The Impact of Corporate Social Responsibility on Sustainable Innovation: A Case in China's Heavy Pollution Industry.

Exploring the impact of corporate social responsibility (CSR) fulfillment and disclosure on enterprises' sustainable innovation capacity can not only expand the research boundary of factors of sustainable innovation and the impact of CSR, but it can also serve as a reference for the decision-making of listed companies in increasing pollution problems. Using a sample of 224 Chinese A-share businesses in the heavy pollution industry listed between 2016 and 2020 and employing an ordinary least square regression, the results provide empirical evidence that CSR is positively associated with sustainable innovation. Second, the business environment can serve as a moderator of the relationship between CSR and sustainable innovation, and the positive relationship between CSR and sustainable innovation is more pronounced in regions with better macroeconomic conditions. Additionally, the improvement of CSR for sustainable innovation is more clear in state-owned firms than in non-state-owned enterprises. After a series of robustness tests that eliminate marketization, law enforcement, and macro-political unpredictability, the results still hold. This study broadens the scope of CSR and sustainable innovation research. In addition, the theoretical and practical significance of this study's findings is referenced in this paper.

Introduction

Environmental and climate concerns caused by the intensification of global industrialization are irreversible. Corporate social responsibility (CSR) is a successful approach that encourages businesses to take on additional duties to support social and sustainable development, given the growing consensus in modern global business on the value of sustainable development (Bauman and Skitka, 2012). Innovation is one of the primary drivers of boosting sustainability development (Silvestre and Neto, 2014) and positively impact green performance (Sharma et al., 2021). Scholars pay close attention to the expanding literature on CSR and sustainable innovation (Shakeel et al., 2020).

Economy (Al-Hadi et al., 2019), legislation (Lau et al., 2018), morality (Mikulka et al., 2020), society, and the environment should be firms’ primary responsibilities (Amor-Esteban et al., 2019). It highlights that CSR is not only to fully consider stakeholders and execute the comprehensive social contract but also to increase economic performance (Cho et al., 2019) while assuming environmental obligations. In recent years, there have been many forms of studies on CSR’s influence and influence elements. National culture and corporate governance (Garcia-Sanchez et al., 2016; Mohamed Adnan et al., 2018), social media (Grygiel and Brown, 2019), ethics (Galvão et al., 2019; Smith et al., 2021), corporate reputation, and customer satisfaction (Li et al., 2019; Yu and Liang, 2020; Bogdan et al., 2021), corporate integrity culture (Wan et al., 2020; Khan et al., 2022). Sustainability strategy, performance, stakeholders, developing nations, climate change, and supply chain management are the research keywords for CSR (Ye et al., 2020). An innovative and sustainable organization respects the environment’s capacity to support and protect its ecosystem’s resources while pursuing economic efficiency (Severo et al., 2017). The research model incorporated CSR as the direction of future development, and sustainable innovation aimed at the environment can boost the economic and environmental performance of businesses (Ahmad et al., 2021).

As a fundamental aspect of sustainable development, the importance of sustainable innovation is self-evident (Silvestre and Ţîrcă, 2019). Prior research shows the connection between people, businesses, and creativity. Environmental sustainability has actively supported the innovation of sustainable business practices (Shahzad et al., 2020). Higher enterprise innovation success is associated with greater CSR (Wu et al., 2018). In recent years, however, there have been few studies on the relationship between sustainable innovation and CSR. Based on social capital and stakeholder theories, CSR impacts innovation performance in an inverted U-shape. The direction of CSR’s influence on innovation may vary among industries (Liu et al., 2021). The association between sustainable innovation and CSR, therefore, requires additional study. Additionally, from an industry perspective, businesses in different industries require different fundamental resources for innovation; therefore, CSR data must be distinguished according to the industry in the study. Existing literature focuses mostly on the fashion sector (Arrigo, 2013), the semiconductor business (Lu et al., 2013), the banking industry (Istianingsih et al., 2020), and the energy industry (Arrigo, 2013; Lu et al., 2013, 2019; Istianingsih et al., 2020).

It is common knowledge that significantly polluting companies have a far greater detrimental influence on the ecological environment than others do (Xie et al., 2022). Because of China’s large consumption of traditional fossil fuels, the country’s environmental quality has worsened significantly, attracting the international community’s utmost concern (Dong et al., 2021). China is the largest manufacturing nation in the world. According to the China Statistical Yearbook 2021, 68% of total energy consumption is attributable to enterprises with elevated levels of pollution. To advance the aims of “carbon peak” and “carbon neutral,” more studies must be undertaken on industries with elevated levels of pollution. Since the reform and opening of China 40 years ago, China’s industry has boomed. The prior development, however, was overly reliant on energy and resource input and production scale expansion. China’s industrial expansion followed a broad pattern of growth, inflicting serious environmental and ecosystem harm (Zhang J. et al., 2017). Traditional industries with high energy consumption, high emissions, and high pollutants will have a considerable influence on the environment. Innovation is one of the driving forces for China’s sustainable industrial development (Yuan and Zhang, 2020). In recent years, China’s environmental rules and regulations have been increasingly stringent, and the sustainable development of China’s heavy pollution sectors has steadily become dependent on green development that considers innovation and environmental considerations. However, Fang et al. (2019) discovered in their research that heavy pollution sectors face the conundrum of “effective but not environmentally friendly innovation.” Consequently, it is vital to investigate further the performance of China’s heavy polluting sectors in terms of sustainable development.

This article picks Chinese A-share listed firms from 2016 to 2020 as its research object and empirically examines the impact of CSR on enterprises’ sustainable innovation capacity. The findings show that the output of green innovation considerably enhances business sustainability. Given the sustainable development of businesses, the following questions are posed in this study. How can CSR foster innovation and sustainability? Does the influence of CSR on the capacity for sustainable innovation vary by corporate environment? What is the state of CSR in the Chinese heavy polluting industry?

The following are the primary contributions of this work. First, it expands the literature on sustainable innovation and CSR, which contributes to the development of a fresh perspective for the study of the factors influencing the sustainable innovation capacity of businesses. Existing studies have investigated more CSR-influencing aspects, however, there remains a dearth of studies on CSR’s role. As opposed to undertaking a standard analysis at the firm or national level, this study focuses on publicly traded enterprises in China’s heavy pollution industry. This study can therefore serve as a substantial contribution to the research on the sustainable development of the heavy pollution sector and give theoretical support for the heavy pollution industry to realize its low-carbon transformation goals. Third, most previous research has ignored the societal dimensions of CSR in general (Chen and Wan, 2020). In this study, the business environment is included as a moderating variable in the research model to investigate the impact of macroeconomic conditions on the relationship between sustainable innovation and CSR. The data passed the test for robustness. Thus, our findings may be useful to policymakers by identifying social normative force and illuminating how it drives businesses. Given that CSR has a significant impact on the interests of stakeholders, this study can also assist stakeholders in making more informed judgments about the sustainable innovation of businesses.

The organization of this investigation is as follows: The literature review and research hypotheses are provided in Section “Literature Reviews and Research Hypotheses”. The third section of this study describes the research design, including the variables, sample, and model selection. The section “Robustness Tests” consists of empirical analysis, findings reporting, and comments. Section “Conclusions and Policy Recommendations” highlights the theoretical and practical implications’ conclusions and policy recommendations.

Literature Reviews and Research Hypotheses

Sustainable Innovation and CSR

According to the stakeholder theory, (CSR) entails that the development of businesses should include stakeholders, including employees, consumers, suppliers, and communities (Turker, 2009). By adhering to principles of CSR, businesses can foster confidence and excellent connections with internal and external stakeholders and effectively drive innovation (Lins et al., 2016).

Everything related to CSR can have a favorable effect on shareholder profitability (Pucheta-Martínez and Gallego-Álvarez, 2021). CSR provides shareholders with economic profits, management and operational knowledge, and motivation to work on CSR. Shareholder-related CSR can increase shareholder confidence in innovative investment opportunities (Iyer and Soberman, 2016). Employee-focused CSR can facilitate employee identification with the organization. When employees acknowledge a firm’s commitment to environmental sustainability, they encourage the organization to regard environmental preservation as a competitive advantage-enhancing opportunity (Ernst and Jensen Schleiter, 2021). Enterprises boost social and environmental performance through pro-environment behavior and stimulate employees’ green behaviors, which has a favorable effect on employees’ innovative technology exploration (Xu et al., 2022). Green human resource management may promote the sustainability of enterprises as an essential technique for influencing the green behavior of employees (Amjad et al., 2021; Zhu et al., 2021). Employee green creativity is regarded as the driving force behind company green innovation, and employee green behavior is a crucial metric for measuring employee green creativity (Jiang et al., 2020). Gaudencio et al. (2017) found that CSR increases employee job satisfaction and organizational commitment and has a beneficial effect on the establishment of a stable innovation team (Ho, 2017). CSR receives greater attention the more optimistic the customer’s attitude. Customers like to buy products that perform well in terms of social responsibility (Iyer and Soberman, 2016). Because of client desires, businesses produce added items through technological innovation. In addition, CSR can influence the behavior and selection of suppliers (Kumar et al., 2014; Zhang M. et al., 2017; Govindan et al., 2018). Companies in a supply chain that apply CSR-related practices can enhance not just their performance, but also that of their supply chain partners (Yang et al., 2020). Businesses may develop societal trust and a positive public image by engaging in CSR. Because of these factors, businesses can foster economic performance and innovative conduct.

Many researchers have conducted studies on sustainable innovation. Sustainable innovation is described as an innovation model with sustainable innovation goals in the creative development process (Cagliano and Behnam, 2019). It exemplifies innovation that is advantageous for environmental quality improvement and social collaboration (Zhang et al., 2022). The enterprises’ green innovation behavior can be considered the performance of sustainable innovation. Important to sustainable development, green innovation promotes innovative technology and concepts (Liao et al., 2022). In addition to ensuring efficient resource usage and effective pollution reduction, the competitive advantage of green innovation rests in achieving optimal economic performance (Fernando et al., 2019). Studies have shown that CSR can assist stakeholders in increasing their profitability and further promote green investment and pro-environment behavior, which is reflected in sustainable innovation’s success. Consequently, we suggest our initial hypothesis:

H1: The output of CSR can significantly enhance the corporate sustainable innovation performance.

The Moderating Role of the Business Environment

Environment and resources limit the development and operation of heavy pollution industries, which are specialized sectors. In other words, high pollution businesses operate in an environment that is dynamic and constantly changing. Instead of operating in a vacuum, organizations are formed by their surroundings (Harrison and Pelletier, 1998). The environment of an organization is its means of survival. Considered one of the aspects determining the sustainable performance of a corporation is the business environment (Alqudah et al., 2021). The optimization of the business environment can foster technological innovation and enhance the product quality and technological level of businesses. In addition, the optimization of the market environment facilitates firm entry and enhances market rivalry, hence interesting incumbent enterprises to do technological research and development. The development of environmental technologies can foster sustainable innovation in industries with high pollution levels.

Based on the preceding study, the following is the second hypothesis:

H2: The promotion effect of the output of CSR on sustainable innovation performance is more significant when the business environment is poor.

Materials and Methods

Sample Selection and Data Sources

This study focuses on the Chinese A-share firms involved in severe pollution industries from 2016 to 2020. Two thousand sixteen is the most recent year for which we have comprehensive data. Given that some data are unavailable at the time of this study, 2020 has been chosen as the conclusion date. The scope of sample selection refers to the CSMAR database and the classification standards of heavily polluting industries in The Guidelines for Environmental Information Disclosure of Listed Companies by the Chinese Ministry of Environmental Protection. To assure the validity of the empirical research, the sample is treated as follows. First, to avoid the influence of outliers, firms with anomalous financial status, ST, *ST, suspended listing, and delisting between 2016 and 2020 were omitted from this study. Second, we eliminate samples devoid of CSR and other variable values. Third, to prevent the influence of extreme values, we eliminate the samples from 2016-to 2020 for which the value of sustainable innovation is zero. The CSR statistics are from the social responsibility reports of HeXun Net-listed enterprises. The data on sustainable innovation comes from the National Intellectual Property Patent Database and the Green List of International Patent Classification maintained by the World Intellectual Property Organization (WIPO). Other data sources include the China Statistical Yearbook, the China Environmental Statistical Yearbook, the annual reports of publicly traded enterprises, and the RESSET database.

Variables

Dependent Variable (Sustainable Innovation)

According to the current body of research, there are no accepted criteria for measuring the sustainable innovation of businesses. The patent data of an enterprise directly reflects its technological innovation accomplishments, and the number of patents can be used to gauge an enterprise’s innovation level (Abraham and Moitra, 2001; Albino et al., 2014). This study selects patent applications for green inventions and green utility models as indicators of sustainable innovation.

Independent Variable (CSR)

The social responsibility assessment system of HeXun Net comprises fifty subdivision indicators. The entire system is based on shareholder responsibility, employee accountability, supplier, customer, consumer responsibility, environmental responsibility, and social responsibility. The findings represent CSR compliance and transparency.

Moderating Variable (Business Environment)

The business environment is selected as the moderating variable in this study. The concept of conducting business is derived from the World Bank’s Doing Business Report. The World Bank evaluates the business climate from a national and regional standpoint. This study requires more granular indicators for provincial regions. Consequently, this article utilizes the research on the evaluation index system of the business environment (Yang and Wei, 2021) to objectively calculate the business environment index of the city where the firms are located. This index system comprises per capita GDP, average salary level, consumption rate, per capita fixed asset investment, and GDP growth rate as indicators and takes into consideration disparities in economic development level and human capital from the standpoint of the macroeconomic environment.

Control Variables

Drawing on the previous empirical research on CSR (Ali and Frynas, 2017; Su, 2019; Chen and Wan, 2020; Wan et al., 2020), this study also selects control variables as follows: the size of the company (SIZE), price-to-book ratio (PB), profitability (LEV), return on total assets (ROA), years of establishment (AGE), cash flow (CASH), shareholding nature (SOE), managerial shareholding ratio (MSH), board independence (INDEP), and duality (DUAL).

The definitions and interpretations of all variables are shown in Table 1.

Table 1

Variable definitions.

Variable name	Symbols	Variable definition
Sustainable innovation	SI	The natural logarithm of one plus the sum of green invention patent applications and green utility model patent applications in the previous year
Corporate social responsibility	CSR	The score of corporate social responsibility of a company
Business environment	ENVIR	The macroeconomic environment index of the city where the enterprise is located
The size of the company	SIZE	The natural logarithm of the company’s total assets
Price-to-Book ratio	PB	The ratio of prices per share to net assets per share
Leverage	LEV	The ratio of a company’s total liability to total assets
Profitability	ROA	The ratio of earnings before interest and tax to total assets
Listed years	AGE	The natural logarithm of one plus the number of listed years of the company
Cash flow	CASH	The ratio of net cash flow of operating activities to total assets
Property rights	SOE	A virtual variable. If a company is owned by the government, the variable is one; otherwise, is zero
Managerial ownership	MSH	The ratio of managers’ shares to total shares
Board independence	INDEP	The proportion of independent directors in all directors
Managers’ duality	DUAL	A virtual variable. If the company’s CEO also serves as the chairperson, the variable is one; otherwise, is zero

Empirical Models

Based on the previous studies (Chen and Wan, 2020; Chen and Ji, 2022; Liao et al., 2022; Xie et al., 2022), this study establishes Equation 1 and use the OLS regression method to investigate the impact of CSR on the sustainable innovation.

Results and Discussions

Descriptive Statistics

The variables’ descriptive statistics for the entire sample are presented in Table 2. As shown in the table, the mean and median values are 1.887 and 1.0986, respectively, whereas the 25% levels and maximum CSR are 0.0000 and 8.9200, showing that there are considerable disparities in SI performance among the studied organizations. The mean and median CSR values are 24.7164 and 19.8650, respectively, while the minimum, 25%, 75%, and maximum CSR values are −11.7700, 15.3025, 26.4050, and 85.7700, indicating that the sampled organizations perform poorly on CSR. Both the capacity for sustainable innovation and performing CSR among the samples have a significant space for development. The 25% and 75% thresholds of ENVIR are 0.3833 and 0.5445, respectively, showing that the macroeconomic contexts in which the studied enterprises operate are distinct. The minimum and maximum CASH values are −7.7700 and 17.5900, respectively, indicating that there are significant variances in operational capability among the examined organizations. The mean for independent SOEs is 0.4375, meaning that 43.75% of the studied enterprises are government-owned. Moreover, there are significant variances in many sample parameters, such as SIZE, PB, LEV, ROA, and MSH, necessitating the inclusion of these control variables in this model. Table 2 additionally provides descriptive analysis results for other variables.

Table 2

Descriptive statistics.

Variables	n	Mean	Std.	Min.	P25	Median	P75	Max
SI	1,120	1.1887	1.3357	0.0000	0.0000	1.0986	1.7918	8.9200
CSR	1,120	24.1764	16.9059	−11.7700	15.3025	19.8650	26.4050	85.7700
ENVIR	1,120	0.4610	0.1120	0.2200	0.3833	0.4510	0.5445	0.7500
SIZE	1,120	22.6404	1.6479	18.7600	21.3852	22.4373	23.7027	28.6400
PB	1,120	2.9944	4.8248	0.0000	1.3146	2.0409	3.2542	78.3400
LEV	1,120	0.4721	0.2007	0.0100	0.3159	0.4734	0.6078	2.2900
ROA	1,120	0.0441	0.1453	−3.1500	0.0143	0.0409	0.0826	1.1300
AGE	1,120	3.0943	0.2124	2.0800	2.9957	3.0910	3.2189	3.7600
CASH	1,120	0.0455	0.6434	−7.7700	−0.0168	0.0072	0.0452	17.5900
SOE	1,120	0.4375	0.4963	0.0000	0.0000	0.0000	1.0000	1.0000
MSH	1,120	0.0965	0.1608	0.0000	0.0000	0.0020	0.1364	0.9600
INDEP	1,120	0.3743	0.0553	0.2300	0.3333	0.3333	0.4286	0.6700
DUAL	1,120	0.2357	0.4246	0.0000	0.0000	0.0000	0.0000	1.0000

Correlation Analysis

The Pearson correlation coefficients between the variables are displayed in Table 3. Consistent with hypothesis H1, the correlation study demonstrates that SI is significantly consistent with CSR at the 1% level, providing early evidence that corporate integrity culture is favorably associated with a firm’s CSR performance.

Table 3

Correlation analysis of variables.

Variables	SI	CSR	ENVIR	SIZE	PB	LEV	ROA	AGE	CASH	SOE	MSH	INDEP	DUAL
SI	1	–
CSR	0.256***	1	–
ENVIR	0.204***	0.124***	1	–
SIZE	0.198***	0.094***	−0.026	1	–
PB	−0.051*	−0.092***	−0.039	−0.167***	1	–
LEV	0.139***	−0.072**	−0.095***	0.374***	0.154***	1	–
ROA	0.041	0.117***	0.036	−0.041	0.002	−0.275***	1	–
AGE	0.152***	0.142***	0.014	−0.012	−0.007	0.145***	−0.046	1	–
CASH	0.064**	0.103***	0.040	−0.068**	0.026	−0.010	0.123***	0.028	1	–
SOE	0.281***	0.139***	−0.083***	0.454***	−0.020	0.417***	−0.100***	0.204***	0.015	1	–
MSH	−0.160***	−0.088***	0.095***	−0.339***	0.015	−0.286***	0.087***	−0.164***	−0.008	−0.522***	1	–
INDEP	0.120***	0.013	0.034	0.085***	−0.017	0.039	−0.010	−0.077**	−0.031	0.082***	−0.053*	1	–
DUAL	−0.056*	−0.030	0.068**	−0.200***	0.024	−0.222***	0.004	−0.032	0.005	−0.257***	0.180***	0.060**	1

***, **, * represents the level of significance at 1, 5, and 10%, respectively.

In general, if the correlation coefficients between independent variables are less than or equal to 0.80, the model may not have significant multicollinearity issues. All correlation coefficients between independent variables in this model are less than 0.454. There is hence no multicollinearity issue. The findings of the univariate correlation analysis are shown above, and the results of the multivariate regression analysis will be presented below.

Multivariate Regression Results

Table 4 displays the findings of a multivariate regression on the effect of CSR e on sustainable innovation. Although the modified R2 (0.147) is insufficient, the F-value shows that the models as a whole are significant (18.506). Model (1)'s regression output comprises independent variables and control variables. CSR has a significantly positive impact on sustainable innovation (0.016, t = 7.018), as shown in the table. This positive correlation implies that hypothesis H1 proposed in this paper’s research hypothesis section has been confirmed by the empirical study. The findings suggest that organizations that prioritize CSR fulfillment and disclosure have a greater capability for sustainable innovation. Since the existing research has identified the influencing factors of CSR, we choose several representative corporate management variables as control variables, and the regression results of the control variables in Model (1) are most consistent with expectations. Among the control variables, the SIZE, AGE, CASH, SOE, and INDEP regression coefficients are significantly positive. A greater number of independent directors, a larger asset size, a longer listing period, more asset liquidity, and more stable equity are all correlated with a higher ability for sustainable innovation. Significantly negative regression coefficients for PB can be observed. The lower the price-to-book ratio, the greater the company’s investment value and growth prospects, and hence its emphasis on sustainable innovation. Several research has previously investigated and proven the inherent positive impact of CSR, innovation, and sustainable development (Silvestre and Ţîrcă, 2019; Sharma et al., 2021; Chen and Ji, 2022; Liao et al., 2022). This study’s findings are consistent with past research in this area. In addition, the data confirm the likelihood that CSR in various industries may have varied effects on sustainable innovation at various times (Liu et al., 2021). This may owe to the various key resources utilized by various sectors. In the context of China’s carbon peak and carbon neutrality objectives, firms in the heavy pollution industry that place a premium on CSR will prioritize their sustainable development and guide stakeholders to engage in sustainable innovation.

Table 4

Regression results of sustainable innovation and CSR.

Variables	Model (1)	Model (2)
CSR	0.016***	0.18***
	(7.018)	(6.422)
ENVIR	−	0.181***
		(6.616)
CSR*ENVIR	−	0.126***
		(5.245)
SIZE	0.059**	0.067**
	(2.177)	(2.058)
PB	−0.007	−0.026
	(−0.874)	(−0.926)
LEV	0.395*	0.069**
	(1.742)	(2.104)
ROA	0.494*	0.05*
	(1.837)	(1.783)
AGE	0.562***	0.083***
	(3.075)	(2.965)
CASH	0.084	0.03
	(1.439)	(1.12)
SOE	0.49***	0.186***
	(4.975)	(5.263)
MSH	0.061	0.006
	(0.225)	(0.185)
INDEP	2.437***	0.096***
	(3.602)	(3.556)
DUAL	0.065	0.013
	(0.708)	(0.457)
Constant	−3.624***	−0.016
	(−3.962)	(−0.58)
Year Effect	Control	Control
Industry Effect	Control	Control
n	1,120	1,120
adj. R2	0.147	0.203
F-value	18.506***	22.965***

(1) T-values are reported in parentheses. (2) ***, **, * represents the level of significance at 1, 5, and 10%, respectively.

The H2 hypothesis investigates the effect of the business environment on the relationship between sustainable innovation and CSR. Before assessing the business environment moderating, we standardize the data. Table 4 displays the regression findings for Model (2). We derived varying business climate scores based on the location of the businesses. The business environment has a considerable impact on the interaction between sustainable innovation and CSR. The enhanced R2 of 0.203 indicates that the model has a better fitting effect. This positive correlation demonstrates that the empirical investigation has confirmed the hypothesis H2 proposed in the research hypothesis section of this work. Objectively, the business environment plays a moderating role. On the one hand, when businesses perform well in terms of CSR, a better macroeconomic climate can bring about greater investment possibilities and human capital to encourage the development of sustainable innovation capability. In contrast, when a business is in a location with a more favorable economic climate, market rivalry and government laws will encourage the business to adhere to CSR and prioritize sustainable development. When businesses are in regions with more favorable economic conditions, they are more likely to have easier access to capital, hence bolstering budgets for sustainable innovation. Consequently, the ability for sustainable innovation may increase. In addition, the coefficients and significance of other control variables in this model are consistent with expectations.

Robustness Tests

Controlling the Effects of Marketization and Law Enforcement

In addition, we control for the effects of marketization and law enforcement, both of which may impact the CSR of local firms. For instance, Du et al. (2016) observed that the amount of law enforcement in an area has a considerable impact on the CSR performance of local businesses and that the enforcement of regulations varies greatly throughout Chinese provinces. Based on a prior study (Wang et al., 2008; Chen and Wan, 2020), one should additionally evaluate a region’s marketization. We use the regional marketization index and the legal environment index in conjunction with prior research (Wan et al., 2020) to assess the marketization process and regional law enforcement in China (Fan et al., 2011). Table 5 displays the outcomes. We add the control variable MARKET to the regression model in column (1). In column (2), the control variable LAW is incorporated into the regression model. In column (3), both MARKET and LAW are included as control variables in the regression model. The results of the three models are comparable, showing that the influence of CSR on sustainable innovation remains positively significant. Therefore, the localization of the market for law enforcement has no bearing on our argument regarding the relationship between CSR and sustainable innovation. The empirical findings remain valid.

Table 5

Controlling for the effects of marketization and law enforcement.

Variables	Marketization (1)	Law enforcement (2)	Both (3)
CSR	0.016***	0.015***	0.016***
	(6.834)	(6.758)***	(6.955)
MARKET	0.029***	−	−0.156***
	(1.389)		(−3.04)
LAW	−	0.02**	0.069***
		(2.867)	(3.946)
SIZE	0.057**	0.051*	0.047*
	(2.069)	(1.875)	(1.726)
PB	−0.007	−0.008	−0.007
	(−0.925)	(−0.963)	(−0.894)
LEV	0.442*	0.479**	0.428*
	(1.928)	(2.1)	(1.876)
ROA	0.481*	0.477*	0.508*
	(1.786)	(1.778)	(1.899)
AGE	0.537***	0.554***	0.67***
	(2.922)	(3.037)	(3.61)
CASH	0.083	0.085	0.09
	(1.426)	(1.454)	(1.558)
SOE	0.534***	0.556***	0.479***
	(5.163)	(5.515)	(4.624)
MSH	0.037	0.032	0.091
	(0.136)	(0.117)	(0.336)
INDEP	2.407***	2.373***	2.383***
	(3.556)	(3.516)	(3.544)
DUAL	0.053	0.039	0.043
	(0.574)	(0.428)	(0.473)
Constant	−3.756***	−3.661***	−3.04***
	(−4.086)	(−4.015)	(−3.625)
Year Effect	Control	Control	Control
Industry Effect	Control	Control	Control
n	1,120	1,120	1,120
adj. R2	0.148	0.152	0.159
F-value	17.139***	17.759***	17.226***

(1) T-values are reported in parentheses. (2) ***, **, * represents the level of significance at 1, 5, and 10%, respectively.

Exclusion of Alternative Explanations

Chen and Ji (2022) discovered that research results may only be confirmed during a period with a negative macro-political environment and fade during other eras. Therefore, it is essential to rule out this other explanation and re-evaluate our samples. In 2017, for instance, the 19th National Congress of the Chinese Communist Party was held. In addition, COVID-19 affected most of China in 2020. Both can be unpredictable macro-political environment elements. As a result, we choose the policy environment index (Yang and Wei, 2021) as a proxy to measure the macro-political environment of a region. To facilitate comparisons, we divide our sample into two groups based on whether macro-political environment uncertainty is high or low and recalculate the regression results. The value of the policy environment index that is below the mean shows macro-political environment uncertainty, whereas a value above the mean indicates political environment uncertainty. The results presented in Table 6 for samples of high and low political uncertainty are consistent with those presented in prior tables. The macro-political environment does not affect our outcomes.

Table 6

Exclusion for political uncertainty.

Variables	Higher (1)	Lower (2)
CSR	0.014***	0.019***
	(3.853)	(6.618)
SIZE	0.205***	−0.116***
	(5.041)	(−3.087)
PB	−0.003	−0.016
	(−0.229)	(−1.358)
LEV	−0.275	1.2***
	(−0.783)	(4.044)
ROA	0.01	1.74***
	(0.029)	(3.557)
AGE	0.75**	0.61***
	(2.128)	(2.818)
CASH	−0.082	0.049
	(−0.376)	(0.847)
SOE	0.325*	0.394***
	(1.818)	(3.13)
MSH	−0.467	0.081
	(−0.666)	(0.282)
INDEP	2.184**	2.492***
	(1.964)	(2.926)
DUAL	0.08	0.035
	(0.446)	(0.339)
Constant	−6.907***	−0.363
	(−4.491)	(−0.317)
Year Effect	Control	Control
Industry Effect	Control	Control
n	443	667
adj. R2	0.164	0.168
F-value	8.863***	13.379***

(1) T-values are reported in parentheses. (2) ***, **, * represents the level of significance at 1, 5, and 10%, respectively.

Conclusion and Policy Recommendations

Conclusion

In recent years, China has established the goals of carbon peak and carbon neutrality, as well as intensified its efforts to change heavy polluting industries to promote energy saving, emission reduction, and sustainable growth. This research investigates the relationship between sustainable innovation and the CSR of China’s big polluters that are publicly traded. Exploring the impact of CSR on the sustainable innovation capacity of enterprises can not only broaden the scope of research on the impact mechanism of sustainable enterprises’ development capacity and the effect consequences of CSR but also serve as a guide for the decision-making of publicly traded companies in the heavy pollution industry. Based on the data of China’s A-share heavy pollution listed companies from 2016 to 2020, we evaluated the effect of CSR fulfillment and disclosure on green patent applications. Through a series of robustness tests, the results are unaffected by marketization, law enforcement, and macro-political unpredictability.

The outcomes reveal: (1) CSR significantly improves the sustainable innovation capacity of businesses and (2) when a business is in a region with a more favorable macroeconomic environment, the effect of CSR on sustainable innovation capacity is more pronounced. Additionally, the improvement of CSR for sustainable innovation is more clear in state-owned firms than in non-state-owned enterprises. CSR has a more favorable effect on sustainable innovation when the board is more independent. These results also indicate that the government and independent directors can serve as a check, a balance, and a supervisor to encourage stakeholders to prioritize CSR and promote sustainable innovation capacity from the sidelines, particularly in China’s heavy pollution industry.

Policy Recommendations

This study’s findings have the following implications for businesses, their managers, and legislators. In the first place, our empirical findings demonstrate that CSR greatly improves sustainable innovation potential. Therefore, corporate managers must acknowledge the significance of CSR. They should place a greater emphasis on the outcomes of sustainable innovation and realize the sustainable development of businesses through sustainable innovation. Due to the stimulation of macroeconomic environmental conditions, firms in the heavy polluting industry will pay greater attention to the fulfillment and disclosure of CSR and support the strengthening of their capacity for sustainable innovation. Currently, heavy polluting industries, particularly manufacturing, are shifting from China’s economically developed eastern areas to the economically depressed center and western regions. If the government does not prioritize local economic growth, it may negatively affect the local ecology. Moreover, by addressing internal governance characteristics, businesses can mitigate the detrimental impact of the regional transfer on sustainable innovation. Thirdly, green human resource management methods can improve the environmental performance and sustainability of businesses (Roscoe et al., 2019; Bazrkar and Moshiripour, 2021). Incorporating sustainability measures into the human resource management system (Sabokro et al., 2021) and recognizing the role of human resource management for the achievement of long-term sustainability in industrial development are therefore options for heavy pollution industries. In conclusion, businesses should integrate green and sustainable practices into their overall development plan.

Limitations and Future Research Directions

This study has limitations that necessitate more investigation. Due to the availability of data, this report only includes information from 2016 to 2020. Due to China’s ongoing efforts in energy saving and emission reduction during the past few years, results may vary over time. In addition, this article examined the impact of CSR on sustainable innovation from the standpoint of CSR. In future, we can also examine the effects of the many components of CSR. Lastly, the proportion of highly educated employees and research and development professionals might be viewed as elements that influence the sustainable innovation capacity of businesses.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author Contributions

XL: investigation, data curation, modeling and experiment. RY: methodology, supervision, and writing. XZ: review and editing. All authors contributed to the article and approved the submitted version.

Funding

The work was supported by the National Natural Science Foundation of China (no. 71802021, no. 71602008), Beijing Natural Science Foundation (no. 9184023), Beijing Municipal Education Commission Foundation (BJSJ2020001, BJSJ2019001, BJSJ2018009), the Fundamental Research Funds for the Central Universities (no. FRF-BD-20-15A, no. FRF-BR-20-01C) and Beijing Philosophy and Social Science Planning Project (no. 21JCC089).

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.