(0.0173)	(0.0382)	(0.0250)	(0.0244)	(0.0278)
GDP per capita	-0.122^**	-0.0811	0.535^***	-0.108	-0.0994	0.0183
	(0.0497)	(0.0525)	(0.140)	(0.0880)	(0.0860)	(0.0980)
Secondary industry ratio	-0.00224	0.00171	-0.221	0.420^*	0.421^*	0.323
	(0.0728)	(0.0755)	(0.203)	(0.227)	(0.215)	(0.374)
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Province FE	Yes	Yes	Yes	Yes	Yes	Yes
Year-industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Year-province FE	Yes	Yes	Yes	Yes	Yes	Yes
Observations	137304	137304	137304	28447	28447	28447
adj. R²	0.396	0.412	0.170	0.318	0.334	0.035
Note: The ownership type in the table is identified using the enterprises’ legal registration codes. The logarithms of all variables are used in the regression except for the dummy variable. Standard errors are clustered at the city level and given in parentheses; ^* p < 0.1, ^ p < 0.05, ^* p < 0.01.

Although the reduction in pollutants’ total emissions has always been an important part of environmental governance in China, considering the sustainable and synergistic growth of the economy and the environment, the degree of enterprises’ clean production, which is measured by the intensity of pollutants, is also an aspect of great concern to us. Therefore, we further explore whether there are differences between SOEs and POEs in the pollution intensity, measured by the ratio of emissions and removal to production, as well as emissions, production, and removal per unit output value. The results for SO₂ are shown in Table 5. Similar to the results in Table 3, we can still find that SOEs produce less, remove more, and thus emit less SO₂ than do POEs. Although the coefficients of columns (3) and (5) are not statistically significant, their signs are still in line with our basic specifications. As for the results for NO_x shown in Table 6, most of the coefficients are insignificant, indicating that there are no significant differences in the unregulated pollutant between SOEs and POEs. However, it is worth noting that the coefficients of the ratio of removal to production and emissions per unit output value are significant at the 10% level in columns (2) and (3), showing that SOEs remove less and emit more NO_x than POEs. These results to some extent indicate that SOEs can perform worse in environmental performance without governments’ mandatory targets.

Table 5

Differences in SO₂ Intensity Between SOEs and POEs
	(1) SO₂ Emissions/ Production	(2) SO₂ Removal/ Production	(3) SO₂ Emissions/ Output	(4) SO₂ Production/ Output	(5) SO₂ Removal/ Output
SOEs	-0.0196^***	0.0434^**	-0.0117	-0.0517^*	0.0111
	(0.00646)	(0.0215)	(0.0105)	(0.0292)	(0.00691)
Output	-0.00619^**	0.0209^**	-0.310^***	-0.745^***	-0.0718^***
	(0.00262)	(0.00962)	(0.00983)	(0.0139)	(0.00962)
Labor	-0.0174^***	0.110^***	0.152^***	0.399^***	0.0558^***
	(0.00366)	(0.0132)	(0.00814)	(0.0199)	(0.00539)
Total capital	-0.0229^***	0.105^***	0.0755^***	0.159^***	0.0427^***
	(0.00301)	(0.0116)	(0.00841)	(0.0159)	(0.00813)
GDP per capita	-0.0341^***	0.179^***	-0.0631^***	-0.127^**	0.0296**
	(0.0116)	(0.0501)	(0.0188)	(0.0522)	(0.0130)
Secondary industry ratio	0.00529	-0.0674	-0.0141	0.0501	-0.0163
	(0.0202)	(0.0691)	(0.0285)	(0.0661)	(0.0222)
Year FE	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes
Province FE	Yes	Yes	Yes	Yes	Yes
Year-industry FE	Yes	Yes	Yes	Yes	Yes
Year-province FE	Yes	Yes	Yes	Yes	Yes
Observations	191890	191890	191890	191890	191890
adj. R²	0.080	0.134	0.430	0.404	0.126
Note: The ownership type in the table is identified using the proportion of registered capital. The logarithms of all variables are used in the regression except for the dummy variable. Standard errors are clustered at the city level and given in parentheses; ^* p < 0.1, ^ p < 0.05, ^* p < 0.01.\

Table 6

Differences in NO_x Intensity Between SOEs and POEs
	(1) NO_x Emissions/ Production	(2) NO_x Removal/ Production	(3) NO_x Emissions/ Output	(4) NO_x Production/ Output	(5) NO_x Removal/ Output
SOEs	0.00700	-0.0297^*	0.0155^*	0.0146	-0.00174
	(0.00460)	(0.0156)	(0.00936)	(0.0312)	(0.00233)
Output	0.00110	-0.0112	-0.223^***	-0.871^***	-0.0113^***
	(0.00266)	(0.00824)	(0.00787)	(0.0184)	(0.00230)
Labor	-0.00141	0.0175^**	0.0758^***	0.338^***	0.00226
	(0.00308)	(0.00884)	(0.00797)	(0.0263)	(0.00140)
Total capital	-0.00619^**	0.0238^***	0.0871^***	0.263^***	0.00776^***
	(0.00256)	(0.00826)	(0.00700)	(0.0200)	(0.00178)
GDP per capita	-0.00608	0.0129	-0.0431^**	-0.142	0.00182
	(0.00740)	(0.0296)	(0.0195)	(0.0900)	(0.00367)
Secondary industry ratio	-0.0102	0.133	0.0736	0.438^**	0.0272
	(0.0245)	(0.0996)	(0.0528)	(0.196)	(0.0192)
Year FE	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes
Province FE	Yes	Yes	Yes	Yes	Yes
Year-industry FE	Yes	Yes	Yes	Yes	Yes
Year-province FE	Yes	Yes	Yes	Yes	Yes
Observations	39310	39310	39310	39310	39310
adj. R²	0.008	0.023	0.341	0.379	0.019
Note: The ownership type in the table is identified using the proportion of registered capital. The logarithms of all variables are used in the regression except for the dummy variable. Standard errors are clustered at the city level and given in parentheses; ^* p < 0.1, ^ p < 0.05, ^* p < 0.01.

5.2 The Impacts of Privatization

We then explore the effects of the privatization of SOEs on the environmental performance of enterprises using the DID method, in which we control for the firm fixed effects simultaneously. The results are shown in Table 7. According to the results in columns (1) to (3), we can see that the coefficient of privatization is positive and significant at the level of 1% in columns (1) and (2). Conversely, the coefficient of privatization is not statistically significant in column (3). The above results indicate that privatized SOEs significantly increased their SO₂ production and emissions after privatization by 7.61% and 8.18%, respectively. There is no significant change in SO₂ removal after privatization, which means there is no significant change in the level of “end-of-pipe” treatment after privatization. Therefore, the deteriorating environmental performance of privatized SOEs mainly comes from changes in the production process. To summarize, compared to non-privatized SOEs, there is a significant decline in the SO₂ performance of privatized SOEs, which further corroborates that SOEs perform better than POEs in environmental performance regulated by governments. These results also indicate that privatized SOEs may be less responsive to policies, which means that their environmental responsiveness strategies may change after privatization. As for the results for NO_x in columns (4) to (6), there is no significant change after privatization. This is in line with our hypothesis that the unregulated environmental performance of SOEs can be similar to their privately-owned counterparts, which makes privatization not have any effects on their unregulated performance.

Table 7

The Impact of Privatization
	(1) SO₂ Emissions	(2) SO₂ Production	(3) SO₂ Removal	(4) NO_x Emissions	(5) NO_x Production	(6) NO_x Removal
Privatize	0.0786^***	0.0733^***	0.0779	0.0147	0.00567	-0.146
	(0.0261)	(0.0248)	(0.0781)	(0.135)	(0.134)	(0.225)
Output	0.210^***	0.219^***	0.148^***	0.125^*	0.133^**	0.0706
	(0.0163)	(0.0154)	(0.0327)	(0.0676)	(0.0631)	(0.116)
Labor	0.101^***	0.104^***	0.115^**	0.0218	0.0254	0.111
	(0.0160)	(0.0148)	(0.0521)	(0.0734)	(0.0714)	(0.212)
Total capital	0.0294^*	0.0347^**	0.0892^*	0.0742	0.0767	0.0583
	(0.0158)	(0.0148)	(0.0472)	(0.0818)	(0.0801)	(0.139)
GDP per capita	-0.0233	-0.0391	0.0745	0.859	0.904	0.676
	(0.0724)	(0.0580)	(0.287)	(0.763)	(0.615)	(1.713)
Secondary industry ratio	0.121	0.132^*	0.146	-0.648	-0.547	1.259
	(0.0871)	(0.0710)	(0.165)	(0.978)	(0.850)	(1.613)
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Year-industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Year-province FE	Yes	Yes	Yes	Yes	Yes	Yes
Firm FE	Yes	Yes	Yes	Yes	Yes	Yes
Observations	102793	102793	102793	12590	12590	12590
adj. R²	0.814	0.835	0.593	0.850	0.879	0.531
Note: The Privatize variable is equal to 1 after the privatization of privatized SOEs; otherwise, it is equal to 0. The logarithms of all variables are used in the regression except for the dummy variable. Standard errors are clustered at the city level and given in parentheses; ^* p < 0.1, ^ p < 0.05, ^* p < 0.01.

To confirm that the control group can be the counterfactual of the treatment group, we conduct a parallel trend test to ensure that privatized SOEs and non-privatized SOEs have a common trend before privatization. It is a pity that we can only obtain the data of NO_x in 2006 and 2007, which makes it impossible for us to test the parallel trend for NO_x. Therefore, we only conduct the parallel trend test for SO₂ here. Since the privatization of privatized SOEs happened at different time points, we conduct the test by including a series of dummy variables of years before and after privatization in the model, referring to the study of Beck et al. (2010). The specific model is as follows:

$${Eviper}_{ict}=\alpha +{\beta }_{1}{D}_{ict}^{-5}+{\beta }_{2}{D}_{ict}^{-4}+\dots +{\beta }_{9}{D}_{ict}^{+4}+{\beta }_{10}{D}_{ict}^{+5}+{\theta }_{t}+{\theta }_{tj}+{\theta }_{tp}+{\theta }_{i}+{\epsilon }_{ict}$$

3

In formula (3), for the dummy variables of years before and after privatization, ${D}_{ict}^{-j}$ (${D}_{ict}^{+j}$) takes the value of 1 in the jth year before (after) the privatization of the privatized SOEs. The value of ${D}_{ict}^{-5}$ (${D}_{ict}^{+5}$) is equal to 1 in all years five years or more before (after) the privatization of the privatized SOEs. The remaining variables in formula (3) have the same meanings as those in formula (1). It should be noted that we exclude the privatization year in our model, so the model uses non-privatized SOEs and the privatization year of privatized SOEs as the benchmark. We plot the distribution of the dummy variable coefficients and their 95% confidence intervals in Fig. 3. Panel A shows the regression results with SO₂ emissions as the dependent variable. Panel B shows the regression results with SO₂ production as the dependent variable. Panel C shows the regression results with SO₂ removal as the dependent variable. On one hand, according to Panel A and Panel B, we can see that there are no significant differences between the treatment group and the control group in SO₂ emissions and production for no more than three years before privatization, and this finding is in line with the parallel trend assumption. After privatization, the SO₂ production and emissions of the treatment group are significantly greater than those of the control group. In addition, the gap between privatized SOEs and non-privatized SOEs shows a clear upward trend with time. The above results indicate that the privatization of SOEs does lead to increases in SO₂ production and emissions. On the other hand, according to the results shown in Panel C, although the coefficients of the SO₂ removal show a significant upward trend after privatization, the coefficients in all years except for the first year after privatization are not significantly different from 0. Therefore, the privatization of SOEs does not lead to significant changes in the removal of SO₂.

People may worry about the selection bias, that is, those privatized SOEs have poor environmental performance before privatization than SOEs whose ownership has not changed. To alleviate this concern, Table 8 shows the differences in environmental performance between privatized SOEs before privatization and SOEs whose ownership has not changed. We can see that there are no significant differences between privatized SOEs and SOEs except for the SO₂ removal. Combined with the results in Table 7, privatized SOEs significantly increase their SO₂ production and emissions after privatization, the coefficients of which are insignificant in Table 8, indicating that the impacts of the privatization are not caused by the selection bias.

Table 8

Differences in Emission, Production and Removal Between Privatized SOEs Before Privatization and SOEs
	(1) SO₂ Emissions	(2) SO₂ Production	(3) SO₂ Removal	(4) NO_x Emissions	(5) NO_x Production	(6) NO_x Removal
Privatized	0.0569	0.0428	-0.202^**	0.0818	0.0785	0.136
	(0.0360)	(0.0354)	(0.0883)	(0.0707)	(0.0700)	(0.0945)
Output	0.274^***	0.283^***	0.140^***	0.119^***	0.117^***	-0.000381
	(0.0235)	(0.0235)	(0.0347)	(0.0261)	(0.0254)	(0.0262)
Labor	0.444^***	0.471^***	0.751^***	0.355^***	0.358^***	0.130^***
	(0.0226)	(0.0234)	(0.0592)	(0.0352)	(0.0349)	(0.0468)
Total capital	0.150^***	0.167^***	0.278^***	0.313^***	0.324^***	0.0577
	(0.0220)	(0.0230)	(0.0463)	(0.0321)	(0.0313)	(0.0371)
GDP per capita	-0.298^***	-0.267^***	0.286^*	-0.353^***	-0.329^***	0.0596
	(0.0463)	(0.0480)	(0.160)	(0.0752)	(0.0755)	(0.0841)
Secondary industry ratio	0.158^**	0.140^*	-0.210	0.530^***	0.532^***	0.317
	(0.0719)	(0.0754)	(0.206)	(0.181)	(0.178)	(0.246)
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Province FE	Yes	Yes	Yes	Yes	Yes	Yes
Year-industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Year-province FE	Yes	Yes	Yes	Yes	Yes	Yes
Observations	76074	76074	76074	11407	11407	11407
adj. R²	0.500	0.514	0.205	0.412	0.428	0.032
Note: The Privatized variable is equal to 1 for privatized SOEs before privatization; otherwise, it is equal to 0. The ownership type in the table is identified using the proportion of registered capital. Privatized SOEs here are privatized SOEs before privatization and SOEs here are SOEs whose ownership has not changed. The logarithms of all variables are used in the regression except for the dummy variable. Standard errors are clustered at the city level and given in parentheses; ^* p < 0.1, ^ p < 0.05, ^* p < 0.01.

Although SOEs have the deficit of being inefficient, it should not be denied that they play a vital role in meeting governments’ requirements and fulfilling policies goals. Using firm-level data, we first explore the differences between SOEs and POEs in environmental performance with and without governments’ mandatory targets to further understand the differentiated environmental strategies adopted by SOEs. In addition, we explore how the privatization of SOEs affects the environmental performance of SOEs in the context of the reform of SOEs. The results show that the environmental performance of SOEs and POEs significantly differs only in aspects with mandatory targets, which means that SOEs follow environmental responsiveness strategies to fulfill the governments’ goals and requirements. Moreover, the privatization of SOEs undermines their environmental performance, which indicates that the discussion on the reform of SOEs should not be limited to the production efficiencies of enterprises but should include multiple angles.

Our study provides important policy implications. SOEs have long been considered to play a vital role in maintaining low unemployment and promoting social stability (Clò et al. 2017; Wang and Jin 2007). Apart from these goals, according to our findings, SOEs are leaders in policy goal accomplishment and compliance, which makes them important for achieving political expectations. Given that the achievement of government goals is a major consideration for SOEs in fulfilling social responsibilities, governments should consider how to set policy objectives to provide incentives for SOEs in providing public services and improving social welfare. With the privatization of SOEs worldwide, though, determining how to avoid the negative effects of privatization as well as to motivate private companies to achieve governments’ goals of providing public goods is another issue.

However, several limitations still exist in our study. First, although we improve our representativeness by utilizing the data from CIED rather than only analyzing the performance of listed companies, we can not obtain information about private industrial enterprises with annual revenue of fewer than 5 million yuan. Second, we examine the existence of responsiveness strategies in terms of environmental performance but more evidence is still needed in other kinds of social responsibilities, such as employee welfare and social sustainability. Third, we prove that the responsiveness to the government’s concern is an important source of SOE’s differentiated strategies, while future research is urged to explore other factors that can induce differentiated corporate strategies.

1 For the report, see https://baijiahao.baidu.com/s?id=1634484734490918890&wfr=spider&for=pc

2 The ninth FYP (1996-2000) set the emission target of SO2 to emit no more than 23.6 million tons in 2000. In the tenth FYP (2001-2005) and the eleventh FYP (2006-2010), the central government demanded to reduce 10% of 2000 SO2 emissions in 2005 and 10% of 2005 SO2 emissions in 2010, respectively. The twelfth FYP (2011-2015) demanded to reduce 8% of 2010 SO2 emissions and 10% of 2010 NOx emissions in 2015, respectively. In the thirteenth FYP (2016-2020), the emission targets of SO2 and NOx are to reduce 15% of 2015 emissions in 2020, respectively. For detailed contents, see http://www.gov.cn/

3 It should be noted that there are still many companies in our sample whose ownership cannot be identified through the above methods, including collectively-owned enterprises; Hong Kong-, Macao- and Taiwan-owned enterprises; foreign-owned enterprises; legal person-owned enterprises; and other enterprises whose controlling shareholder’s proportion of registered capital is below 50%. As mentioned before, our focus is on a comparison between SOEs and POEs, for which reason we eliminated these companies that cannot be identified as SOEs or POEs in our study.

The authors declare that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.

Abreu, M. C. S. d. (2009). How to define an environmental policy to improve corporate sustainability in developing countries. Business Strategy and the Environment, 18(8), 542–556, doi:https://doi.org/10.1002/bse.625.
Albalate, D., Bel, G., González-Gómez, F., & Picazo-Tadeo, A. J. (2020). Contract Renewal in Urban Water Services, Incumbent Advantage, and Market Concentration. [https://doi.org/10.1111/puar.13282]. Public Administration Review, n/a(n/a), doi:https://doi.org/10.1111/puar.13282.
Arena, C., Michelon, G., & Trojanowski, G. (2018). Big Egos Can Be Green: A Study of CEO Hubris and Environmental Innovation. British Journal of Management, 29(2), 316–336, doi:https://doi.org/10.1111/1467-8551.12250.
Bai, C.-E., Lu, J., & Tao, Z. (2006). The Multitask Theory of State Enterprise Reform: Empirical Evidence from China. American Economic Review, 96(2), 353–357, doi:10.1257/000282806777212125.
Bai, C.-E., Lu, J., & Tao, Z. (2009). How does privatization work in China? Journal of Comparative Economics, 37(3), 453–470, doi:https://doi.org/10.1016/j.jce.2008.09.006.
Beck, T., Levine, R., & Levkov, A. (2010). Big Bad Banks? The Winners and Losers from Bank Deregulation in the United States. The Journal of Finance, 65(5), 1637–1667, doi:https://doi.org/10.1111/j.1540-6261.2010.01589.x.
Becker, R. A., Pasurka Jr, C., & Shadbegian, R. J. (2013). Do environmental regulations disproportionately affect small businesses? Evidence from the Pollution Abatement Costs and Expenditures survey. Journal of Environmental Economics and Management, 66(3), 523–538, doi:https://doi.org/10.1016/j.jeem.2013.08.001.
Berkowitz, D., Ma, H., & Nishioka, S. (2017). Recasting the Iron Rice Bowl: The Reform of China's State-Owned Enterprises. Review of Economics and Statistics, 99(4), 735–747, doi:10.1162/REST_a_00637.
Berrone, P., Fosfuri, A., & Gelabert, L. (2017). Does Greenwashing Pay Off? Understanding the Relationship Between Environmental Actions and Environmental Legitimacy. Journal of Business Ethics, 144(2), 363–379, doi:10.1007/s10551-015-2816-9.
Berry, M. A., & Rondinelli, D. A. (1998). Proactive corporate environmental management: A new industrial revolution. Academy of Management Perspectives, 12(2), 38–50, doi:10.5465/ame.1998.650515.
Brandt, L., Jiang, F., Luo, Y., & Su, Y. (2020). Ownership and Productivity in Vertically-Integrated Firms: Evidence from the Chinese Steel Industry. The Review of Economics and Statistics, 1–49, doi:10.1162/rest_a_00923.
Brandt, L., Van Biesebroeck, J., & Zhang, Y. (2012). Creative accounting or creative destruction? Firm-level productivity growth in Chinese manufacturing. Journal of Development Economics, 97(2), 339–351, doi:https://doi.org/10.1016/j.jdeveco.2011.02.002.
Brown, J. D., Earle, John S., & Telegdy, Á. (2006). The Productivity Effects of Privatization: Longitudinal Estimates from Hungary, Romania, Russia, and Ukraine. Journal of political economy, 114(1), 61–99, doi:10.1086/499547.
Bundy, J., Shropshire, C., & Buchholtz, A. K. (2013). Strategic Cognition and Issue Salience: Toward an Explanation of Firm Responsiveness to Stakeholder Concerns. Academy of management Review, 38(3), 352–376, doi:10.5465/amr.2011.0179.
Chen, Y.-S. (2008). The Positive Effect of Green Intellectual Capital on Competitive Advantages of Firms. Journal of Business Ethics, 77(3), 271–286, doi:10.1007/s10551-006-9349-1.
Chen, Y.-S., & Chang, C.-H. (2013). The Determinants of Green Product Development Performance: Green Dynamic Capabilities, Green Transformational Leadership, and Green Creativity. Journal of Business Ethics, 116(1), 107–119, doi:10.1007/s10551-012-1452-x.
Chen, Y., Igami, M., Sawada, M., & Xiao, M. (2021). Privatization and Productivity in China. RAND Journal of Economics, Forthcoming, doi:http://dx.doi.org/10.2139/ssrn.2695933.
Chuang, S.-P., & Huang, S.-J. (2015). Effects of Business Greening and Green IT Capital on Business Competitiveness. Journal of Business Ethics, 128(1), 221–231, doi:10.1007/s10551-014-2094-y.
Chun, R. (2009). Ethical Values and Environmentalism in China: Comparing Employees from State-Owned and Private Firms. Journal of Business Ethics, 84(3), 341–348, doi:10.1007/s10551-009-0202-1.
Clò, S., Ferraris, M., & Florio, M. (2017). Ownership and environmental regulation: Evidence from the European electricity industry. Energy Economics, 61, 298–312, doi:https://doi.org/10.1016/j.eneco.2016.12.001.
Dahlmann, F., Branicki, L., & Brammer, S. (2019). Managing Carbon Aspirations: The Influence of Corporate Climate Change Targets on Environmental Performance. Journal of Business Ethics, 158(1), 1–24, doi:10.1007/s10551-017-3731-z.
Dögl, C., & Holtbrügge, D. (2014). Corporate environmental responsibility, employer reputation and employee commitment: an empirical study in developed and emerging economies. The International Journal of Human Resource Management, 25(12), 1739–1762, doi:10.1080/09585192.2013.859164.
Dooley, R. S., & Fryxell, G. E. (1999). Are Conglomerates Less Environmentally Responsible? An Empirical Examination of Diversification Strategy and Subsidiary Pollution in the U.S. Chemical Industry. Journal of Business Ethics, 21(1), 1–14, doi:10.1023/A:1006221229985.
Duanmu, J.-L., Bu, M., & Pittman, R. (2018). Does market competition dampen environmental performance? Evidence from China. Strategic Management Journal, 39(11), 3006–3030, doi:https://doi.org/10.1002/smj.2948.
Fryxell, G. E., & Lo, C. W. H. (2001). Organizational Membership and Environmental Ethics: A Comparison of Managers in State-owned Firms, Collectives, Private Firms and Joint Ventures in China. World Development, 29(11), 1941–1956, doi:https://doi.org/10.1016/S0305-750X(01)00078-X.
Harrison, A., Meyer, M., Wang, P., Zhao, L., & Zhao, M. (2019). Can a Tiger Change Its Stripes? Reform of Chinese State-Owned Enterprises in the Penumbra of the State. National Bureau of Economic Research Working Paper Series, No. 25475, doi:10.3386/w25475.
Hart, S. L. (1995). A Natural-Resource-Based View of the Firm. Academy of management Review, 20(4), 986–1014, doi:10.5465/amr.1995.9512280033.
Hsieh, C.-T., & Klenow, P. J. (2009). Misallocation and Manufacturing TFP in China and India*. The quarterly journal of economics, 124(4), 1403–1448, doi:10.1162/qjec.2009.124.4.1403.
Hsieh, C.-T., & Song, Z. (2015). Grasp the Large, Let Go of the Small: The Transformation of the State Sector in China. National Bureau of Economic Research Working Paper Series, No. 21006, doi:10.3386/w21006.
Jefferson, G. H., & Su, J. (2006). Privatization and restructuring in China: Evidence from shareholding ownership, 1995–2001. Journal of Comparative Economics, 34(1), 146–166, doi:https://doi.org/10.1016/j.jce.2005.11.008.
Jensen, M. (2001). Value Maximisation, Stakeholder Theory, and the Corporate Objective Function. European Financial Management, 7(3), 297–317, doi:https://doi.org/10.1111/1468-036X.00158.
Jiang, L., Lin, C., & Lin, P. (2014). The determinants of pollution levels: Firm-level evidence from Chinese manufacturing. Journal of Comparative Economics, 42(1), 118–142, doi:https://doi.org/10.1016/j.jce.2013.07.007.
Kang, J. (2013). The relationship between corporate diversification and corporate social performance. [https://doi.org/10.1002/smj.2005]. Strategic Management Journal, 34(1), 94–109, doi:https://doi.org/10.1002/smj.2005.
Kang, Y.-S., & Kim, B.-Y. (2012). Ownership structure and firm performance: Evidence from the Chinese corporate reform. China Economic Review, 23(2), 471–481, doi:https://doi.org/10.1016/j.chieco.2012.03.006.
Kassinis, G., & Vafeas, N. (2006). Stakeholder Pressures And Environmental Performance. Academy of Management journal, 49(1), 145–159, doi:10.5465/amj.2006.20785799.
Kikeri, S., Nellis, J. R., & Shirley, M. M. (1992). Privatization: the lessons of experience: World Bank Publications.
Li, X., & Chan, C. G.-W. (2016). Who Pollutes? Ownership type and environmental performance of Chinese firms. Journal of Contemporary China, 25(98), 248–263, doi:10.1080/10670564.2015.1075718.
Liang, J., & Langbein, L. (2021). Are State-Owned Enterprises Good Citizens in Environmental Governance? Evidence From the Control of Air Pollution in China. Administration & Society, 53(8), 1263–1292, doi:10.1177/00953997211005833.
Liang, J., & Ma, L. (2020). Ownership, Affiliation, and Organizational Performance: Evidence from China’s Results-Oriented Energy Policy. International Public Management Journal, 23(1), 57–83, doi:10.1080/10967494.2019.1595796.
Lu, J., Tao, Z., & Yang, Z. (2010). The costs and benefits of government control: Evidence from China's collectively-owned enterprises. China Economic Review, 21(2), 282–292, doi:https://doi.org/10.1016/j.chieco.2009.08.004.
Lyon, T. P., & Maxwell, J. W. (2008). Corporate Social Responsibility and the Environment: A Theoretical Perspective. Review of environmental economics and policy, 2(2), 240–260, doi:10.1093/reep/ren004.
Maung, M., Wilson, C., & Tang, X. (2016). Political Connections and Industrial Pollution: Evidence Based on State Ownership and Environmental Levies in China. Journal of Business Ethics, 138(4), 649–659, doi:10.1007/s10551-015-2771-5.
Miles, M. P., & Covin, J. G. (2000). Environmental Marketing: A Source of Reputational, Competitive, and Financial Advantage. Journal of Business Ethics, 23(3), 299–311, doi:10.1023/A:1006214509281.
Muhammad, N., Scrimgeour, F., Reddy, K., & Abidin, S. (2015). The Impact of Corporate Environmental Performance on Market Risk: The Australian Industry Case. Journal of Business Ethics, 132(2), 347–362, doi:10.1007/s10551-014-2324-3.
Mykhayliv, D., & Zauner, K. G. (2013). Investment behavior and ownership structures in Ukraine: Soft budget constraints, government ownership and private benefits of control. Journal of Comparative Economics, 41(1), 265–278, doi:https://doi.org/10.1016/j.jce.2012.03.001.
Omran, M. (2009). Post-privatization corporate governance and firm performance: The role of private ownership concentration, identity and board composition. Journal of Comparative Economics, 37(4), 658–673, doi:https://doi.org/10.1016/j.jce.2009.02.002.
Pargal, S., & Wheeler, D. (1996). Informal Regulation of Industrial Pollution in Developing Countries: Evidence from Indonesia. Journal of political economy, 104(6), 1314–1327, doi:10.1086/262061.
Patten, D. M. (2002). The relation between environmental performance and environmental disclosure: a research note. Accounting, Organizations and Society, 27(8), 763–773, doi:https://doi.org/10.1016/S0361-3682(02)00028-4.
Porter, M. E., & van der Linde, C. (1995). Toward a New Conception of the Environment-Competitiveness Relationship. Journal of Economic Perspectives, 9(4), 97–118, doi:10.1257/jep.9.4.97.
Rowe, A. L., & Guthrie, J. (2010). The Chinese Government's Formal Institutional Influence On Corporate Environmental Management. Public Management Review, 12(4), 511–529, doi:10.1080/14719037.2010.496265.
Sánchez-Medina, P. S., Díaz-Pichardo, R., Bautista-Cruz, A., & Toledo-López, A. (2015). Environmental Compliance and Economic and Environmental Performance: Evidence from Handicrafts Small Businesses in Mexico. Journal of Business Ethics, 126(3), 381–393, doi:10.1007/s10551-013-1945-2.
Schmid, A., & Rubin, D. K. (1995). Brazil emerges as Latin America's new environmental hotspot. (cover story). ENR: Engineering News-Record, 234(21), 42–42.
Shrivastava, P. (1995). Environmental technologies and competitive advantage. Strategic Management Journal, 16(S1), 183–200, doi:https://doi.org/10.1002/smj.4250160923.
Sprenger, C. (2011). The choice of ownership structure: Evidence from Russian mass privatization. Journal of Comparative Economics, 39(2), 260–277, doi:https://doi.org/10.1016/j.jce.2010.09.003.
Sukhtankar, S. (2016). Does firm ownership structure matter? Evidence from sugar mills in India. Journal of Development Economics, 122, 46–62, doi:https://doi.org/10.1016/j.jdeveco.2016.04.002.
Talukdar, D., & Meisner, C. M. (2001). Does the Private Sector Help or Hurt the Environment? Evidence from Carbon Dioxide Pollution in Developing Countries. World Development, 29(5), 827–840, doi:https://doi.org/10.1016/S0305-750X(01)00008-0.
Tihanyi, L., Aguilera, R. V., Heugens, P., Essen, M. v., Sauerwald, S., Duran, P., et al. (2019). State Ownership and Political Connections. Journal of Management, 45(6), 2293–2321, doi:10.1177/0149206318822113.
Walker, K., & Wan, F. (2012). The Harm of Symbolic Actions and Green-Washing: Corporate Actions and Communications on Environmental Performance and Their Financial Implications. Journal of Business Ethics, 109(2), 227–242, doi:10.1007/s10551-011-1122-4.
Wang, H. (2000). Pollution charges, community pressure, and abatement cost of industrial pollution in China (Vol. 2337): World Bank Publications.
Wang, H., & Jin, Y. (2007). Industrial Ownership and Environmental Performance: Evidence from China. Environmental and Resource Economics, 36(3), 255–273, doi:10.1007/s10640-006-9027-x.
Wang, H., Mamingi, N., Laplante, B., & Dasgupta, S. (2003). Incomplete Enforcement of Pollution Regulation: Bargaining Power of Chinese Factories. Environmental and Resource Economics, 24(3), 245–262, doi:10.1023/A:1022936506398.
Wong, S. C. Y. (2004). Improving Corporate Governance in Soes: An Integrated Approach. Corporate Governance International, 7(2).
Yao, S., & Yang, J. (2017). Geographical Distance and Environmental Information Disclosure: The Perspective of Public Pressure Transmission Efficiency. Asia-Pacific Journal of Financial Studies, 46(3), 445–462, doi:https://doi.org/10.1111/ajfs.12176.
Yu, M. (2015). Processing Trade, Tariff Reductions and Firm Productivity: Evidence from Chinese Firms. The Economic Journal, 125(585), 943–988, doi:https://doi.org/10.1111/ecoj.12127.
Zeng, S. X., Xu, X. D., Yin, H. T., & Tam, C. M. (2012). Factors that Drive Chinese Listed Companies in Voluntary Disclosure of Environmental Information. Journal of Business Ethics, 109(3), 309–321, doi:10.1007/s10551-011-1129-x.
Zhang, C. (2017). Political connections and corporate environmental responsibility: Adopting or escaping? Energy Economics, 68, 539–547, doi:https://doi.org/10.1016/j.eneco.2017.10.036.
Zhang, L., Ren, S., Chen, X., Li, D., & Yin, D. (2020). CEO Hubris and Firm Pollution: State and Market Contingencies in a Transitional Economy. Journal of Business Ethics, 161(2), 459–478, doi:10.1007/s10551-018-3987-y.

No competing interests reported.

Do state-owned enterprises really have better environmental performance in China? Environmental regulation and corporate environmental strategies

Status:

Version 1

Abstract

Figures

1. Introduction

2. Hypotheses Development

3. Methodology And Data

3.1 Variables and Measures

3.2 Data Sources

3.3 Summary Statistics

4. Econometric Models

4.1 Differences Between Ownership Types

4.2 Impact of Privatization

5. Empirical Findings

5.1 The Existence of SOEs’ Environmental Responsiveness Strategies

5.2 The Impacts of Privatization

6. Conclusions

Notes

Statements And Declarations

References

Additional Declarations

Status:

Version 1