Environmental, Social and Corporate Governance (ESG) and Total Factor Productivity: The Mediating Role of Financing Constraints and R&D Investment

Abstract

In recent years, “environment, society and governance” (ESG) has attracted widespread attention. As an investment philosophy focused on long-term value creation and non-financial performance indicators, ESG addresses internal governance challenges and fosters high-quality economic and social development. This study uses panel data analysis of 9125 observations from 1305 eligible companies to examine the relationship between ESG ratings, financing constraints, corporate research and development (R&D), and total factor productivity (TFP). It focuses on heavily polluting enterprises listed on the Shanghai and Shenzhen A-shares from 2012 to 2022. The findings show that (1) ESG ratings significantly impact TFP for the better, and (2) financial limitations act as a go-between for the ESG ratings and TFP connection, and (3) corporate R&D also serves as a mediator between ESG ratings and TFP. These findings offer valuable insights for shaping corporate ESG strategies, driving green transformation, enhancing productivity, advancing sustainable development, and supporting high-level environmental protection.

1. Introduction

As environmental pollution worsens and climate change becomes increasingly urgent, governments worldwide are implementing sustainable strategies to address these challenges [1]. The Chinese government has emphasized the importance of transforming the growth model and accelerating the green transformation. The key to this initiative is establishing an ESG rating system, which has become the core of corporate strategic planning and management. ESG performance is now widely recognized as a vital indicator of corporate sustainability and long-term value, driving extensive academic research into how various ESG components influence corporate productivity. With increasingly strained resources and limited technological capabilities, companies urgently need to improve innovation efficiency while accelerating the transition to a green development model [2].

There has been growing attention to sustainability and ESG activities in recent years. According to research by Gies et al., companies that perform well in the ESG domain tend to outperform their peers and have higher profits [3]. Many studies have shown that companies benefit their bottom line when doing well in ESG areas [4,5]. A higher ESG rating can increase access to preferential financing channels and investment opportunities [6,7]. Companies prioritizing ESG practices tend to be more transparent, improving their reputation and attracting favourable financing terms, ultimately boosting productivity [8]. In addition, a study by Wang et al. emphasizes the critical role of ESG concepts in attracting foreign direct investment when companies expand internationally [9]. Tan et al. found that ESG is critical to driving the green transformation of listed companies because it promotes sustainable development, mitigates financing constraints, and encourages environmentally friendly practices [10].

Total Factor Productivity (TFP) measures technological progress and overall efficiency as a critical measure of how well each manufacturing input works together [11]. “As a core indicator of the economy, total factor productivity (TFP) plays a key role in promoting sustainable development of enterprises” [12]. Literature confirms that TFP is critical to improving productivity and increasing enterprise value [13]. In addition, research has found that incorporating ESG factors into corporate investment strategies can effectively improve corporate productivity’s steady growth [14,15]. Therefore, ESG factors should be integrated into corporate investment strategies and business models to improve total factor productivity [16].

Prior studies have mainly used the perspectives of internal and external corporate governance to analyze the variables impacting total factor productivity (TFP). Notably, Financing constraints and research and development (R&D) are widely recognized as critical determinants of TFP [17,18]. Corporate research and development (R&D) is crucial to high-quality economic development [19]. R&D activities enhance enterprise productivity by fostering technological advancements, improving processes, and optimizing resource allocation [20,21]. As companies invest more in green and low-carbon technologies, their focus on sustainable innovation strengthens management practices and helps balance economic, ecological, and social benefits [22,23]. Literature indicates that financing constraints are significant obstacles for companies pursuing long-term growth [24]. Given that R&D activities require substantial financial support, these constraints directly impact enterprise productivity [25].

Despite existing research, the impact of ESG on TFP has yet to be adequately examined, particularly regarding the mechanisms related to financing constraints and corporate research and development (R&D). This study addresses this gap by integrating firm-level financing constraints and R&D into the analytical model to elucidate how ESG ratings influence TFP. This research seeks to enhance our comprehension of the impact of ESG practices on business productivity by examining the intricate link between ESG and TFP.

This research examines polluting firms listed on the Shenzhen and Shanghai stock markets between 2012 and 2022. The sample comprises 1305 eligible companies, yielding 9125 observations, with data from the CSMAR database. This research aims to clarify the specific pathways through which ESG impacts total factor productivity, ultimately providing valuable insights to enhance the sustainability of Chinese companies.

This paper’s contributions are summarized as follows: This study uniquely contributes to the field by comprehensively analyzing the direct and indirect impacts of ESG performance on total factor productivity (TFP). While previous studies often focus on individual ESG components, this research integrates ESG performance as a holistic factor, highlighting how ESG practices enhance corporate productivity by easing financing constraints and boosting research and development (R&D) investments. These findings broaden the academic discourse on ESG’s role in corporate sustainability and address the specific gap in how ESG practices impact productivity in highly polluting industries. This study presents a novel framework that uses the mediating effects of financing constraints and R&D investment to emphasize ESG performance as a critical driver of TFP. This framework reveals that improved ESG performance enables companies to reduce financing constraints and increase R&D investment, indirectly fostering productivity growth—an area not fully explored in existing literature, especially within China’s highly polluting enterprises. Analyzing data from 2012 to 2022 on high-pollution firms listed on Shanghai and Shenzhen A-shares, the study demonstrates that strong ESG performance significantly enhances TFP by alleviating financing limitations and increasing R&D. Based on these findings, it seems that businesses that score higher on ESG metrics have an easier time finding cheap capital and putting more emphasis on research and development. The study’s insights provide strategic guidance for both managers and policymakers. For companies, prioritizing ESG improvements can bolster productivity and secure long-term sustainability. For governments, policy incentives promoting ESG practices could enhance efficiency in heavily polluting sectors. Our research demonstrates that ESG performance fosters social and environmental responsibility and improves corporate production efficiency, highlighting the need to incorporate ESG issues into company operations.

The following is the outline of the paper: The second part presents the research hypotheses and a comprehensive literature review. The third portion introduces and explains the research model, variables, and methodology. The fourth part presents the empirical analysis’s findings. Final thoughts provide a synopsis of the results, some discussion of their theoretical and managerial significance, and some recommendations for further study.

2. Research Hypotheses

(1)

ESG ratings and total factor productivity

Total factor productivity (TFP) refers to the additional and often unpredictable productivity generated under given input conditions [26]. Few studies have examined how overall ESG performance relates to TFP; most have concentrated on specific ESG components and how they affect TFP [15,27]. The idea of high-quality development is gaining traction, and as a result, financial markets now assess businesses using ESG sustainability metrics in addition to traditional financial and economic metrics [14].

According to stakeholder theory, a company’s connections with its many stakeholder groups greatly impact its capacity to grow and develop [28]. As consumers of resources and energy, companies pursuing long-term goals naturally need to comply with ESG principles [29]. As “social and ecological economic agents”, companies must go beyond the pursuit of profits, consider the needs of different stakeholders, and assume greater environmental, social, and governance responsibilities.

First, good ESG performance can enhance interactions with stakeholders, attract more social attention, and promote cooperative relationships [30]. This helps to reduce transaction and agency costs, ensure a stable customer base, and improve competitiveness and total factor productivity. Second, if companies actively follow ESG standards, they will help create a positive image among investors and stakeholders [31], enhancing their reputation and attracting high-quality investment and financing opportunities. This also helps to reduce the cost of obtaining strategic resources and provides favourable conditions for improving total factor productivity [31]. Ding et al. (2024) found that high ESG performance can create unique intangible assets and promote sustainable, growth-oriented development [5]. Governments often introduce supportive policies for companies that excel in ESG, signalling green operations in the capital market [32]. This approach meets the environmental expectations of consumers and investors and builds trust with stakeholders such as governments, communities and investors. Another study also confirmed that companies with good ESG performance prioritize strengthening governance, enhancing employee interactions, cutting agency expenses, and streamlining organizational processes—a win-win situation for everyone involved. Companies that do well in ESG areas can also increase the openness of their internal communications, thereby promoting human resources development and increasing total factor productivity [33]. These companies usually perform well in non-financial sustainability. Good ESG practices can improve transparency and build investor trust [34]. During external crises or public relations challenges, companies with good ESG records usually receive public tolerance [35]. Due to better risk management and sustainable returns, total factor productivity improves, investors’ perception of risk is reduced, and financing costs are lowered. In light of this, the following assumptions are made in this study:

Hypothesis 1: 

ESG rating indices can improve total factor productivity.

(2)

ESG ratings, financing constraints and total factor productivity

Fazzari and Athey (1987) argue that financing constraints arise in imperfect capital markets where internal and external financing are not perfectly fungible, and external financing costs are generally higher [36]. According to Midrigan and Xu, businesses have significant financial limitations, making it difficult to secure adequate funding, forcing them to place the cost of investment above the project’s intrinsic value. This misallocation will lead to suboptimal investment decisions, ultimately reducing total factor productivity [37] Access to government funding and preferential bank loans can effectively ease financing constraints, boosting total productivity. Companies that actively embrace social responsibility also tend to maintain higher-quality financial information. The reputational risk of financial fraud serves as a strong deterrent, encouraging greater transparency. As a result, improved financial reporting helps alleviate financing constraints, increases working capital, and ultimately enhances total factor productivity [38].

The concept of asymmetric information shows that investors are more inclined to choose companies that actively fulfil their ESG obligations when there is insufficient transparency in capital markets [39]. High ESG performance compensates for the lack of transparency in external information disclosure by providing comprehensive non-financial information, and reducing information asymmetry with external investors [40].

In times of confidence crises in the capital market or industry, companies that embrace ESG responsibilities tend to be more stable, which can lower investors’ risk premium requirements [41]. Additionally, companies with strong ESG performance, acting as an implicit contract, consider stakeholder needs in their operations [42]. ESG performance influences investors’ perceptions of an enterprise’s emotional and cognitive reputation through reputation signalling, earning stakeholder trust and easing financing constraints [43]. As the concept of high-quality development grows, investors increasingly consider a company’s ESG performance—beyond financial indicators like profitability—to assess risk management and ensure sustainable returns.

Improvements in financing constraints reduce liquidity risk and transaction costs, allowing enterprises to prioritize project value over financing concerns when making investment decisions [44]. This promotes the efficient use of financial resources, facilitates the conversion of liquid assets into illiquid ones, and encourages more significant investment in high-return technological innovation projects, thereby enhancing total factor productivity [45]. Additionally, strong ESG performance strengthens stakeholder oversight and reduces innovation’s potential cost while increasing total factor productivity through better operational management and lower opportunity costs. Based on the above, we propose Hypotheses 2 and 3

Hypothesis 2: 

The higher the ESG rating index, the lower the financing constraints.

Hypothesis 3: 

Financing constraints will serve as a mediating factor through which ESG ratings impact total factor productivity, meaning that higher ESG ratings are expected to reduce financing constraints, enhancing total factor productivity.

(3)

ESG ratings, research and development (R&D) and total factor productivity

Maximizing enterprise value is the ultimate goal of business operations [46]. Companies fulfilling their ESG responsibilities will actively engage in more research, development, and innovation activities to ensure sustainable operations and long-term growth [19,47]. Executing an ESG strategy facilitates the growth of social networks, the accumulation of resources, and the enhancement of investment in R&D [48]. According to stakeholder theory, companies depend on stakeholders, who exert a binding force on operations [49]. In order to maintain legitimacy, companies must meet society’s expectations, which encourages long-term, healthy decision-making, reduces management’s focus on short-term goals, and promotes the allocation of more research and development [49]. Moreover, socially responsible companies use advanced technology to continuously innovate and optimize products, improving customer satisfaction [50]. Agency theory highlights principal–agent conflicts, where managers tend to favour low-risk, cash-flow stable projects for personal reputation [51]. However, a sound corporate governance structure (an essential element of ESG) can supervise management activities, inhibit selfish behaviour, and balance the interests of shareholders and managers, thereby allowing risks to be taken in R&D [52].

Resource dependence theory suggests that higher ESG levels strengthen ties with stakeholders and help firms access critical resources needed for innovation. Stakeholders from different industries can provide valuable knowledge and technology to enrich a firm’s internal knowledge base and improve innovation and R&D translation [53]. Companies that fulfil their ESG responsibilities can gain support from stakeholders and drive innovative growth.

Total factor productivity (TFP) may be enhanced by technological improvement that results from research and innovation [54]. Poor ESG performance can expose companies to policy penalties, hinder stakeholder resource sharing, and limit access to financial support. Conversely, companies that innovate to reduce pollution and improve ESG management can attract stakeholder resources and expand production, ultimately improving TFP. Research by Xiang (2020) shows that environmental performance disclosure requirements may significantly increase companies’ R&D investment [55].

Li et al. (2022) believe that if companies actively disclose environmental information, such behaviour can promote corporate research development and innovation, thereby reducing environmental pollution [56]. A “dual carbon” strategy that encourages more investment in research and development can boost technical innovation and environmental performance, which strengthens businesses’ knowledge bases and speeds up the process by which new information is turned into productivity [57]. Guellec’s research indicates a favourable association between business R&D spending and productivity in OECD nations [58]. Through proactive ESG management, firms may enhance environmental performance, governance, and overall productivity. Based on the above previous studies, we propose Hypotheses 4 and 5.

Hypothesis 4: 

ESG ratings positively influence research and development (R&D).

Hypothesis 5: 

Research and development (R&D) will serve as a mediating factor in the impact of ESG ratings on total factor productivity; higher ESG ratings are expected to increase research and development (R&D), thereby improving total factor productivity.

3. Research Methods and Data

3.1. Sample Selection and Data Sources

This research examines firms with significant pollution levels listed on China’s Shanghai and Shenzhen stock markets, utilizing corporate financial annual reports as the data source and examining data from 2012 to 2022. It investigates the relationships among ESG ratings, financing constraints, corporate research and development (R&D), and total factor productivity (TFP) within this context. To improve the accuracy of the data, the data are processed as follows: (1) Exclude companies with incomplete data; (2) Exclude ST and *ST companies; (3) The variables’ upper and lower 1% quantiles are adjusted to reduce the impact of abnormal values. Ultimately, we acquired 9125 observations from 1305 qualifying companies. The financial data of publicly traded companies are sourced from the Wind Financial Terminal and the Guotai’an CSMAR database.

3.2. Variable Setting

(1)

ESG Ratings (ESG)

Shanghai Huazheng’s ESG rating is the independent variable in this research. The Shanghai Huazheng ESG rating system collects a large amount of data for analysis and evaluation, considering the internationally mainstream ESG evaluation systems and combining them with the characteristics of the Chinese market [5,59]. The results are finally converted into a rating score and comprehensive rating for investors to understand easily. Its main advantages are comprehensive coverage, fast data updates, and compliance with China’s national conditions.

(2)

Total Factor Productivity (TFP)

Total factor productivity (TFP) is a measure that evaluates the production efficiency and technological progress of a nation, region, or organization. It indicates alterations in the overall efficacy of production inputs. Total factor productivity is often computed using an estimation of the production function. The most used methodologies are the Olley–Pakes (OP) technique and the Levinsohn–Petrin (LP) method. The LP method uses intermediate inputs (such as material consumption) as a proxy for enterprise production efficiency to solve the endogeneity problem. This method assumes that the choice of intermediate inputs is a monotonic function of enterprise production efficiency. This paper uses TFP calculated using the LP method to proxy for total factor productivity [45,60].

(3)

Mediating variable

Financial constraints (SA): Financing constraints reflect a company’s ability to secure external funds. Companies often rely on external financing for investment, research, and expansion. Firms with strong ESG performance may experience lower financing costs and greater access to capital due to their commitment to social responsibility and sound governance, which helps build investor trust. This study utilizes prior research and employs the SA index as a surrogate variable for business funding limitations [61].

Research and development (RD): corporate research and development (R&D) is a crucial driver of total factor productivity (TFP) growth. Innovation activities like R&D and introducing new technologies and processes can directly improve a firm’s productivity and market competitiveness. Companies that score higher on ESG metrics are often more committed to the long term. They are willing to invest in innovation activities, which not only improve their products and services but also improve the efficiency of their production processes. The mediating variable formed through corporate research and development can more specifically illustrate how ESG practices can help companies improve productivity. This study stands in for corporate R&D spending by using the logarithm of such spending [62].

(4)

Control variables

This paper selects enterprise size (Size), equity concentration (Top1), enterprise liquidity (Cr), enterprise age (Lnage), and enterprise leverage (lev) and includes enterprise ROA as a control variable while taking into consideration the fixed impacts of year and industry.

Table 1. Research Model.

Variable	Symbol	Definition
Explained variable	Total Factor Productivity (TFP)	Obtained by LP method measurement
Explanatory variable	ESG Ratings (ESG_score)	Huazheng’s ESG rating is divided into nine levels, from C to AAA, and, through a specific scoring method, assigned a score of 1 to 9.
Mediating variable	Financial constraints (SA)	Interest expense/total debt
	Research and development (RD)	The logarithm of the sum of research and development expenses
Control variable	Enterprise size (Size)	Take the natural logarithm of the total assets at the end of the period.
	Equity concentration (Top1)	The greatest shareholder’s percentage of the total shares
	Enterprise liquidity (Cr)	The current asset-to-current liability ratio
	Enterprise age (Lnage)	Calculate the logarithm to the base 10 of the age of the enterprise.
	Enterprise leverage (Lev)	The ratio of total liabilities to total assets
	Enterprise return on total assets (Roa)	Net profit/average total assets

contains the pertinent variables.

3.3. Research Model

This study selected a two-way fixed-effect model to control for individual effects and time effects, a common method in panel data analysis, to reduce endogenous problems and the impact of omitted variables.

To examine the influence of ESG ratings on total factor productivity, Model (1) is developed:

$$TF{P}_{it}=\alpha +{\beta }_{1}ESG\_scor{e}_{it}+\sum control++u_t+{\mu }_i+{\epsilon }_{it}$$

(1)

This research studies the impact mechanism of ESG ratings on total factor productivity, using a mediation effect test to generate Models (2)–(5):

$$R{D}_{it}=\alpha +{\beta }_{1}ESG\_scor{e}_{it}+\sum control++u_t+{\mu }_i+{\epsilon }_{it}$$

(2)

$$S{A}_{it}=\alpha +{\beta }_{1}ESG\_scor{e}_{it}+\sum control++u_t+{\mu }_i+{\epsilon }_{it}$$

(3)

$$TF{P}_{it}=\alpha +{\beta }_{1}ESG\_scor{e}_{it}+{\beta }_{2}S{A}_{it}+\sum control++u_t+{\mu }_i+{\epsilon }_{it}$$

(4)

$$TF{P}_{it}=\alpha +{\beta }_{1}ESG\_scor{e}_{it}+{\beta }_{2}R{D}_{it}+\sum control++u_t+{\mu }_i+{\epsilon }_{it}$$

(5)

In this article, $ESG\_score$ is the explanatory variable (independent variable), while TFP is the variable to be elucidated (dependent variable). $SA$ denotes the financing constraints of the enterprise, and $RD$ represents the research and development of the enterprise, acting as the mediating variable in this study. $\sum Control$ signifies the control variable, $u_t$ indicates the temporal fixed effect, $u_t$ represents the individual fixed effect, and ${\mu }_i$ is the stochastic disturbance term. By including individual and time-fixed effects, this model may reduce the impact of missing variables and improve the accuracy of its estimates by considering the impact of fixed individual and temporal features on the dependent variables.

4. Results

4.1. Descriptive Statistics

According to

Table 2. Descriptive statistics of major variables.

Variable	N	Mean	SD	Min	Max
TFP_LP	9125	8.293	1	3.864	11.81
ESG_score	9125	72.52	5.740	36.62	90.93
Size	9125	22.37	1.353	16.12	28.64
Lnage	9125	2.304	0.755	0	3.434
Top1	9125	34.32	14.80	0.290	89.99
Cr	9125	0.479	0.184	0.019	1
Lev	9125	0.425	0.241	0.014	10.08
Roa	9125	0.039	0.067	−0.244	0.236
SA	9125	3.546	0.313	1.800	4.349
RD	7127	17.88	1.646	0	23.89

, a total of 9125 samples were collected for this study. The mean total factor productivity (TFP) from 2012 to 2022 is 8.293, with most of the sample distributed between 7.632 and 8.871, and the median is 8.221, indicating no significant skewness. The average value of the liquidity indicator Cr is 0.479, indicating that enterprises are relatively efficient in the use of funds and have relatively good liquidity. Most enterprises have a positive ROA, indicating that they are doing well, and there are relatively few loss-making enterprises. The remaining control variables are also within an acceptable range.

According to the research, the VIF may show if the two variables (the independent and the control) are multiple collinear. In most cases, a VIF below 10 is deemed satisfactory [5]. All of the primary variables’ VIF values are below 10, and the average is below 3, as shown in

Table 3. Multicollinearity analysis.

Variable	VIF	1/VIF
Lnage	5.88	0.169926
SA	4.89	0.204666
Size	2.87	0.348008
Lev	1.73	0.579115
RD	1.57	0.63632
Roa	1.42	0.706248
Cr	1.33	0.754124
ESG_score	1.16	0.864662
Top1	1.15	0.869989
Mean VIF	2.44

. This proves that the variables utilized in this study are unaffected by multiple collinearity.

4.2. Analysis of Regression Results

To test Hypothesis 1, which looks at how ESG ratings affect overall factor productivity, we ran a multiple regression analysis, and the findings are in

Table 4. Regression results of ESG ratings and total factor productivity.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
VARIABLES	TFP_LP	TFP_LP	TFP_LP	TFP_LP	TFP_LP	TFP_LP	TFP_LP
ESG_score	0.0090 ***	0.0054 ***	0.0052 ***	0.0052 ***	0.0041 ***	0.0039 ***	0.0031 **
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Size		0.4626 ***	0.4680 ***	0.4634 ***	0.4973 ***	0.4981 ***	0.4774 ***
		(0.03)	(0.03)	(0.03)	(0.03)	(0.03)	(0.02)
Lnage			−0.0560 *	−0.0452	0.0631 *	0.0724 **	0.0862 ***
			(0.03)	(0.03)	(0.03)	(0.03)	(0.03)
Top1				0.0015	0.0016	0.0016	0.0012
				(0.00)	(0.00)	(0.00)	(0.00)
Cr					1.2410 ***	1.2299 ***	1.0682 ***
					(0.10)	(0.10)	(0.10)
Lev						−0.0638	0.1049 ***
						(0.05)	(0.04)
Roa							1.5344 ***
							(0.12)
Observations	9125	9125	9125	9125	9125	9125	9125
R-squared	0.887	0.914	0.915	0.915	0.925	0.925	0.930
Control	NO	YES	YES	YES	YES	YES	YES
Company FE	YES	YES	YES	YES	YES	YES	YES
Year FE	YES	YES	YES	YES	YES	YES	YES

Robust standard errors in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

. Despite progressively including control factors, the analysis shows that the primary explanatory variable’s (ESG_score) coefficient is still substantially positive at 1% and 5%. A one-unit increase in the ESG score is associated with a 0.0031 increase in the enterprise’s TFP_LP when other variables are controlled for. These results support Hypothesis 1, which states a positive correlation between ESG and TFP.

4.3. Mediating Model

The impact of financing constraints as a mediating variable is shown in

Table 5. Mediating effects of corporate finance constraints.

	(1)	(2)	(3)
VARIABLES	TFP_LP	SA	TFP_LP
SA			−0.3343 **
			(0.17)
ESG_score	0.0039 ***	−0.0009 ***	0.0036 ***
	(0.00)	(0.00)	(0.00)
Size	0.4981 ***	0.0110	0.5018 ***
	(0.03)	(0.01)	(0.02)
Lnage	0.0724 **	0.0608 ***	0.0927 **
	(0.03)	(0.01)	(0.04)
Top1	0.0016	−0.0001	0.0016
	(0.00)	(0.00)	(0.00)
Cr	1.2299 ***	−0.0630 ***	1.2088 ***
	(0.10)	(0.02)	(0.10)
Lev	−0.0638	−0.0796 **	−0.0904 **
	(0.05)	(0.04)	(0.05)
Observations	9125	9125	9125
R-squared	0.925	0.982	0.926
Control	YES	YES	YES
Company FE	YES	YES	YES
Year FE	YES	YES	YES

Robust standard errors in parentheses, *** p < 0.01, ** p < 0.05.

. The negative correlation between the two variables is statistically significant at the 1% level of analysis. ESG performance affects its total factor productivity, as shown in column (3), where the financing constraint coefficient also shows a negative correlation at the 5% significance level. Although ESG_score has a significant coefficient in column (3), the mediating effect of financing constraints causes a 1-unit increase in the company’s ESG_score to increase total factor productivity by 0.0036 units. Therefore, Hypotheses 2 and 3 are supported.

Results from the analysis of corporate R&D’s mediating influence are shown in

Table 6. Mediating effects of corporate research and development.

	(1)	(2)	(3)
VARIABLES	TFP_LP	RD	TFP_LP
RD			0.0555 ***
			(0.01)
ESG_score	0.0031 **	0.0097 ***	0.0035 **
	(0.00)	(0.00)	(0.00)
Size	0.4774 ***	0.7856 ***	0.4152 ***
	(0.02)	(0.06)	(0.02)
Lnage	0.0862 ***	−0.3167 ***	0.1568 ***
	(0.03)	(0.11)	(0.03)
Top1	0.0012	−0.0032	0.0012
	(0.00)	(0.00)	(0.00)
Cr	1.0682 ***	−0.0004	1.0210 ***
	(0.10)	(0.22)	(0.09)
Lev	0.1049 ***	−0.1757	0.1864 **
	(0.04)	(0.16)	(0.07)
Roa	1.5344 ***	0.8067 ***	1.5510 ***
	(0.12)	(0.28)	(0.14)
Observations	9125	7127	7127
R-squared	0.930	0.825	0.945
Control	YES	YES	YES
Company FE	YES	YES	YES
Year FE	YES	YES	YES

Robust standard errors in parentheses, *** p < 0.01, ** p < 0.05.

. According to the regression findings, the ESG score has a positive correlation with business R&D, and this correlation is significant at the 1% level. At the 5% significance level, there is a positively linked coefficient for corporate R&D in column (3), suggesting that corporate R&D mediates the relationship between ESG and total factor productivity. Corporate research and development acts as a mediator, increasing TFP by 0.0035 units for every.

4.4. Robustness Test and Endogeneity Problem

The postponed ESG score functioned as an ancillary variable in this research, selected for its relevance, uniformity, and independence from the company’s prior year’s activities [5,63]. In the two-stage regression test of the instrumental variable, the Cragg–Donald Wald F statistic is 742.909, which is higher than the 16.38 at the 10% significance level, indicating that there is no weak instrument problem and that the lagged one-period ESG score is valid as an instrumental variable. According to

Table 7. Solving the endogenous problem: instrumental variable method.

	(1)	(2)
	ESG_Score	TFP_LP
L. ESG_score	0.3247 ***
	(0.01)
ESG_score		0.0114 ***
		(0.00)
Size	0.6492 ***	0.4605 ***
	(0.18)	(0.01)
Lnage	−0.3725	0.0823 ***
	(0.43)	(0.03)
Top1	−0.0115	0.0015 **
	(0.01)	(0.00)
Cr	−0.1132	1.1094 ***
	(0.67)	(0.04)
Lev	−3.3780 ***	0.2057 ***
	(0.72)	(0.04)
Roa	1.8804	1.6222 ***
	(1.16)	(0.07)
_constant	37.0082 ***
	(3.81)
N	7637	7517
adj. R2	0.639	0.546

Robust standard errors in parentheses, *** p < 0.01, ** p < 0.05.

, at the 1% significance level, there is a significant positive correlation between the instrumental variable and the original explanatory variable, confirming the instrument's relevance. The second-stage regression verifies that ESG scores benefit from total factor productivity at the 1% significance level, so the endogeneity test is passed, and Hypothesis 1 remains valid. The test is passed, and Hypothesis 1 remains valid.

All fundamental coefficients were validated as significant. The mediating variables were analyzed using the Sobel test to corroborate the results further. As shown in

Table 8. Testing of the financing constraints mediating effects.

		Coef	Std Err	Z	p > Z
Sobel		−0.0002138	0.00006813	−3.138	0.00170054
Goodman-1	(Aroian)	−0.0002138	0.00006899	−3.099	0.00194096
Goodman-2		−0.0002138	0.00006726	−3.179	0.00148007

, the Sobel test yielded a Z-value of 3.138, exceeding the critical threshold of 2.58, with a corresponding p-value of less than 0.01. This result confirms the statistical significance of the mediation effect.

As indicated in

Table 9. Testing of the research and development mediating effects.

		Coef	Std Err	Z	p > Z
Sobel		0.00077464	0.00019748	3.923	0.00008756
Goodman-1	(Aroian)	0.00077464	0.00019786	3.915	0.00009039
Goodman-2		0.00077464	0.00019709	3.93	0.00008481

, the Sobel test produced a Z-value of 3.923, well above the critical threshold of 2.58. The corresponding p-value is below 0.01, confirming the validity of the Sobel test.

5. Conclusions

5.1. Summary

This research examines the A-share markets in Shanghai and Shenzhen from 2012 to 2022 to determine how polluting businesses’ total factor productivity (TFP) changed as a result of their corporate ESG ratings. The results show that good ESG performance can significantly increase TFP, a conclusion that confirms our previous hypotheses and reinforces the findings of earlier studies [15,64]. This suggests that companies with effective ESG practices utilize resources more efficiently, increasing productivity. Several factors contribute to this positive correlation: improved reputation and trust among stakeholders, increased access to financing that reduces constraints, enhanced operational efficiency through sustainable practices, and a long-term strategic focus that fosters innovation. Additionally, the mediating roles of financing constraints and corporate R&D underscore the significance of ESG in boosting productivity. These findings emphasize the multifaceted benefits of incorporating ESG factors into corporate strategy, which can enhance productivity and promote sustainable development for firms and the broader economy. This study broadens our understanding of how ESG ratings affect TFP, particularly for listed firms in developing countries and highly polluting industries, highlighting the urgent need for transformation and reinforcing the case for investing in ESG initiatives.

5.2. Revelations

This study provides empirical evidence that there is a positive relationship between ESG and TFP and offers some key insights. Companies should prioritize enhancing their ESG ratings as a strategic avenue for promoting TFP growth. Strong ESG performance helps reduce financing constraints, improve capital access, and boost research and development investment, all of which ultimately drive TFP. To this end, companies should fully integrate ESG principles into their operations, strengthen environmental responsibility, reduce financing barriers, and increase R&D investments to enhance competitiveness and meet social obligations. Additionally, the government should strengthen ESG disclosure standards, offer clear guidance on corporate ESG practices, and provide policy incentives for high-performing companies while imposing appropriate restrictions on those with lower ratings. For highly polluting industries, specific policies should encourage the adoption of robust ESG practices, motivating these enterprises to contribute actively to TFP improvement.

5.3. Limitations and Future Research

The limitations of this study are as follows: First, this study was conducted in China and only focused on polluting enterprises listed in China. In order to fully understand the impact of ESG in various economic environments, future research should expand the sample scope to include companies from different industries and countries. Second, this study also discusses the relationship between ESG and TFP, particularly emphasising the mediating role of financing constraints and corporate research and development. Future research should investigate more mechanisms better to understand the impact of ESG on productivity outcomes.