Article Not peer-reviewed version

Evaluating the Influence of ESG Ratings

on the Corporate Operational
Performance: Evidence from
Commercial Banks in China

Honglin Wang *

Posted Date: 28 August 2024

doi: 10.20944/preprints202408.1978.v1

Keywords: ESG rating; Commercial Banks; Operating Performance; Empirical Analysis

Preprints.org is a free multidiscipline platform providing preprint service that

is dedicated to making early versions of research outputs permanently
available and citable. Preprints posted at Preprints.org appear in Web of
Science, Crossref, Google Scholar, Scilit, Europe PMC.

Copyright: This is an open access article distributed under the Creative Commons
Attribution License which permits unrestricted use, distribution, and reproduction in any
medium, provided the original work is properly cited.
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

Disclaimer/Publisher’s Note: The statements, opinions, and data contained in all publications are solely those of the individual author(s) and
contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting
from any ideas, methods, instructions, or products referred to in the content.

Article

Evaluating the Influence of ESG Ratings on the

Corporate Operational Performance: Evidence from
Commercial Banks in China
Honglin Wang
University of Macau; [email protected] or [email protected]

Abstract: With growing global emphasis on sustainable development and social responsibility,
Environmental, Social, and Governance (ESG) ratings have emerged as critical metrics to evaluate
commercial banks’ operating performance. The paper utilizes panel data spanning 2009 to 2022 from
large state-owned and joint-stock commercial banks in China. It employs a two-way fixed effect
model and robustness tests, alongside tests for heterogeneity analysis and moderating effect to
conduct empirical analysis. The findings of the fixed effect model indicate that improvements in
ESG performance have advantageous impacts on commercial banks’ operating performance,
particularly with higher growth rate for operating income Robustness tests including mixed
regressions, Tobit tests, variable substitutions, and tests of endogenous bias have validated that
initial regression findings still hold. Tests of heterogeneity analysis reveal that among large state-
owned commercial banks, banks with higher leverage, larger major shareholder ownership, and
larger banks scales, ESG performance can significantly affect the banks’ operating performance, and
the relationship exhibits positive. Furthermore, the paper finds that GDP growth rate, CPI and
analysts’ coverage can strengthen the relationship between ESG performance and commercial banks’
operating performance through moderating effects. The research provides policy proposals and
strategic recommendations tailored for policymakers, banks, and investors, building on these insights.

Keywords: ESG rating; commercial banks; operating performance; empirical analysis

1. Introduction
In recent years, Chinese companies significantly demonstrated a notable acceleration in the
development of environmental, social, and governance (ESG) performance, in line with the ‘dual
carbon' goals. The Central Financial Work Conference of October 2023 emphasized the critical role
that finance plays in supporting domestic economic growth and improving core competitiveness. At
the same times, five major articles proposed in the conference have laid a theoretical foundation for
high-quality development in finance. The five major articles consist of Fintech Finance, Green
Finance, Inclusive Finance, Pension Finance and Digital Finance.
As a responsible major country, China made a solemn commitment to devote itself to the global
low-carbon transformation at the 75th United Nations General Assembly in 2020 and it introduced
‘dual carbon’ goal, whose purposes to achieve carbon peak in 2030 and carbon neutrality in 2060.
‘Dual carbon’ goal aims to reduce carbon dioxide emissions, which helps create green production
and operation models for businesses as well as a green transformation of the socioeconomic structure.
Emergence of green finance, to a certain extent, contributes to ‘dual carbon’ goals and set higher
standards in operations for enterprises, including business models and aspects of environmental,
social and governance.
ESG concept is spreading across the entire world, emphasizing that enterprises ought to align
with current trends and concentrate on green environmental protection, energy conservation and
low-carbon development meanwhile pursuing high-quality development. As a component of the
ESG concept, ESG performance measures an enterprise’s sustainability along three dimensions:

© 2024 by the author(s). Distributed under a Creative Commons CC BY license.

Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

environment, society, and corporate governance. The dimensions are scored for their respective
performance, and these scores are then calculated into an ESG score, which aims to assign
comprehensive ratings for companies to evaluate their ESG performance. Incorporating ESG ratings
into business operations has been evidenced to have a positive impact on mitigating misconduct,
accelerating the green transformation of business models and achieving sustainable development
goals for enterprises.
As a pillar of the financial industry, China's commercial banks play a crucial role in driving
economic progress. This paper employs Stata to conduct regression analysis, investigating the impact
of ESG ratings on operating performance of Commercial banks in China. Innovations of the paper
are to explore determinants of commercial banks’ operating performance, for the purpose of
extending existing research and filling a gap between ESG ratings and financial institutions. Secondly,
econometric methodologies deployed in the paper are statistically rigorous. It mainly adopts fixed-
effect models controlling for year and individual effects, supplemented by bank-level variables.
Besides, robustness tests contain mixed regressions, Tobit models, and endogeneity analysis.
Meanwhile, heterogeneity analysis exists in the paper, considering different natures of commercial
banks, shareholder structures, and leverage ratios. Additionally, moderating effect analysis
incorporates GDP growth rates and CPI, along with analysts’ coverage, to explore their moderating
effects on banks’ operating performance. The database involved in this paper is a panel dataset, which
can reflect the real-time data and cover various banks. The research background aligns with the 'dual
carbon' goals and reflects the ESG trends. The findings provide practical insights for bank managers,
policymakers, and investors, offering guidance on strategic decisions and investment frameworks in
line with sustainable development objectives.

2. Literature Review

2.1. Research on ESG Performance

2.1.1. The Role of ESG in Enterprises and Financial Markets

Under the guidance of the ‘dual carbon’ goal, ESG performance has attracted many scholars to
study and explore it. A strand of literature is mainly around how the ESG performance constrain
inappropriate behaviors and decision-making process within organizations, while maximizing
enterprises values.
Improving ESG performance is beneficial for curbing corporate fraud as evidenced by Chinese
listed A-shares from 2012 to 2022, at the same time maintaining a good ESG performance can enhance
information transparency and alleviate financial restrictions under mediating effect [1]. The higher a
company’s ESG engagement, the more it can significantly constrain management's dereliction of
duty, which is more evident in companies with fewer institutional investors and information
asymmetry [2]. Thus, ESG practices enable companies to reinforce sustainability and maximize
shareholders’ value.
Besides, enhancement in ESG performance can significantly stabilize earnings in firms and
reduce inefficient investment activities, which is help of improving investment efficiency as stated in
[3]. ESG ratings can provide guidance on decision-making for managers and significantly regulate
risk-taking events involved in the organizations, which is more common in environments with low
information transparency, weak corporate governance, and low pressure from external regulatory
environment [4].
In the context of financial crisis and COVID-19, evidence shows that ESG performance helps to
improve the operating performance of listed companies. when confronted with challenges posed by
the global pandemic, firms with good ESG performance appears more resilient and can exert more
discernible influence on withstanding with the challenges, which is more significant in small
companies, state-owned enterprises and highly competitive market environment [5].
Firms with excellent and poor ESG ratings have important spillover effects on financial market
which is more pronounced in firms with excellent ESG ratings, besides ESG rating can be regarded as
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

the benchmark for regulation and incorporated into systematic risk assessments, aiming to effectively
stabilize companies’ financial positions [6]. Besides, investment portfolios containing major stocks with
good ESG performance often outperform than those with poor ESG performance during financial crisis,
whereas the effect is relatively weak in the normal period [7]. The results imply that superior ESG
performance significantly eliminates financial related risks and can cope with the crisis.
Overall, ESG performance plays an important role in regulating decision-making process,
mitigating risk-taking, promoting effective investing practices. Besides, ESG performance can help
enterprises and investors manage with the impact of the global financial crisis and the COVID-19
epidemic. A robust ESG performance indicates an enterprise has the enhanced capacity to withstand
systemic shocks.

2.1.2. The Determinants of ESG Performance

A review of extant literature reveals a multitude of factors that exert a profound and pervasive
influence on ESG performance. From the standpoint of financial markets, factors include consumer
concentration, investor attention and analysts’ coverage. At the enterprise level, factors that shape
ESG performance encompass digital transformation, equity incentive plans, mergers and acquisitions,
and the financial gaps. From the standpoint of government, the impact of government environmental
protection expenditures on ESG performance is a crucial consideration.
Consumer concentration impairs a company's operating cash flow and innovation capacity,
resulting in a notable decline in ESG performance which is particularly pronounced in large scale
enterprises, non-state-owned enterprise, low competitive market environments and young
executives [8]. The inhibitory effect also exists in the relation between investor attention and ESG
performance. There is a mutual inhibitory effect between investor attention and a company's ESG
performance, meanwhile enhanced ESG performance of the company is help of improving ESG
performance of neighboring companies [9]. In details, for each additional unit increase in investor
attention, the company’s ESG performance will decline by 0.25 units, and vice versa.
Besides, analyst coverage is an indicator of analyst attention. Strong ESG performance can attract
more analyst coverage, which is contributed by mechanism effect of diminishing information
asymmetry [10]. Meanwhile, high degree of analyst attention improves information transparency
between enterprises and public [11]. Media attention and analyst coverage have essential influences
on improving ESG performance, meanwhile companies with high-quality ESG performance tends to
attract attention of media and analysts, which in turn maximizing corporate value contributed by
stakeholder pressure [12]. Analyst coverage significantly enhances corporate ESG performance and
incentives enterprises to engage ESG practices, through attracting media attention and more site
visits [11]. Thus, it can be reached out conclusions that it exists mutual promotion relationship
between analyst coverage and ESG performance.
Against the backdrop of the accelerated development of digital technology, digitalization is
playing a central role in the operations of companies, while digital transformation has the capacity to
improve ESG performance, especially for state-owned enterprises and conventional companies with
elevated pollution levels [13]. That means digital transformation is predicted to lead to a notable
improvement in those companies’ ESG performance.
Executive equity incentive plans (EEIPs) are the one of various methods that enterprises can
employ to motivate and retain employees, referring that companies grant equity incentive plans to
executives, who can purchase company’s shares at a price lower than the market price. Executive
equity incentive plans (EEIPs) have a beneficial influence on ESG performance, and the effect is
evident in EEIPs with a long effective term and high vesting targets, at the same time the impact
appears to be more pronounced in non-state-owned firms and in those with exemplary external
governance [14].
Mergers and acquisitions (M&A) are a necessary part of business. Some studies show that M&A
activity has a positive impact on ESG performance, whereas the impact is not instantaneous, and
instead the company’s ESG score has been upgraded in the subsequent year after the transaction’s
conclusion [15].
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

From perspectives of government, an increase in government investment on environmental

protection brings a new green innovation wave, that leads to improvements in both the ecological
environment and the social landscape, which ultimately enhances country governance standards and
has a positive impact on a country’s ESG performance [16].
ESG controversy is another factor that affects ESG performance. Financial gaps of enterprises
incentive to improve their ESG performance, which is diminished by ESG controversies [17]. ESG
controversies refer that different opinions exist in firms regarding the ESG practices, which causes
that the likelihood of the company engaging ESG practices is reduced, and financial gaps are less
likely to be closed.

2.2. Research on Commercial Banks’ Operating Performance

Commercial banks can satisfy different needs of market entities and drive domestic economies up
as an important component of financial market. After reading literature of recent five years, some
factors that affects operating performance of commercial banks can be found. The determinants mainly
focus on green credit, internet finance, fin-tech products and corporate social responsibility (CSR).
Green credit, a part of green finance refers that financial institutions subsidize low interest rates
and simplify credit approval procedures for green industries, which is related to energy conservation
and low carbon emission. At the same time, they elevate the threshold of loans and impose restrictive
methods on enterprises which contains high pollutant, emission and energy-consumption projects.
Thus, green credit regarded as a type of financial policy effectively serves to constrain pollutants,
aiming to realize ‘dual carbon’ goals and promote sustainable development in enterprises and society.
Green credit can impact diverse facets of a commercial bank. Some evidence demonstrates that
implementation of green credit can reduce various bank-related risks to a certain extent, including
credit and reputation risks since green credit can enhance banks’ core competencies [18]. Besides,
commercial banks that implement green credit policy can effectively improve their financial
performance and boost green development, which stems from green credit can improve returns on
banks and attract higher economic profits [19].
CSR is another factor that influences on banks’ profitability. Specifically, fulling CSR obligations
brings negative impacts on banks’ financial performance in a short term, whereas fulfilment of CSR
improves operating performance, under moderating effect of green credit [20].
The emergence of internet finance brings people lots of conveniences and encompass various
features. For instance, online payment platform simplifies payment procedures. Internet lending
platform can optimize lending approval process through shortening process time and avoiding
traditional lending procedures. Internet finance is beneficial to promoting enhancements in
commercial banks’ comprehensive performance, i.e. profitability, which is more evident in city
commercial banks, whereas it does not bring impacts on state-owned commercial banks and the
limitation of internet finance is that it impedes liquidity in commercial banks [21].
As the understanding of financial literacy deepens, the number of fintech products is increasing,
with different implications for commercial banks as issuers. In details, perceived usefulness of fintech
products can enhance consumer satisfaction, customers’ expectations, in contrast perceived
difficulties of the products play a vital role in improving banks services quality [22].

2.3. The Impact of ESG on the Operating Performance of Commercial Banks

Some enterprises’ profits are prone to be impacted by external environments, such as Covid-19
and financial crisis. In the meantime, green finance is a part of five major articles proposed on
Conference Finance Work. It suggests that enterprises should implements ESG practices while
pursuing profits. ESG performance includes ESG ratings and its sub-index scores, including
environment, society and governance. Extensive literature examines the relationship between ESG
performance and firms value, whereas little research is related to banking sectors and some analysis
reach out mixed evidence.
Enhancement in ESG performance is conducive to improving corporate performance.
Enterprises implement sustainability management initiatives, which improves social sustainable
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

development but also enhances their financial performance constructs [23]. ESG performance has a
positive correlation between with corporate performance, which is more significant for large-scale
companies and those involved with high-risk cases [24]. Moreover, improvements in listed
companies’ ESG performance can increase their market value and have a positive role in operation
capabilities for state-owned listed companies through the mediation effect, which ultimately has
impacts on market value [25].
Besides, ESG performance of enterprises can positively promote improvements in corporate
financial performance, contributed by alleviating financial constraints and enhancing operation
efficiencies with the organization, which is more significant in non-state-owned enterprises, medium-
small enterprises and manufacturing enterprises [26]. Banks which engages in environmentally
friendly activities exert the greatest influences on their value, at the same times banks that prioritize
ESG initiatives can reduce the cost of equity and improve cash flows whereas cost of debt is not
impacted [27]. Furthermore, a trade-off exists between bank value and risk-taking, a more stable
financial system, because high ESG scores are accompanied by excessive investment of ESG practices,
thereby leading to lower bank values [28].
Overall, existing literature have conducted certain research on ESG performance and the
determinants of operating performance in banking sectors. However, as far as the author’s research
is concerned, the existing literature still has some shortcomings. The two main points are as follows:
firstly, most of the literature fail to consider factors that affects operating performance of commercial
banks and panel data of listed banks is not deployed for thoughtful analysis. Secondly, institution-
related variables such as analyst’s coverage and economic variables such as GDP as moderators, are
not utilized in the most of existing literature, nor does it test the moderating effect. The deficiencies
serve as an innovation and a breakthrough point for the research paper.

3. Empirical Analysis Process

3.1. Data Sources and Sample Selection

This article investigates the link between environmental, social and governance (ESG) ratings
and Chinese commercial banks’ operating performance from 2009 - 2022. It selects 15 Chinese
commercial banks as research targets.
The China Financial Regulatory Bureau classifies the banking sector into six state-own
commercial banks: Industrial and Commercial Bank of China, Agricultural Bank of China, Bank of
China, China Construction Bank, Bank of Communications, and Postal Saving Bank of China.
Additionally, there are nine joint-stock commercial banks, including Everbright Bank, HuaXia Bank,
Minsheng Bank, Ping An Bank, Shanghai Pudong Development Bank, Industrial Bank, China
Merchants Bank, China Zheshang Bank, and China CITIC Bank.
The bank-level characteristics are derived from the Wind financial database and the indicator
consist of the debt-to-asset ratio (DTA), return on total assets (ROA), leverage ratio (LR), total asset
turnover rate (TAT), etc. The same period of China Shanghai Huazheng ESG Rating is chosen for this
research, since the rating can reflect commercial banks’ comprehensive ESG performance, cover
extensive banks and align with China’s national conditions [29]. Then, a dynamic panel dataset is
formed, together with banks’ characteristics data and the sample size is 2940.

3.2. Variable Description

The study employs two explained variables, one core explanatory variable, five control
variables, three threshold variables and three moderating variables. Table A1 presents the variable
names (abbreviations), types, and variables constructions.

3.2.1. Explained Variables: Operating Performance of Chinese Commercial Banks

Return on assets (ROA) and return on equity (ROE) are two accounting-based indicators, which
is employed to assess the operating performance of the banks in this research. ROE measures the
returns generated by each unit of net capital invested by shareholders. ROA provides a
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

comprehensive measure of the profit generated by the total assets of a business, i.e. the total funds
invested by shareholders and creditors. Higher indicators signify that the bank is more profitable,
which translates to a higher return on investment for investors. In this paper, ROA is selected to
measure the operating performance of the bank.

3.2.2. Core Explanatory Variables: Shanghai Huazheng ESG Rating

The paper chooses Chinese Commercial Banks' Shanghai Huazheng ESG rating as core
explanatory variable. Huazheng ESG rating system classifies companies into nine levels in accordance
with their ESG performance, which is specified as AAA, AA, A, BBB, BB, B, CCC, CC, C, then
transforms them into ESG rating scores from 9 – 1 [29]. For instance, if the highest level of Huazheng
ESG rating is AAA, ESG score of the bank is 9. On the contrary, when the ESG score is 1, the bank's ESG
ratings is correspondingly C. The better the company’s ESG performance, the higher its score.

3.2.3. Control Variables

The article selects five financial indicators as control variables, consisting of DTA, TAT, OI, IA
and FA. The purpose is to avoid the influences of potential exogenous factors on the regression model
and to better capture the factors that affect the operating performance of commercial banks in China.
Specifically, the control variables are as follows:
Debt-to-asset ratio (DTA): The ratio measures financial risks in organizations, and the ratio in
the banking sector is higher than other sectors. Debt-to-asset ratio is often used as a benchmark for
predicting debt crisis; Total asset turnover ratio (TAT): The ratio measures operation capacity. High
indicator signifies the better liquidity and operating efficiency in the companies; Operating income
growth rate (Growth): This metric depicts the growth rate of financial institutions. The higher the
operating income growth rate, the more value the financial institution creates and the better its
growth potential; Intangible assets ratio (ITA): This indicator is defined as the ratio of intangible
assets to total assets. The typical intangible assets contain R&D, goodwill and patents, etc.; Fixed
assets ratio (FA): It is defined as the ratio of fixed assets to total assets. Fixed assets are tangible items
owned by companies, utilized in operations for long-term financial benefits.

3.2.4. Moderators
Three indicators are selected as moderators for the analysis, consisting of year-on-year growth
rate of GDP, CPI index and analyst coverage, and their impacts on ESG ratings and operating
performance of commercial banks in China are evaluated. The first two indicators are related to
economic factors, while the latter is an indicator of institutional investors activity.
GDP(YoY): Gross domestic product (GDP) is the most effective indicator of a country’s economy.
YoY measures year-on-year growth rate of a GDP in comparison to the previous year. Consumer price
index (CPI): CPI is a common macroeconomic indicator used for measuring inflation [30]. Analyst
coverage: The number of analysts and research reports focused on the specified bank as proxy
variables and the indicator is calculated by the natural logarithm of the number of analysts who paid
attention to the bank during the year, plus one [11].

3.2.5. Threshold Variables

The article selects bank characteristic indicators: leverage ratio, firm size and largest
shareholder's shareholding ratio as threshold variables. In terms of average means of the variables,
15 sample banks are divided into two different groups for the regression analysis.
Leverage ratio (Lev): This refers to the ratio of equity to total assets in the balance sheet. Bank size
(Size): The ratio measures the bank scale and calculated by natural logarithm of banks' total assets.
The share of the largest shareholders (Top): In this paper, the number of shares held by the top 10 percent
of shareholders, as a percentage of the total number of shares issued by the company, is selected.
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

3.3. Basic Characteristics of the Data

This section presents the fundamental characteristics of the dataset. As evidenced in Table A2,
the total number of individual variable observations is 210, while the overall sample observations
amount to 2940. The average ROA of the sample banks is 0.97%, the minimum is 0.34% and the
maximum is 1.47%. The mean value of the debt-asset ratio is 93.29%, which is significantly higher
than the other segment ratios.

4. Modelling
This paper firstly adopts fixed effects regression as the baseline regression model for the study
and considers the impact of individual and year effects on the model. The function of baseline
regression model as shown below.

𝑅𝑂𝐴 = 𝛼 + 𝛽 𝐸𝑆𝐺 + ∑ 𝛽 𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑠 + 𝜇 + 𝜃 + 𝜀 (1)

𝑅𝑂𝐴 is operating performance of bank i in year t. 𝛽 denotes the coefficients of variables,
measuring sensitivities of ROA to changes in each variable. 𝐸𝑆𝐺 , is the ESG score, transformed from
Shanghai Huazheng ESG ratings. 𝜇 and 𝜃 denotes the identity effect and year effects, controlled for
the model. 𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑠 denotes the bank characteristics statistics. 𝜀 is error term of the model.
Secondly, the robustness tests used to prove the robustness and validity of the conclusions,
which is carried out in three statistical methodologies: the first is mixed regression, i.e., OLS
regression; the second is Tobit test, which is utilized to examine the regression model under the
truncated dependent variables; and the third involves variable substitution, replacing ROA with ROE
to conduct new regression experiments.
Meanwhile, commercial banks in China exhibit distinctive characteristics, such as: different
ownership, capital structure and so on. Therefore, it is not appropriate to apply a single regression
model to the entire sample. To mitigate the influence of specific characteristics on the outcome, the
paper decides to adopt tests for heterogeneity analysis to investigate the relationship. This research
implements a heterogeneity analysis by grouping 15 types of commercial banks according to their
distinctive characteristics. Furthermore, the analysis considers three exogenous variables, two of
which pertains to macroeconomic indicators and one of which pertains to institutions indicator which
aims to investigate the roles of these variables in the relationship between ESG performance and
operating performance in the banks.

5. Discussions
The two-way fixed effects model represents the benchmark regression model of this paper,
namely multivariate regression model implemented after controlling for individual and year effects.
The two-way fixed effect model eliminates a certain extent of omitted variable bias caused by
individual and year fixed effects [31]. Besides, endogenous bias is common in econometrics, referring
to the fact that a single explanatory variable is correlated with an error term, or two error terms are
correlated, which can lead to generate inaccurate estimates, misleading conclusions and wrong
theoretical explanations [32]. Besides, the paper also noted that although the sources of endogenous
bias are various, different approaches can address these issues. One approach deployed in the paper
is dynamic generalized method of moments (GMM) model and Stata as statistical software can be
utilized to perform GMM model tests in the panel dataset.
This paper employs a GMM model, mixed regression, Tobit test and explanatory variable
replacement to assess the robustness of the results. The whole group is divided into two groups based
on different characteristics and tested for heterogeneity. Moreover, three exogenous variables are
considered to explore the relationship between ESG ratings and commercial banks’ performance.
This section displays empirical findings of the models. Table 3 presents the outputs of the two-
way fixed effects model, while Tables 4–7 illustrate findings of the robustness tests, Tables 8–11
display the results of the heterogeneity tests. Finally, Tables 12–15 present the results of the
moderating effect analysis.
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

5.1. Baseline Regression

This study aims to explore the relationship between ESG performance and commercial banks’
operating performance (OP). Table A3 presents outcomes of the baseline regression model. The
results show that there is a positive relationship between ESG ratings and ROA, which is significant
at 1%, 5% and 10%. And R2 is increasing when control variables are added incrementally in the
models, which means that the model’s accuracy is gradually improved after introducing five control
variables. The full model’s result show that the coefficient of ESG rating is 0.0242, meaning that one
unit increase in ESG rating of commercial banks brings approximately 2% in the ROA. Explanatory
power of ESG rating is strengthened when all control variables are included, and the relation is
statistically significant at any levels as stated in the table. Therefore, ESG ratings have positive
impacts on the 15 commercial banks’ operating performance, which can be regarded as an essential
determinant of operating performance.
The evidence is highly consistent with [33], that ESG performance measured by the combined
score exhibits a positive and significant relationship with firm profitability, signifying high ESG
investing brings promised returns for the firm based on value and profitability. As for investors, they
are willing to focus on high ESG-scored banks and invest in the banks which can help to earn profits.

5.2. Robustness Tests

Table A4 describes the outputs of mixed regressions. Mixed regression as a part of robustness
tests is used for validating whether empirical findings are consistent among the series of tests. The
evidence shows that coefficients of ESG ratings are around 4%, significant at any levels. It means that
ESG ratings have positive impacts on commercial banks’ ROA. The results are consistent with
baseline regressions. As shown in Table 4, it was noted that R2 remained around 0.5, which is lower
than that of the baseline regressions, indicating that the fitness of the mixed regression model is
slightly inferior to the fixed-way effect model. Standards errors in the models are relatively high as
well. For instance, standard errors are 0.0063 in model 6, slightly higher than the errors in baseline
regressions. Even if, explanatory power of ESG ratings still strong as reflected in the coefficients,
which is 0.04. It represents that one-unit upgrade in banks’ ESG rating improves their profitability
measured by ROA approximately 4% consequently, with other variables constant. Control variables
other than DTA, such as TAT, OI, IA, FA, can significantly explain variations in operating
performance of commercial banks in China.
From perspectives of corporate governance, ESG engagement helps to normalize decision-
making processes and restrict misconducts within organizations [1; 19], which improves investment
efficiency and eliminates unnecessary costs. Hence, it facilitates the maximization of commercial
banks’ operating performance.
Tobit model, an essential method in econometrics, aims to effectively assess the relation between
omitted or truncated continuous independent variables and dependent variables. The study
constructed six groups for Tobit models to investigate the impacts of ESG ratings on operating
performance, through including five control variables incrementally. Table A5 describes the
regression outputs of Tobit models. It can be found that the banks improve ESG ratings for one unit,
leading to higher ROA about 2.6%, with other variables unchanged. And some control variables hold
significance for ROA.
Explained variables substitution is the third robustness test through replacing ROA with ROE.
This study chooses ESG ratings as core explanatory variables and ROE as the modified dependent
variable. Then, it retakes fixed-effects regression, mixed regression, and Tobit model, to validate the
empirical findings of baseline regression. Table A6 presents result of the test, which reaches out R2
values are 0.857 and 0.736. The finding demonstrates that the model’s fitness is enhanced, and the
outcomes are more dependable than those of the original model. ESG ratings impose positive
implications on commercial banks’ operating performance since the coefficients are positive and
significant at the 10%, 5% and 1%.
This paper introduces the general Generalized Method of Moments (GMM) model and the
Differential GMM model to address endogeneity bias in the model, which aims to achieve results that
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

are close to unbiased estimation and enhance the reliability of the constructed model. Three models
differing in their control variables are formed in the GMM model construction. There are six
regression models were constructed to assess the impact of each variable on the operating
performance of commercial banks. As shown in Table A7, it demonstrates that ESG ratings still
positively influence on ROA in the two GMM models and the five variables have varying degrees of
explanatory power with respect to the operating performance.

5.3. Heterogeneity Analysis

This paper examines the relationship between ESG ratings and the operating performance of Chinese
commercial banks with varying characteristics. To this end, the study employs bank nature, leverage, the
largest shareholder's ownership, and firm size as threshold variables for heterogeneity analysis.
Table A8 presents regression outputs based on differences in ownership nature. The sample is
divided into two groups in terms of ownership nature, one is state-owned commercial banks, and
another is join-stock commercial banks. The results show that the coefficient of ESG ratings in state-
owned banks is 0.03, significant at 10%. That’s to say, one unit upgrade in ESG rating brings around
2% higher in operating performance of state-owned banks. Whereas, in joint-stock banks group, ESG
rating cannot significantly influence on operating performance of the banks.
The rationale for this is that the financial regulation of large state-owned commercial banks is
more stringent than that of joint-stock commercial banks. The banks can proactively address the call
of the central financial work conference through the implementation of a green finance policy. At the
same time, the banks play a leading role in financial institutions and need to maintain social
reputation through full engagement of ESG practices. This helps to standardize business models and
decision-making process, thus thereby contribute to improve operating performance. On the other
hand, as stated in the results it is not important for joint-stock commercial banks improve their
operating performance by maintaining good ESG performance. The rationale is that joint-stock banks
are inclined to preserve shareholders’ interests and maximize their own interests, rather than
fulfilling their obligations in the fields of corporate governance, environmental protection and social
responsibility. The empirical results provide clear guidance for policymakers and regulators who
should implement measures to strengthen the information disclosure on ESG performance of joint-
stock commercial banks and encourage to participate ESG - oriented governance.
This section examines the impacts of differences in leverage ratio on the relationship between
ESG ratings and operating performance of commercial banks. The threshold variable value is 6.56.
Table A9 presents the outcomes of heterogeneity analysis based on leverage ratio. The evidence finds
that in high leveraged banks, there is a positive relationship between ESG rating and ROA as reflected
in the coefficient, indicating ESG rating can significantly explain the operating performance at 1%,
5% and 10% level. In contrast, for banks with below-average leverage, ESG ratings as an explanatory
variable perform poorly and don't hold significance. Specifically, debt to asset ratio is significantly
correlated to return on asset in high leveraged banks. The rationale is that leverage ratio and debt-to-
asset ratio are measures of debt. Higher indicators can regulate decision-making process and limit
misconducts in banks, which helps to cut off unnecessary costs and enhance investment efficiency in
operations, consequently improving operating performance.
The section examines the relationship between the largest shareholders’ shareholding ratio, ESG
ratings and operating performance. Table A10 presents ESG ratings are positively related to
commercial banks’ operating performance, whose threshold variable is above sample mean.
However, ESG ratings have no explanatory power for the ROA of commercial banks in the below-
average sample. In addition, debt to asset ratio, total asset turnover ratio and operating income
growth rate can significant and positive implications on the operating performance of Chinese
commercial banks.
The phenomena can be elaborated by [34], that precise signals of efforts are required by the
largest shareholder under concentrated share ownership, which can receive better monitor and bring
values in the firm. It means that the largest shareholders need more accurate signals before acting.
Therefore, controversies among ESG concepts are minimal in banks with concentrated share
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

10

ownership and agency costs are reduced in a large degree. The likelihood of ESG practices thus has
been maximized and it brings more bank values.
The sample banks have been classified into two groups based on their average size, to undertake
the fixed effect regression. The first group consists of banks that were larger than the mean, while the
second group included those that were below the mean. The threshold variable value is 12.77. Table
A11 presents outcomes of the analysis based on bank size. The results indicate that there is a positive
correlation between the ESG rating and operating performance of the bank, especially in banks with
large size. It implies that one unit of increase in ESG rating can bring higher returns about 2.76%. But
the relation is limited in small or medium - scale banks.
The main rationale is that a large scale signifies that the banks possess sufficient assets and cash
flows to bear operating expenses, for the purpose of improving ESG performance, except for other
costs such as: administrative expenses, costs of goods sold. Sustainability practices are positively
related to enhanced social sustainability performance and better firm financial performance [23].
Thus, banks institutions which engages ESG practices are likely to receive better financial
performance.

5.4. Moderation Analysis

Table A12 presents outputs of moderating analysis. This section examines the moderating effect
of GDP growth rate on the relationship between ESG ratings and the operating performance of
commercial banks in China. The moderator is constructed through intersection term between ESG
ratings and GDP growth. The test involved variables including return on assets, ESG ratings, five
control variables and one intersection term, to conduct fixed effect regression models. The results
indicate that GDP growth rate, a moderator can strengthen explanatory power of ESG ratings while
it has positive moderating effects on improvement in the operating performance. The coefficient of
intersection term is positive, and the mean is 0.3%, which means that when GDP growth rate
increases one unit, ROA will improve to the same level with other variables constant. The moderator
holds statistical significance at any levels, and its standard errors are relatively small. The positive
relation between ESG performance and GDP is in align with [35], that better ESG performance
signifies better economic growth prospects in long term.
Table A13 presents regression outputs with CPI as a moderator. CPI measures the price of
domestic products, and the growth rate of CPI determines the country’s inflations. The evidence finds
that it exists a positive moderating effect between CPI and operating performance because the
coefficient of intersection term is positive in the six models, significant at 5% and 10%. In details, each
unit increase in CPI can drive ESG performance up by 0.3 percent in the full model, which has positive
impacts on the Chinese commercial banks’ operating performance. However, negative constants in
the model weaken the overall improvements in operating performance of commercial banks.
A rising CPI indicates that the domestic economy is performing well, and that domestic products’
price are becoming more expensive, which influences on operating performance of commercial banks.
When growth rate of CPI exceeds a threshold value, central bank of China introduces higher interest
rates to control climbing prices. The measure is to attract more deposits from clients and banks need
to pay higher interests on the deposits to them as return, which results in lower overall performance
of banks in China.
Table A14 presents the regression results of a fixed effect model with analyst coverage as a
moderator. To measure analyst coverage, the study employs the natural logarithm of the number of
analysts' attention plus one. Except for Model 4, the results demonstrate that the coefficients of the
intersection term exhibit a positive and statistically significant at the 1%, 5%, and 10% levels. This
suggests that analyst coverage may serve to reinforce the correlation between ESG ratings and the
operating performance of commercial banks.
It can be inferred from the evidence that there is a positive correlation between analyst coverage
and ESG ratings. This indicates that the number of analysts focusing on a particular bank may, to
some extent, prompt improvements in ESG performance [36]. This is attributable to the discipline
effect of analyst coverage. The increased attention from financial analysts has the effect of restraining
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

11

misconducts within organizations and enhancing awareness of ESG practices, which in turn
improves banks’ ESG performance. Superior banks’ ESG performance can attract more attentions
from analysts, who are willing to write analyst reports for the banks. High ESG performance is linked
with the enhanced corporate stock price synchronicity [37]. Investors tend to focus on banks with
better ESG ratings as well and purchase stocks in banking sectors with outstanding ESG performance,
which incentives to upgrade banks’ value. Ultimately, the overall operational performance of
commercial banks is enhanced.

6. Countermeasure and Suggestions

Given that the empirical outcomes, the paper proposes corresponding countermeasures from
the perspective of policy makers, banks, and investors who consists of individuals and professionals.
These countermeasures address the following points:

6.1. Countermeasures and Suggestions from the Perspective of Policymakers

For policymakers, it is recommended that ESG related policies be formed to incentive banks for
ESG engagement. In addition, they need to emphasize the importance of ESG practices and enhance
awareness of ESG performance for enterprises. And they should guide joint-stock commercial banks
to fulfil ESG practices.
For regulatory departments, it is recommended to strengthen the regulation of ESG performance
disclosure and improve information transparency in financial markets, which should be considered to
reduce negative impacts of systemic risk on enterprises’ operating performance, particular for banks.

6.2. Countermeasures and Suggestions from the Perspective of Banks

(1) Commercial banks should maintain good ESG performance and practice all aspects of ESG,
such as: social responsibility, environment protection and corporate governance for regulating
operating behaviors and improving ESG scores, which significantly enhances their operating
performance.
(2) Commercial banks whose shareholdings’ ratio of largest shareholder is below than the
industry average need to enhance awareness of ESG practices and eliminate controversies towards
ESG within organization.
(3) Banks must recognize that the implementation of ESG practices are the responsibility of the
banks themselves. They are recommended to implement green finance in depth from a long-term
perspective, instead of simply fulfilling ESG standards.
(4) Banks are suggested enhancing information disclosure on ESG practices and scores, which
can help improve information transparency on themselves, thereby reducing the degree of
information asymmetric in financial markets.
(5) Banks are encouraged enhancing ESG performance, which can effectively deal with negative
influences of tail risks in financial market.
(6) Bank managers are recommended to be ESG concepts oriented and regulate their conducts
for strengthening sustainable development of banks.

6.3. Countermeasures and Suggestions from the Perspective of Investors

6.3.1. For Individual Investors

(1) Investors should integrate ESG considerations into their investing strategies according to
investment capacity and high ESG-rated stocks can be chosen as underlying assets.
(2) Investors are encouraged practicing ESG investing, which helps to mitigate the downside
risks to stock prices in a financial market recession.
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

12

6.3.2. For Institutional Investors

(3) Investors are encouraged to undertake further research on ESG performance and develop
quantitative investment models that incorporates ESG ratings.
(4) Investors can use ESG investment concepts to optimize underlying asset pool by means of
strategic asset allocation.
(5) Investors are suggested investing in banks with superior ESG performance and increasing
their weights in the investment portfolios, while reducing the weight of banks with poor ESG
performance. This leads to improve the overall performance of the investment portfolios.

Funding: This research received no external funding.

Institutional Review Board Statement: Not applicable.

Data Availability Statement: The raw data supporting the conclusions of this article will be made available by
the author on request.

Acknowledgments: The author acknowledges anonymous peer reviewers for their dedicated time and expertise.
Their valuable feedback has enhanced this research paper’s quality and credibility.

Conflicts of Interest: The author declares no conflicts of interest.

Appendix A

Table A1. Variable Summary.

Variable Names Symbol Variable Construction Type

Return on Assets ROA Net Income/Average Assets
Explained Variable
Returns on Equity ROE Net Income/Average Equity
Assign a score of 0 to 9 based on ESG Explanatory Variable
ESG Performance ESG
ratings
Debt to Asset Ratio DTA Total Debt/Total Asset * 100%
Total Asset Turnover TAT Total Sales/Average Assets
Operating Income Growth
Growth [Operating Income / OI (Previous) – 1] Control Variable
Rate
Intangible Assets Ratio ITA Intangible Assets/Total Assets
Fixed Assets Ratio FA Tangible Assets/ Total Assets
Bank Size Size Natural Logarithm of Total Assets
Leverage Ratio Lev Equity Capital/Total Capital
Number of shares held by the top ten Threshold Variable
Largest shareholders’
Top% shareholders / Total number of
shareholding
outstanding shares
Gross domestic product/GDP
GDP Growth Rate GDP%
(Previous Year)
CPI CPI Consumer Price Index Moderator
Ln (numbers of analysts tracking on a
Analyst Coverage Analyst
bank +1)
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

13

Table A2. Descriptive statistics.

Obs. Mean Median SD Min Max

ROE 210 15.5578 14.915 4.9613 6.0100 31.7900

ROA 210 0.9739 0.9760 0.2410 0.3400 1.4700

ESG 210 4.9762 6.0000 1.9428 0.0000 7.0000

DTA 210 93.2949 93.244 1.4740 90.5884 98.3502

TAT 210 0.0293 0.0290 0.0042 0.0201 0.041

Growth 210 13.5827 10.452 13.6645 -15.5968 64.4801

IA 210 0.0696 0.0460 0.0683 0.0025 0.4761

FA 210 0.6109 0.5600 0.3064 0.1743 1.7018

Lev 210 6.5629 6.6450 0.9618 3.6221 8.6912

Top 210 76.0414 74.865 18.0457 33.6912 100

Size 210 8.5954 8.6910 1.1920 0.0000 10.5868

YoY 210 7.1333 7.2336 2.2206 2.2386 10.6359

CPI 210 0.9976 0.9981 0.0226 0.9377 1.0403

Analyst 210 3.2277 3.2958 0.6472 0.0000 4.3944

Table A3. Regression outputs of fixed-effect model.

(1) (2) (3) (4) (5) (6)

0.0246** 0.0286*** 0.0240*** 0.0241*** 0.0242*** 0.0242***

ESG
（0.0094） (0.0086) (0.0058) (0.0056) (0.0057) (0.0056)

0.0578*** 0.0401*** 0.0341*** 0.0341*** 0.0346***

DTA
(0.0088) (0.0060) (0.0062) (0.0062) (0.0066)

35.3398*** 30.9510*** 30.9926*** 31.0652***

TAT
(2.3147) (2.4923) (2.4957) (2.5468)

0.0024*** 0.0024*** 0.0021***

OI
(0.0007) (0.0007) (0.0007)

0.1238 0.1938
IA
(0.1617) (0.1698)

-0.0591
FA
(0.0457)

0.8508*** -4.5634*** -3.9254*** -3.2641*** -3.2770*** -3.2888***

Constant
(0.0487) (0.8254) (0.5569) (0.5827) (0.5836) (0.5825)

R2 0.034 0.211 0.645 0.669 0.670 0.674

Id FE Yes Yes Yes Yes Yes Yes

Year FE Yes Yes Yes Yes Yes Yes

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

14

Table A4. Regression outputs of mixed regression model

(1) (2) (3) (4) (5) (6)

0.0517*** 0.0530*** 0.0376*** 0.0388*** 0.0386*** 0.0372***

ESG
(0.0080) (0.0085) (0.0069) (0.0069) (0.0068) (0.0063)

0.0052 -0.0032 -0.0083 -0.0084 0.0006

DTA
(0.0110) (0.0088) (0.0095) (0.0093) (0.0088)

32.4879*** 29.4341*** 31.7364*** 32.1948***

TAT
(3.0096) (3.2255) (3.2332) (2.9922)

0.0023*** 0.0025** 0.0040***

OI
(0.0010) (0.0010) (0.0010)

-0.5815*** -0.7389***
IA
(0.1803) (0.1690)

0.2232***
FA
(0.0377)

0.7155 0.2216 0.1347 0.6607 0.6413 -0.3485

Constant
(0.0426) (1.0414) (0.8337) (0.9009) (0.8807) (0.8318)

R2 0.1695 0.1704 0.4711 0.4700 0.4960 0.5707

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.

Table A5. Regression outputs of Tobit model

(1) (2) (3) (4) (5) (6)

0.0298*** 0.0325*** 0.0255*** 0.0256*** 0.0256*** 0.0255***

ESG
(0.0091) (0.0084) (0.0057) (0.0055) (0.0055) (0.0055)

0.0537*** 0.0381*** 0.0322*** 0.0323*** 0.0326***

DTA
(0.0088) (0.0060) (0.0062) (0.0062) (0.0062)

35.3680*** 31.0520*** 31.0607*** 31.0885***

TAT
(2.2722) (2.4386) (2.4357) (2.4300)

0.0024*** 0.0024*** 0.0022***

OI
(0.0006) (0.0007) (0.0007)

0.0839 0.1188
IA
(0.5787) (0.1663)

-0.0281
FA
(0.0441)

0.8235*** -4.2018*** -3.7474*** -3.1046*** -3.1165*** -3.1279***

Constant
（0.0601） （0.8278） （0.5555） （0.5792） (0.5787) (0.5771)

R2
*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

15

Table A6. Regression outputs of explained variables substitution

(1) Baseline (2) Mixed (3) Tobit Model

0.1907* 0.5497*** 0.2394**

ESG
(0.1056) (0.1030) (0.1033)

2.9263*** 2.4823*** 2.8780 ***

DTA
(0.1179) (0.1437) (0.1168)

346.9189*** 429.9061*** 351.6303***

TAT
(47.0384) (49.1286) (45.7105)

0.0656*** 0.0755*** 0.0678***

OI
(0.0131) (0.0160) (0.0128)

-2.4858 -12.0753*** -3.7498

IA
(3.2082) (2.7740) (3.0929)

-1.1438 3.0700*** -0.4194

FA
(0.8454) (0.6197) (0.8317)

-268.5775*** -233.4620*** -264.9642***

Constant
(10.9986) (13.6577) (10.8926)

R2 0.857 0.736

Id FE Yes

Year FE Yes

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.

Table A7. Regression outputs of endogenous tests

General GMM Differential GMM

0.0549*** 0.0362*** 0.0372*** 0.0182 0.0208** 0.0206**

ESG
(0.0078) (0.0059) (0.0068) (0.0143) (0.0090) (0.0086)

0.0143 -0.0175** 0.0006 0.0745*** 0.0419*** 0.0407**

DTA
(0.0109) (0.0076) (0.0087) (0.0288) (0.0136) (0.0192)

38.3273*** 32.1948*** 28.4013*** 29.9910***

TAT
(3.8845) (3.5403) (5.1930) (7.9934)

0.0014* 0.0040*** 0.0014 0.0009

OI
(0.0009) (0.0010) (0.0009) (0.0010)

-0.7389*** 0.4477
IA
(0.1768) (0.3737)

0.2233*** -0.0953
FA
(0.0421) (0.0696)

-0.6440 1.2992* -0.3485 -6.0722** -3.8775*** -3.7735**

Constant
(1.0332) (0.7091) (0.8079) (2.6759) (1.2278) (1.6782)
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

16

Id FE Yes Yes Yes Yes Yes Yes

Year FE Yes Yes Yes Yes Yes Yes

* Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.

Table A8. Regression outputs based on differences in the banks’ ownership nature

State-Owned Joint-Stock

0.0282* 0.0137
ESG
(0.0048) (0.0089)

0.0305*** 0.0410***
DTA
(0.0071) (0.0093)

39.4020*** 30.3006***
TAT
(2.6480) (3.6861)

0.0000 0.0025***
OI
(0.0008) (0.0010)

-0.7100*** 0.3226
IA
(0.2526) (0.2199)

-0.0321 0.0275
FA
(0.0311) (0.0995)

-3.0194*** -3.9225
Constant
(0.6417) (0.8911)

R2 0.902 0.612

Id FE Yes Yes

Year FE Yes Yes

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.

Table A9. Regression outputs based on differences in leverage ratios

Lev > 6.56 Lev < 6.56

0.0351*** 0.0012
ESG
(0.007) (0.0064)

0.0260*** -0.0241
DTA
(0.006) (0.0145)

44.1324*** 21.8888***
TAT
(2.687) (4.0619)

0.0001 0.0001
OI
(0.001) (0.0012)

-0.0997 1.5032***
IA
(0.154) (0.3952)
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

17

-0.0213 -0.4174***
FA
(0.040) (0.1521)

-2.8585*** 2.6141*
Constant
(0.516) (1.3518)

R2 0.804 0.700

Id FE Yes Yes

Year FE Yes Yes

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.

Table A10. Regression outputs based on differences in shareholdings of largest shareholders.

Top% > 35.65 Top% < 35.65

0.0446*** 0.0092
ESG
(0.0073) (0.0082)

0.0474*** 0.0270***
DTA
(0.0111) (0.0085)

21.6224*** 41.3878***
TAT
(4.2179) (3.7575)

0.0031*** 0.0002
OI
(0.0010) (0.0095)

0.1196 0.1104
IA
(0.2540) (0.3214)

0.0061 -0.0697
FA
(0.0817) (0.0624)

-4.3870*** -2.7441***
Constant
(1.0169) (0.8078)

R2 0.739 0.735

Id FE YES YES

Year FE YES YES

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.

Table A11. Regression outputs based on bank size

Size > 12.77 Size < 12.77

0.0274*** 0.0066
ESG
(0.0061) (0.011)

0.0169** 0.0439***
DTA
(0.0084) (0.013)
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

18

40.0110*** 25.7275***
TAT
(3.4476) (5.407)

0.0006 0.0017
OI
(0.0011) (0.001)

-0.3758 0.5018*
IA
(0.3328) (0.264)

-0.0239 -0.1254
FA
(0.0533) (0.130)

-1.8322** -3.9644***
Constant
(0.7519) (1.317)

R2 0.760 0.422

Id FE YES YES

Year FE YES YES

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.

Table A12. Regression outputs with GDP growth rate as a moderator

(1) (2) (3) (4) (5) (6)

-0.8349*** -0.7167*** -0.3845*** -0.2574** -0.2490* -0.2733**

ESG
(0.1636) (0.1504) (0.1060) (0.1256) (0.1271) (0.1275)

0.0084*** 0.0073*** 0.0040*** 0.0028** 0.0027** 0.0029**

ESG * GDP
(0.0016) (0.0015) (0.0010) (0.0012) (0.0012) (0.0012)

0.0528*** 0.0384*** 0.0359*** 0.0359*** 0.0366***

DTA
(0.0084) (0.0058) (0.0062) (0.0062) (0.0062)

33.4157*** 31.1806*** 31.1986*** 31.3045***

TAT
(2.2897) (2.4683) (2.4738) (2.4644)

0.0015* 0.0015* 0.0010

OI
(0.0008) (0.0008) (0.0081)

0.0739 0.1541
IA
(0.1619) (0.1688)

-0.0714
FA
(0.0446)

0.8507*** -4.0938*** -3.7026*** -3.4230*** -3.4260*** -3.4535***

Constant
(0.0456) (0.7846) (0.5408) (0.5809) (0.5822) (0.5800)

R2 0.155 0.301 0.671 0.678 0.678 0.683

Id FE Yes Yes Yes Yes Yes Yes

Year FE Yes Yes Yes Yes Yes Yes

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

19

Table A13. Regression outputs with CPI as a moderator.

(1) (2) (3) (4) (5) (6)

-0.8349*** -0.7167*** -0.3845*** -0.2574** -0.2490* -0.2733**

ESG
(0.1636) (0.1504) (0.1060) (0.1256) (0.1271) (0.1275)

0.0084*** 0.0073*** 0.0040*** 0.0028** 0.0027** 0.0029**

ESG*CPI
(0.0016) (0.0015) (0.0010) (0.0012) (0.0012) (0.0012)

0.0528*** 0.0384*** 0.0359*** 0.0359*** 0.0366***

DTA
(0.0084) (0.0058) (0.0062) (0.0062) (0.0062)

33.4157*** 31.1806*** 31.1986*** 31.3045***

TAT
(2.2897) (2.4683) (2.4738) (2.4644)

0.0015* 0.0015* 0.0010

OI
(0.0008) (0.0008) (0.0081)

0.0739 0.1541
IA
(0.1619) (0.1688)

-0.0714
FA
(0.0446)

0.8507*** -4.0938*** -3.7026*** -3.4230*** -3.4260*** -3.4535***

Constant
(0.0456) (0.7846) (0.5408) (0.5809) (0.5822) (0.5800)

R2 0.155 0.301 0.671 0.678 0.678 0.683

Id FE Yes Yes Yes Yes Yes Yes

Year FE Yes Yes Yes Yes Yes Yes

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.

Table A14. Regression outputs with analyst coverage as a moderator.

(1) (2) (3) (4) (5) (6)

-0.0349** -0.0132 0.0165 0.0254* 0.0133 0.0133

ESG
(0.016) (0.018) (0.013) (0.013) (0.013) (0.013)

0.0280*** 0.0210*** 0.0079** 0.0046 0.0095*** 0.0095***

ESG*Analyst
(0.003) (0.004) (0.003) (0.003) (0.003) (0.003)

0.0281*** 0.0337*** 0.0311*** 0.0322*** 0.0321***

DTA
(0.011) (0.008) (0.007) (0.007) (0.0321)

30.7041*** 28.4123*** 24.4019*** 24.3839***

TAT
(2.349) (2.418) (2.408) (2.421)

0.0022*** 0.0018** 0.0018**

OI
(0.001) (0.001) (0.001)

0.2155 0.2119
IA
(0.158) (0.162)

FA 0.0050
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

20

(0.044)

0.6869*** -1.9185* -3.2767*** -2.9832*** -3.0128*** -3.0057***

Constant
(0.061) (0.995) (0.711) (0.702) (0.699) (0.704)

R2 0.389 0.412 0.708 0.723 0.750 0.750

Id FE Yes Yes Yes Yes Yes Yes

Year FE Yes Yes Yes Yes Yes Yes

*Notes: *, **, *** indicate that coefficients in the regression models are significant at 10%, 5% and 1% respectively;
Values in parentheses are standard errors.

References
1. Li, D.; Ma, C.; Yang, J.; Li, H. ESG performance and corporate fraud. Finance Research Letters 2024, 62, 105212.
https://doi.org/10.1016/j.frl.2024.105212.
2. He, F.; Du, H.; Yu, B. Corporate ESG performance and manager misconduct: Evidence from China.
International Review of Financial Analysis 2022, 82, 102201. https://doi.org/10.1016/j.irfa.2022.102201.
3. Lian, Y.; Weng, X. ESG performance and investment efficiency 2024. Finance Research Letters, 62, 105084.
https://doi.org/10.1016/j.frl.2024.105084.
4. He, F.; Ding, C., Yue, W.; Liu, G. ESG performance and corporate risk-taking: Evidence from China.
International Review of Financial Analysis 2023, 87, 102550. https://doi.org/10.1016/j.irfa.2023.102550.
5. Gao, M.; Geng, X. The role of ESG performance during times of COVID-19 pandemic. Scientific Reports 2024,
14(1). https://doi.org/10.1038/s41598-024-52245-7.
6. Bax, K.; Bonaccolto, G.; Paterlini, S. Spillovers in Europe: The role of ESG. Journal of Financial Stability 2024,
72. https://doi.org/10.1016/j.jfs.2024.101221
7. Broadstock, D. C.; Chan, K.; Cheng, L. T. W., & Wang, X. The role of ESG performance during times of
financial crisis: Evidence from COVID-19 in China. Finance Research Letters 2021, 38, 101716.
https://doi.org/10.1016/j.frl.2020.101716.
8. Zheng, S.; Zhang, Q.; Zhang, P. Can customer concentration affect corporate ESG performance? Finance
Research Letters 2023, 58, 104432. https://doi.org/10.1016/j.frl.2023.104432.
9. Gu, J. Investor attention and ESG performance: Lessons from China’s manufacturing industry. Journal of
Environmental Management 2024, 355, 120483. https://doi.org/10.1016/j.jenvman.2024.120483.
10. Lin, Z. Do Analysts Care About ESG Performance? Evidence from Tracking Decisions. SSRN 2023.
11. Wu, H.; Zhang, K.; Li, R. ESG score, analyst coverage and corporate resilience. Finance Research Letters 2024,
105248.
12. Zheng, Y.; Wang, B.; Sun, X.; Li, X. ESG performance and corporate value: Analysis from the stakeholders’
perspective. Frontiers in Environmental Science 2022.
13. Zhao, X.; Cai, L. Digital transformation and corporate ESG: Evidence from China. Finance Research Letters
2023, 58, 104310. https://doi.org/10.1016/j.frl.2023.104310.
14. Zeng, Y.; Zhao, X.; Zhu, Y. Equity incentives and ESG performance: Evidence from China. Finance Research
Letters 2023, 58, 104592. https://doi.org/10.1016/j.frl.2023.104592.
15. Barros, V.; Verga Matos, P.; Miranda Sarmento, J.; Rino Vieira, P. M&A activity as a driver for better ESG
performance. Technological Forecasting and Social Change 2022, 175, 121338.
https://doi.org/10.1016/j.techfore.2021.121338.
16. Niu, B. Government environmental protection expenditure and national ESG performance: Global
evidence. Innovation and Green Development 2024, 3(2), 100117. https://doi.org/10.1016/j.igd.2023.100117.
17. DasGupta, R. Financial performance shortfall, ESG controversies, and ESG performance: Evidence from
firms around the world. Finance Research Letters 2022, 46, 102487. https://doi.org/10.1016/j.frl.2021.102487.
18. Luo, S.; Yu, S.; Zhou, G. Does green credit improve the core competence of commercial banks? Based on
quasi-natural experiments in China. Energy Economics 2021, 100, 105335.
https://doi.org/10.1016/j.eneco.2021.105335.
19. Lian, Y.; Gao, J.; Ye, T. How does green credit affect the financial performance of commercial banks? ——
Evidence from China. Journal of Cleaner Production 2022, 344, 131069.
https://doi.org/10.1016/j.jclepro.2022.131069.
Preprints.org (www.preprints.org) | NOT PEER-REVIEWED | Posted: 28 August 2024 doi:10.20944/preprints202408.1978.v1

21

20. Zhou, G.; Sun, Y.; Luo, S.; Liao, J. Corporate social responsibility and bank financial performance in China:
The moderating role of green credit. Energy Economics 2021, 97, 105190.
https://doi.org/10.1016/j.eneco.2021.105190.
21. Dong, J.; Yin, L.; Liu, X.; Hu, M.; Li, X.; Liu, L. Impact of internet finance on the performance of commercial
banks in China. International Review of Financial Analysis 2020, 72, 101579.
https://doi.org/10.1016/j.irfa.2020.101579.
22. Chen, X.; You, X.; Chang, V. FinTech and commercial banks’ performance in China: A leap forward or
survival of the fittest? Technological Forecasting and Social Change 2021, 166, 120645.
https://doi.org/10.1016/j.techfore.2021.120645.
23. Sroufe, R.; Gopalakrishna-Remani, V. Management, Social Sustainability, Reputation, and Financial
Performance Relationships: An Empirical Examination of U.S. Firms. Organization & Environment 2019,
32(3), 331–362. https://doi.org/10.1177/1086026618756611.
24. Chen, S.; Song, Y.; Gao, P. Environmental, social, and governance (ESG) performance and financial
outcomes: Analyzing the impact of ESG on financial performance. Journal of Environmental Management
2023, 345, 118829. https://doi.org/10.1016/j.jenvman.2023.118829.
25. Zhou, G.; Liu, L.; Luo, S. Sustainable development, ESG performance and company market value:
Mediating effect of financial performance. Business Strategy and the Environment 2022, 31(7), 3371–3387.
https://doi.org/10.1002/bse.3089.
26. Weixiang, Y.; Yu, Z.; Defeng, M. Study on the Influence of ESG Rating on the Financial Performance of
Listed Companies. Journal of Nanjing Audit University 2023, 71-80.
27. Azmi, W.; Hassan, M. K.; Houston, R.; Karim, M. S. ESG activities and banking performance: International
evidence from emerging economies. Journal of International Financial Markets, Institutions and Money 2021,
70, 101277. https://doi.org/10.1016/j.intfin.2020.101277.
28. Tommaso, C. D.; Thornton, J. Do ESG scores effect bank risk taking and value? Evidence from European
banks. Corporate Social Responsibility and Environmental Management 2020, 1-13. https://doi:10.1002/csr.1964.
29. Meng, T.; Yahya, M. H.; Ashhari, Z. M.; Yu, D. ESG performance, investor attention, and company
reputation: Threshold model analysis based on panel data from listed companies in China. Heliyon 2023, 9,
10. https://doi.org/10.1016/j.heliyon.2023.e20974
30. Nguyen, T.-T.; Nguyen, H.-G.; Lee, J.-Y.; Wang, Y.-L.; Tsai, C.-S. The consumer price index prediction using
machine learning approaches: Evidence from the United States. Heliyon 2023, p. e20730.
https://doi.org/10.1016/j. Heliyon. 2023.e20730
31. Breuer, M.; Dehaah, E. Using and Interpreting Fixed Effects Models. Journal of Accounting Research 2024, pp.
1183-1226. https://doi.org/10.1111/1475-679X.12559.
32. Ullah, S.; Akhtar, P.; Zaefarian, G. Dealing with endogeneity bias: The generalized method of moments
(GMM) for panel data. Industrial Marketing Management 2018, 71, 69–78.
https://doi.org/10.1016/j.indmarman.2017.11.010
33. Aydoğmuş, M.; Gülay, G.; Ergun, K. Impact of ESG performance on firm value and profitability. Borsa
Istanbul Review 2022, pp. 119-127. https://doi.org/10.1016/j.bir.2022.11.006
34. Huddart, S. The Effect of a Large Shareholder on Corporate Value. Management Science 1993, pp. 1407-1421.
https://doi.org/10.1287/mnsc.39.11.1407.
35. Diaye, M.-A.; Ho, S.-H.; Oueghlissi, R. ESG performance and economic growth: a panel co-integration
analysis. Empirica 2022, pp. 99-122. https://doi.org/10.1007/s10663-021-09508-7
36. Zhang, C.; Wu, X. (2023). Analyst coverage and corporate ESG performance. Sustainability 2023,15(17),12763.
https://doi.org/10.3390/su151712763
37. Hu, J.; Zou, Q.; Yin, Q. (2023). Research on the effect of ESG performance on stock price synchronicity:
Empirical evidence from China's capital markets. Finance Research Letters 2023, 55, p. 103847.
https://doi.org/10.1016/j.frl.2023.103847

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those
of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s)
disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or
products referred to in the content.