Received: 11 December 2017 Revised: 25 June 2018 Accepted: 13 July 2018

DOI: 10.1002/bse.2224

RESEARCH ARTICLE

Do environmental, social, and governance activities improve

corporate financial performance?
Jun Xie1 | Wataru Nozawa1 | Michiyuki Yagi2 | Hidemichi Fujii3 |
4
Shunsuke Managi

1
Department of Urban and Environmental
Engineering, School of Engineering, Kyushu Abstract
University, Fukuoka, Japan This study investigated the relationship between corporate efficiency and corporate
2
Center for Social Systems Innovation, Kobe
sustainability to determine whether firms concerned about environmental, social, and
University, Kobe, Japan
3
Faculty of Economics, Kyushu University,
governance (ESG) issues can also be efficient and profitable. We applied data
Fukuoka, Japan envelopment analysis to estimate corporate efficiency and investigated the nonlinear
4
Urban Institute and Department of Urban relationship between corporate efficiency and ESG disclosure. Evidence shows that
and Environmental Engineering, School of
Engineering, Kyushu University, Fukuoka, corporate transparency regarding ESG information has a positive association with
Japan corporate efficiency at the moderate disclosure level, rather than at the high or low
Correspondence
disclosure level. Governance information disclosure has the strongest positive linkage
Shunsuke Managi, Urban Institute and
Department of Urban and Environmental with corporate efficiency, followed by social and environmental information disclosure.
Engineering, School of Engineering, Kyushu
Moreover, we explored the relationship between particular ESG activities and
University, 744 Motooka, Nishi‐ku, Fukuoka
819‐0395, Japan. corporate financial performance (CFP), including corporate efficiency, return on assets,
Email: [email protected]
and market value. We found that most of the ESG activities reveal a nonnegative
Funding information
Japan Society for the Promotion of Science relationship with CFP. These findings may provide evidence about voluntary corporate
(JSPS) KAKENHI, Grant/Award Number: social responsibility strategy choices for enhancing corporate sustainability.
JP26000001; 4th Environmental Economics
Research Fund of the Ministry of Environ-
KEY W ORDS
ment, Japan
corporate efficiency, corporate social responsibility, data envelopment analysis, ESG activities, ESG
disclosure, stakeholder engagement, sustainable development

1 | I N T RO D U CT I O N To date, a large amount of empirical literature has examined the

relationship between corporate financial performance (CFP) and
In the past several decades, a profound trend has begun to emerge in corporate sustainability to identify the implications of stakeholder‐
corporate sustainability, from voluntary engagement in sustainable oriented management for CFP (Brower & Rowe, 2017; Deng, Kang,
activities to de facto requirements due to both social expectations and & Low, 2013; Garcia, Mendes‐Da‐Silva, & Orsato, 2017; John, 1992;
regulatory pressure (Brockett & Rezaee, 2012). The number of firms that Margolis, Elfenbein, & Walsh, 2009; Orlitzky, Schmidt, & Rynes,
employ sustainability strategies and disclose environmental, social, and 2003; Z. Wang & Sarkis, 2017). Regarding this relationship, two views
governance (ESG) information continues to increase, which has caused prevail in the literature. On the one hand, according to the Porter
fundamental changes to occur in business models and management the- hypotheses (Ambec, Cohen, Elgie, & Lanoie, 2013; Porter & van der
ory. Conventional shareholder‐oriented management (Friedman, 1970) Linde, 1995), corporate social responsibility (CSR) activities, especially
aims to enhance financial performance and maximize the benefits of environmental activities, are a source of innovation that creates extra
shareholders. In contrast, sustainable business promotes stakeholder‐ revenue that can cover the additional costs, so an appropriate CSR
oriented management (Freeman & McVea, 2001), which takes all the strategy can be positively related to CFP. The majority of empirical
stakeholders, including shareholders, consumers, customers, communi- studies show a positive linkage between CFP and CSR (Deng et al.,
ties, and other related groups, into consideration and focuses on reduc- 2013; John, 1992; Margolis et al., 2009; Orlitzky et al., 2003). On
ing externalities and maximizing social value regarding ESG issues. the other hand, CSR activities bring additional costs caused by agency

Bus Strat Env. 2018;1–15. wileyonlinelibrary.com/journal/bse © 2018 John Wiley & Sons, Ltd and ERP Environment 1
2 XIE ET AL.

problems and inefficient resource allocation, which will put the firm at policy, equal opportunity policy, and independent directors. We used
an unfavorable position in the free and competitive market (Friedman, a relatively large sample that covers major countries worldwide and
1970; Sternberg, 1997). A summary of empirical studies found that 11 sectors from the Bloomberg ESG database. We attempt to
negative relationships also exist between corporate social perfor- determine how ESG information transparency is related to corporate
mance and financial performance (Margolis & Walsh, 2003). In addi- efficiency, whether the disclosure level has different effects on
tion to positive and negative relationships, a neutral relationship has corporate efficiency and which types of activities would have a
also been found between CFP and CSR (McWilliams & Siegel, 2001; positive effect on CFP.
Moore, 2001). We have contributed to the existing literature in three aspects.
The uncertainty in the findings leaves unresolved questions. The First, we used corporate efficiency as an indicator of CFP to avoid
reasons for the failure to reach a consensus regarding the implications the problem of a single metric. Data envelopment analysis (DEA)
of CSR for CFP are related to several aspects. First, the wide range of allows us to estimate corporate efficiency by considering all the
CFP metrics makes it difficult to determine a general relationship productive factors. Not only revenue and assets, which are the
regarding this issue. Market‐based metrics are the most frequently frequently used CFP metrics, but also labor and cost of goods sold
applied approaches (Peloza, 2009), which include share price, mutual (COGS) are included in the DEA model. This comprehensive
fund returns, and market value (Fatemi, Glaum, & Kaiser, 2016). CFP evaluation of CFP includes more possible CSR implications than a
is widely assessed by accounting‐based metrics, such as return on single metric.
assets (ROA) and return on equity (Ferrell, Liang, & Renneboog, Second, the nonlinear relationship between corporate efficiency
2016; Lee, Cin, & Lee, 2016). Choosing the appropriate mechanism and ESG disclosure was investigated by nonparametric regression
to meaningfully support corporate decision making and policy analysis and piecewise linear regression analysis. We found that a moderate
can be problematic. The frequently used single metric fails to capture disclosure level is positively related to corporate efficiency, rather
potential CSR implications. than a high or low disclosure level.
Second, a variety of CSR concepts and categories complicate Third, we applied the same pattern of analysis to the components
drawing conclusions about the CSR–CFP relationship. CSR is a of ESG issues. We compared the three main aspects of corporate
multidimensional construct (Waddock & Graves, 1997). Firms have sustainability in terms of the differences in both information
different motivations for practicing CSR strategies and use a wide disclosure and activities and found that governance disclosure has
range of resources, which lead to different CFP implications. Studies the strongest positive effect on corporate efficiency. We tested the
that focus on different types of CSR activities may reach different effect of 26 ESG activities, including 11 environmental activities, 6
conclusions about the CSR–CFP relationship. social activities, and 9 governance activities, on corporate efficiency.
Third, as Barnett and Salomon (2006) have noted, the relationship Moreover, a comparison with ROA and market value was used to test
between CFP and CSR is neither strictly positive nor strictly negative. the consistency of the corporate efficiency indicator.
There may have been problems with the previous studies on this issue Figure 1 shows the research framework of this study. The research
that simply concluded a linear relationship between CFP and CSR. design involves two steps. In the first step, we estimated corporate effi-
Studies that have found an inverted U‐shaped relationship between ciency by using the DEA model, with revenue as output and fixed assets,
CFP and corporate environmental performance also provide evidence numbers of employees, and COGS as inputs. In the second step, we
for the nonlinear relationship (Fujii, Iwata, Kaneko, & Managi, 2013; attempted to clarify the CSR–CFP relationship. First, the nonlinear rela-
Trumpp & Guenther, 2017). tionship between corporate efficiency and ESG disclosure was investi-
In this study, we classified CSR strategy into two categories, ESG gated. Then, we explored which types of activities would have a
information disclosures and ESG activities, such as green building positive effect on corporate efficiency, ROA, and market value.

FIGURE 1 Research framework of this

study. COGS: cost of goods sold; DEA: data
envelopment analysis; ESG: environmental,
social, and governance; OLS: ordinary least
squares
XIE ET AL. 3

The paper proceeds as follows. Section 2 summarizes the relevant disclosure is positively related to CFP. Clarkson, Li, Richardson, and
literature on the relationship of CFP with ESG disclosures and Vasvari (2011) found that environmental disclosure improved CFP in
activities. On the basis of the previous studies, we propose several polluting industries in the United States. In addition to environmental
hypotheses about ESG. Section 3 explains the methodology, including disclosure, governance disclosure also benefits the principal. Larger
the DEA model and the nonparametric and parametric regression firms that introduce more stringent disclosure regimes will employ
models. Section 4 describes the sample and the variables used in this more capable managers, which leads to better financial performance
study. Section 5 presents the empirical results and discussion. Finally, (Hermalin & Weisbach, 2012). Corporations that publish an ESG
we conclude in Section 6. report have a higher stock market return (Weber, 2014). Higher ESG
transparency boosts firm value (Yu, Guo, & Van Luu, 2018), and
integrated ESG reporting leads to a higher market valuation than
stand‐alone reporting (Mervelskemper & Streit, 2017). Conversely,
2 | L I T E R A T U R E RE V I E W A N D H Y P O T H E S E S
Qiu, Shaukat, and Tharyan (2016) did not find any relation between
environmental disclosure and CFP but found that social disclosure
2.1 | CFP and ESG disclosure was positively related to CFP. However, negative relationships have
Several theories explain the reasons why firms voluntarily disclose also been found between information disclosure and CFP (Lorraine,
ESG information. First, the legitimacy theory perspective argues that Collison, & Power, 2004; Santos & Escanciano, 2002).
ESG disclosure is aimed at gaining social legitimacy for environmental Against this backdrop, even though mixed relationships exist, it is
or social impacts caused by the firm's operation. A firm's ESG legiti- important to note that previous studies have seldom investigated the
macy is a strong motive for ESG disclosure (Lokuwaduge & nonlinear relationship between corporate efficiency and ESG disclo-
Heenetigala, 2017). Furthermore, under the pressure of social media sure. They also failed to consider the three types of disclosures
and the attention of stakeholders, ESG disclosure is a tool for impres- together to compare the differences among them. One exception is
sion management that can maintain and enhance corporate reputation the study by Nollet, Filis, and Mitrokostas (2016). They argued that
(Brammer & Pavelin, 2008; Higginson, Simmons, & Warsame, 2006; ESG disclosure possessed a quadratic relationship with CFP and found
Patten, 1992). Second, information asymmetry between firm man- a U‐shaped relationship between governance disclosure and CFP.
agers and outside investors is another motivation for voluntary corpo- Consistent with this finding, we further investigate how corporate
rate disclosure (Shane, Spicer, Shane, & Spicer, 1983). The lemon efficiency is related to ESG disclosure and distinguish the relationships
problem arises from information asymmetry between managers and at different disclosure levels. As such, we hypothesize the following.
outside investors. Investors will undervalue good performing compa-
Hypothesis 1a. Corporate efficiency presents a nonlin-
nies and overvalue bad performing companies based on the limited
ear relationship with overall ESG disclosure, and a posi-
information. Thus, the capital market will fail to optimally allocate
tive relationship exists at a certain disclosure level.
resources. One solution for the lemon problem is to send reliable sig-
nals to the market to diminish the information asymmetry between Hypothesis 1b. Corporate efficiency has a nonlinear
managers and outside investors. Socially responsible investment has relationship with the ESG components of disclosures.
increased dramatically worldwide in the past decades. Investors with Among the three components of ESG disclosure,
intrinsic social preferences are more likely to hold socially responsible governance disclosure has the strongest effect on corpo-
funds, rather than conventional funds of higher financial performance rate efficiency.
(Riedl & Smeets, 2017). Thus, corporations have become more prone
to disclosing ESG information. Third, corporate information disclosure
is a discretionary‐based disclosure, which reflects managerial discre-
2.2 | CFP and ESG activities
tion regarding how much information managers choose to disclose ESG activities include both managerial discretion and compulsory obli-
(Verrecchia, 1990, 2001). There is a trade‐off faced by managers gations. The relationship between CFP and CSR strategy choices has
regarding to what extent they should disclose corporate information, been extensively discussed in previous studies. In terms of environ-
favorable or unfavorable, to value the firm in a rational market. Firms mental activities, the traditional view argues that environmental regu-
(the seller) prefer to disclose favorable information and withhold unfa- lations are an additional cost to the company that reduce profitability
vorable information to enhance their evaluation in the market. On the and lead to low efficiency. Conversely, the Porter hypothesis argues
other hand, outside investors (the buyer) discount the firm's value by that stringent, but flexible, environmental regulation may stimulate
considering the undisclosed information as unfavorable information. companies to innovate technologically or managerially. These innova-
This compels firms to reveal more information even though it may tions create efficiencies that offset the additional costs and, in the
be unfavorable. future, generate more revenue. However, whether environmental
As the theories above explain, ESG disclosure may bring firms a regulations can stimulate additional revenue that exceeds the addi-
competitive advantage in a rational market. However, previous tional cost and lead to higher corporate efficiency remains to be
empirical studies have found a mixed relationship between CFP and determined (Ambec et al., 2013). Another concern is that firms may
ESG disclosure. Aerts, Cormier, and Magnan (2008) found that more also symbolically conform to environmental policies without actual
environmental disclosure could be translated to precise earnings efforts to pursue the environmental goals (Haque & Ntim, 2018).
forecasts by analysts, which implies that improved environmental Empirically, adopting stringent global environmental standards will
4 XIE ET AL.

lead to much higher market value compared with less stringent regula- θ* ¼ minθ subject to
tions (Dowell, Hart, & Yeung, 2000). In a high growth industry, positive Σnj¼1 λj xrj ≤ θ xi0 ; i ¼ 1; 2; …; m;
links between environmental performance and profitability will be Σnj¼1 λj yrj ≥ yr0 ; r ¼ 1; 2; …; s; (1)
strengthened (Michael & Paul, 1997). According to the previous stud- Σnj¼1 λj ¼ 1
ies discussed above, we propose that only policies that cut costs or λj ≥ 0; j ¼ 1; 2; …; n:
bring extra revenue will be positively related to corporate efficiency.
In Equation (1), ∑nj¼1 λj ¼ 1 indicates the variable returns to scale fea-
Hypothesis 2. Environmental activities that are cost ture. θ* is the input‐oriented efficiency of each decision‐making unit
cutting are positively related to corporate efficiency. and ranges from 0 to 1. If θ* = 1, then the input cannot be reduced,
which implies that the company's performance is at the optimal level.
In terms of social activities, good corporate social performance,
If θ* < 1, then the input should be reduced by θ* to become efficient,
such as corporate charitable giving, will have a positive effect on
which indicates that the firm is less efficient than its peers.
financial performance in the long run (Brammer & Millington, 2008).
Moreover, these long‐term benefits could be stronger in the sectors
that depend more on reputation, brands, and large quantities of 3.2 | Regression model
natural resources (Eccles, Ioannou, & Serafeim, 2014). However, social On the basis of the corporate efficiency score estimated by the DEA
activities bring the same concerns regarding additional costs as model above, we specified three regression models to investigate
environmental activities. Thus, we propose Hypothesis 3. the relationship between corporate efficiency and CSR strategies.
First, we used nonparametric regression to investigate the rela-
Hypothesis 3. Social activities that enhance corporate
tionship between corporate efficiency and ESG disclosure score. In
reputation and attract capable employees with little addi-
the regression model, Effi is the corporate efficiency and xi is disclo-
tional cost are positively related to corporate efficiency.
sure score, which includes the overall ESG disclosure score and the
In terms of governance activities, many studies explored the rela- components E, S, and G of the disclosure score, as shown in
tionship between board structure and CFP. Positive relationships were Equation (2). The Epanechnikov kernel estimator was used to smooth
found for several aspects. In the case of independent directors, Luan the regression result.
and Tang (2007) found that having an independent director has a sig-
nificantly positive impact on a firm's performance. Zhu, Ye, Tucker, Eff i ¼ mðxi Þ þ εi ; i ¼ 1; …; N: (2)

and Chan (2016) found that, in China, empowering independent direc-

Second, to validate the nonparametric regression result and examine
tors is positively associated with firm value. Regarding the issue of
to what degree the ESG disclosure scores are related to corporate effi-
women on the board, many studies found that board gender diversity
ciency, we further divided the ESG disclosure score into quarters (low,
may improve CFP (Erhardt, Werbel, & Shrader, 2003; Hoobler,
lower‐middle, upper‐middle, and high disclosure levels) to test the
Masterson, Nkomo, & Michel, 2016; Terjesen, Couto, & Francisco,
slope of each interval by using piecewise linear regression. Xi repre-
2016; S.‐Y. Wang, 2012). In certain industries, such as the banking
sents the firm characteristics.
industry, good corporate governance also has a positive effect on
CFP (Esteban‐Sanchez, de la Cuesta‐Gonzalez, & Paredes‐Gazquez, Eff i ¼ β0 þ βi Disclosurei þ αi Xi þ μi (3)
2017). Following the previous studies, we further added global sus-
tainable governance principles as a factor in our analytical model to Third, ordinary least squares regression was applied to investigate the

investigate how governance activities are linked with corporate effi- relationship of corporate efficiency with ESG activities. The coeffi-

ciency. We propose the following hypothesis. cients βi explain how corporate efficiency is related to ESG activities.

Hypothesis 4. Governance activities that involve an Eff i ¼ β0 þ βi ESGi þ αi Xi þ μi : (4)

independent and diversified board of directors are posi-
tively related to corporate efficiency. ROAi ¼ β0 þ βi ESGi þ αi Xi þ μi : (5)

Tobin′ s Qi ¼ β0 þ βi ESGi þ αi X i þ μi : (6)

3 | METHODOLOGY
We mainly investigated how corporate efficiency is related to ESG
activities. Furthermore, we also specified models by using ROA and
3.1 | Estimation of corporate efficiency market value as dependent variables. ROA is calculated from earnings
To estimate corporate efficiency, we applied DEA, a multivariable before interest and taxes, interest expenditure, tax expenditure, and
estimation method widely used to evaluate the efficiency of total assets. We used Tobin's Q (total market value divided by total
companies and utilities. We applied the input‐oriented model (Banker, assets) as the indicator of market value and took the natural logarithm
Charnes, & Cooper, 1984) by minimizing inputs and maintaining the value to eliminate the effect of outliers (Aouadi & Marsat, 2016; Jo &
output at its current level, as shown in Equation (1), where xi denotes Harjoto, 2011). Xi represents the control variables. We added sector
inputs, including COGS, net fixed assets, and number of employees, and country dummy variables to control for both sector and country
and yr denotes the output, the revenue. features. Liang and Renneboog (2017) have noted that CSR practices
XIE ET AL. 5

TABLE 1 Descriptive statistics of main variables

Variables # of Obs. Minimum Maximum Mean SD

Net fixed assets (million U.S. dollar) 6,631 0.0020 251,605 1,943 8,001
Employee (person) 6,631 2 2,200,000 13,678 46,938
COGS (million U.S. dollar) 6,631 0.0001 365,086 3,142 11,999
Revenue (million U.S. dollar) 6,631 0.0009 485,651 4,378 15,248
ESG score 6,615 2 83 22 13
E score 3,618 1 94 22 16
S score 4,389 3 91 27 15
G score 6,615 9 75 48 7

Note. For estimating the corporate efficiency score more precisely, we selected the sample size as large as possible. After dropping missing values, the
observations for the nonparametric regression analysis are less than the sample size for the corporate efficiency estimation. The statistical description
for each sector is shown in Table S1. COGS: cost of goods sold; ESG: environmental, social, and governance.

vary significantly across countries, according to different legal origins corporate efficiency for each firm. We used revenue as the output.
that play an important role in driving financial and economic out- In terms of inputs, COGS, net fixed assets, and number of employees
comes. We also controlled for research and development (R&D) were used to estimate corporate efficiency. Given the features of the
expenditures, firm size, and financial risk. Larger firms may take advan- different sectors, corporate efficiency was calculated by sector.
tage of economies of scale to enhance revenue and market value. We Table 1 shows the descriptive statistics for all the observations. The
used the value of the natural logarithm of the number of employees as descriptive statistics for each sector are shown in Table S1.
the indicator of firm size. Firms are vulnerable to financial burden. In the second step of the regression analysis, we used the ESG dis-
Firms with high financial leverage are more likely to lose market share closure score, components of ESG disclosure scores, and ESG activity‐
and experience a negative effect on profitability and market value related data to analyze their relationship with corporate efficiency. As
(Opler & Titman, 2009). To control for financial risk, we used the shown in Table 1, the average governance disclosure score (G score) is
natural logarithm value of financial leverage (total liability divided by higher than both the environmental disclosure score (E score) and the
equity) to indicate financial risk. social disclosure score (S score). ESG‐related activities consisted of
ESG policies and initiatives. We gathered 26 ESG activities, including
11 environmental activities, 6 social activities, and 9 governance activi-
4 | D AT A ties. The variables illustrating whether the firm implemented the related
policies or initiatives are all dummy variables, for which 0 indicates that
We used the Bloomberg Environmental, Social and Governance Data- the firm does not practice the activity and 1 indicates that the firm does
base, which offers both the financial data and the ESG information of perform the activity. For governance activities, we also collected infor-
global companies. Bloomberg Professional Service provides the ESG mation regarding the percentage of women on the board, independent
information collected from firms' annual reports, sustainability reports, directors, CEO duality, and audit committee meetings. The descriptions
press releases, and third‐party research and covers information on for all ESG activities are listed in Table 2. On the basis of the sample for
board structure, such as the percentage of women on the board and corporate efficiency estimation, we selected observations by dropping
independent directors. The transparency of ESG information is indi- the missing values for each regression model. The sample size varies
cated by the index called the ESG disclosure score, in which every data as the model changes. For detailed descriptive statistics for all ESG
point is weighted in terms of importance and tailored to different activity variables, see Tables S2, S3, and S4.
industry sectors. The Bloomberg ESG disclosure score ranges from
0.1, which indicates the lowest disclosure level, to 100, for firms that
disclose every data point collected by Bloomberg (Siew, 2015).
5 | RESULTS AND DISCUSSIONS
We gathered financial data and ESG information, including the
overall ESG disclosure score, components of ESG disclosure scores,
and ESG activity‐related data of global companies. This study 5.1 | Corporate efficiency
consisted of two steps: The first step involved estimating the corpo- Figure 2 shows the distribution of the efficiency scores in each sector.
rate efficiency for each firm by using financial data. The second step The distribution of the entire sample shows a trend that the number of
involved analyzing the relationship between corporate efficiency and companies decreases as the efficiency score increases, except for the
ESG disclosure and ESG activities. companies with extremely low efficiency. The corporate efficiency dis-
In the first step, the financial data of 6,631 companies in 2015 tribution of materials is close to a normal distribution, where there are
from 74 countries1 and 11 sectors were used to estimate the many medium‐level efficiency companies, whereas the numbers of
1
companies decrease as the efficiency score becomes closer to 0 or
Most observations come from the top three economies, the United States
(2,030 obs.), China (1,042 obs.), and Japan (1,908 obs.). Among the remaining 1. The industrial, consumer discretionary, consumer staple, health
observations, European countries account for a large proportion. care, and information technology sectors present similar right‐skewed
6 XIE ET AL.

TABLE 2 Descriptions of ESG activities

Environmental activities Description

Verification type Indicates whether the company's environmental policies were subject to an independent
assessment for the reporting period.
Green building policy Indicates whether the company has taken any steps towards using environmental technologies
and/or environmental principles in the design and construction of its buildings.
Sustainable packaging Indicates whether the company has taken any steps to make its packaging more
environmentally friendly.
Environmental quality management policy Indicates whether the company has introduced any kind of environmental quality
management and/or environmental management system to help reduce the
environmental footprint of its operations.
Environmental supply chain management Indicates whether the company has implemented any initiatives to reduce the
environmental footprint of its supply chain. Environmental footprint reductions
could be achieved by reducing waste, by reducing resource use, by reducing
environmental emissions, by insisting on the introduction of environmental
management systems, etc. in the supply chain.
Climate change policy Indicates whether the company has outlined its intention to help reduce global emissions
of the greenhouse gases that cause climate change through its ongoing operations
and/or the use of its products and services.
Climate change opportunities discussed Indicates whether the management discussion and analysis and its equivalent section of
the company's annual report discuss business opportunities related to climate change.
Risks of climate change discussed Indicates whether the management discussion and analysis and its equivalent section
of the company's annual report discuss business risks related to climate change.
Emissions reduction initiatives Indicates whether the company has implemented any initiatives to reduce its
environmental emissions to air
New products—climate change Indicates whether the company has developed and/or launched products, during the
current period only, which address future impacts of climate change and/or which
mitigate customers' contributions to climate change by reduced greenhouse gas
emissions. The products may or may not be new to the market.
Energy efficiency policy Indicates whether the company has implemented any initiatives to make its
use of energy more efficient.
Social activities
Equal opportunity policy Indicates whether the company has made a proactive commitment to ensure
nondiscrimination against any type of demographic group.
Human rights policy Indicates whether the company has implemented any initiatives to ensure the
protection of the rights of all people it works with.
Training policy Indicates whether the company has implemented any initiatives to train new and existing
employees on career development, education, or skills. Training initiatives should apply to
all employee levels, not just to those employees at management level.
Employee CSR training Discloses whether the company conducts training courses for employees on CSR.
Health and Safety Policy Indicates whether the company has recognized its health and safety risks and
responsibilities and is making any effort to improve the management of employee
health and/or employee safety.
Fair remuneration policy Indicates if the company has demonstrated a group wide commitment to ensure
payment of a fair (could be defined as minimum, living, or some other criteria)
wage to all group employees, even in those countries that do not legally
require a minimum wage.
Governance activities
UN Global Compact Signatory Indicates whether the company is a signatory of the United Nations Global Compact.
GRI Criteria Compliance Indicates whether the company is in compliance with GRI criteria.
Global Reporting Initiatives Checked Indicates whether the company's application level was checked by the GRI.
Percentage of independent directors Independent directors as a percentage of total board membership.
CEO duality Indicates whether the company's CEO is also chairman of the board, as reported
by the company.
Audit committee meetings Number of meetings of the board's audit committee during the reporting period.
In the case of Japanese companies, this number will reflect the number of audit
committee meetings for those companies with a committee‐based structure or
the number of board of auditor meetings for those companies with a board
of auditor based structure.
Percentage of women on board Percentage of women on the board of directors, as reported by the company.
Executive compensation linked to ESG Indicates whether executive compensation is linked to ESG goals.
Business ethics policy Indicates whether the company has established ethical guidelines and/or a compliance policy
for its nonmanagement/executive employees in the conduct of company business.

Note. CSR: corporate social responsibility; ESG: environmental, social, and governance; GRI: Global Reporting Initiative.
XIE ET AL. 7

FIGURE 2 Corporate efficiency by sectors. Corporate efficiency of 11 sectors is estimated by using the input‐oriented DEA model. The
efficiency score ranges from 0, indicating the least efficient, to 1, indicating the most efficient. Given a sector's heterogeneity, we estimated
the efficiency score by sector. Most observations come from the United States (2,030 Obs.), Japan (1,908 Obs.), and China (1,042 Obs.)

distributions, where most of the companies are concentrated in the negative relationship at the low disclosure level (Figure 3c), similar to
low efficiency score interval. Conversely, the numbers of high effi- that for the ESG score. There is also a negative trend at the high dis-
ciency companies in the financial, utilities, and real estate sectors are closure level, but it is not reliable because of the wide confidence
much higher than those of companies with medium and low efficiency interval. As shown in Figure 3d, the G score first shows a negative
scores. The energy and telecommunication services have similar distri- relationship with corporate efficiency. At the middle disclosure level,
butions, where firms with high and medium efficiency scores dominate the positive relationship between the G score and corporate efficiency
the sector. is the strongest among all three types of disclosures.
We further used piecewise linear regression to validate the rela-
tionship between corporate efficiency and ESG disclosure. The disclo-
5.2 | Corporate efficiency and ESG disclosure sure score was divided into quarters to test the strength of the
The relationship between corporate efficiency and ESG disclosure was relationship, where the low disclosure level is from 0 to 25 points,
investigated, as illustrated in Figure 3. Assuming that there is no linear the lower‐middle disclosure level is from 25 to 50 points, the upper‐
relationship between them, we used nonparametric regression analy- middle disclosure level is from 50 to 75 points, and the high disclosure
sis to explore the relationship between corporate efficiency and dis- level is from 75 to 100 points. Moreover, the change of slope was
closure score. As shown in Figure 3a, we found a positive tested by the coefficient equality test to distinguish the relationship
relationship between corporate efficiency and overall ESG score at at different disclosure levels. The results were consistent with the
the middle disclosure level, a strong negative relationship at the high nonparametric regression analysis.
disclosure level, and a weak negative relationship at the low disclosure As shown in Table 3, the overall ESG score is negatively related to
level. Figure 3b shows that the E score also has a mixed relationship corporate efficiency when the ESG score is lower than 25 points. At
with corporate efficiency. At the low and middle disclosure levels, the lower‐middle disclosure level, the ESG score has a significant
the E score presents an ascending trend with corporate efficiency. and positive effect on corporate efficiency. At the upper‐middle dis-
When the E score is greater than 75 points, corporate efficiency tends closure level, the relationship becomes negative again but is not signif-
to decrease as the E score increases. For the S score, we found a weak icant, and at the high disclosure level, corporate efficiency shows a
8 XIE ET AL.

FIGURE 3 Relationship between corporate efficiency and ESG scores. The figures present nonparametric regression results between corporate
efficiency and ESG score (a, Obs. = 6,556), environmental (b, Obs. = 3,618), social (c, Obs. = 4,389), governance (d, Obs. = 6,615) disclosure scores,
respectively. The disclosure score ranges from 0.1 to 100, and 100 points indicate full disclosure of all the ESG items. Corporate efficiency ranges
from 0 to 1, and 1 point is the best. ESG: environmental, social, and governance [Colour figure can be viewed at wileyonlinelibrary.com]

TABLE 3 Piecewise linear fitting result of corporate efficiency and ESG disclosure score
Coefficient
Piecewise interval of
ESG disclosure score ESG score E score S score G score
[1] 0–25 −0.0010* (0.0006) 0.0034*** (0.0006) −0.0019*** (0.0006) −0.0167* (0.0088)
[2] 25–50 0.0055*** (0.0005) 0.0013** (0.0006) 0.0037*** (0.0005) 0.0051*** (0.0007)
[3] 50–75 −0.0002 (0.0021) 0.0003 (0.0017) −0.0007 (0.0014) 0.0104*** (0.0009)
[4] 75–100 −0.0200*** (0.0078) −0.0090** (0.0041) 0.0123 (0.0100) (Omitted)
R&D 0.0072*** (0.0007) 0.0061*** (0.0006) 0.0068*** (0.0006) 0.0071*** (0.0007)
Firm size −0.0033 (0.0023) 0.0246*** (0.0030) 0.0191*** (0.0028) −0.0068*** (0.0022)
Leverage 0.0373*** (0.0031) 0.0409*** (0.0046) 0.0384*** (0.0041) 0.0357*** (0.0031)
Sector YES YES YES YES
R2 0.2854 0.346 0.3166 0.3048
# of Obs. 6,424 3,552 4,284 6,424
Coefficient equality test F ‐statistics
[1]–[2] 46.91*** 3.35* 32.66*** 6.03**
[2]–[3] 5.97** 0.26 6.6** 17.02***
[3]–[4] 4.08** 3.22* 1.55 (Omitted)

Note. The disclosure scores are quartered into four equal splines, namely, low disclosure level [1], lower‐middle disclosure level [2], upper‐middle disclosure
level [3], and high disclosure level [4]. The first part of the table shows the result of the piecewise linear regression. The second part of the table lists the
result of the coefficient equality test for each adjacent spline. ESG: environmental, social, and governance; R&D: research and development. Significance
levels are indicated as follows.
*p < 0.1. **p < 0.05. ***p < 0.01.

significantly negative association with the ESG score. The nonlinear asymmetry problem may be associated with the low disclosure level.
relationship implies that the degree of ESG information transparency However, a high disclosure level may bring additional costs and more
results in different corporate efficiency effects. The information limitations for corporate operations. As a result, neither the low nor
XIE ET AL. 9

the high ESG information disclosure level has positive effects on cor- the highest, followed by the S and E scores. These results are consis-
porate efficiency. In this case, only the moderate disclosure level has tent with the nonparametric regression analysis and prove our hypoth-
a positive effect on corporate efficiency. esis that a positive relationship between ESG disclosures exists at
As for the components of the ESG score, different trends are certain disclosure levels. As shown in Table 3, this disclosure level is
found for the different types of disclosures. In terms of the E score, between 25 and 50 points. However, at the upper‐middle disclosure
at both the low and lower‐middle disclosure levels, there is a signifi- level, only the G score has a significant and positive effect on corporate
cant and positive relationship between corporate efficiency and E efficiency. The other types of disclosure have no significant relationship
score. However, the E score is negatively related to corporate effi- with corporate efficiency. At this disclosure level, the disadvantage of
ciency at the high disclosure level, and there is no significant associa- withholding information may be offset by the benefit of disclosing
tion at the upper‐middle disclosure level, which is the same trend seen information, especially for environmental and social information. Gov-
with the overall ESG score. For the S score, the results at the low and ernance information still has a positive effect on corporate efficiency
middle disclosure levels are similar to those for the overall ESG score, and is the strongest factor for companies at the upper‐middle disclo-
that is, there is a weak negative relationship at the low disclosure level sure level, which implies that the importance of operation‐related gov-
and a positive relationship at the lower‐middle disclosure level. How- ernance information is greater than that of environmental and social
ever, we did not observe a significant relationship at either the high information. On the basis of these results, we conclude that our results
disclosure level or the upper‐middle disclosure level. Because there support Hypothesis 1a and 1b regarding the nonlinear relationship and
are no observations at the high disclosure level for the G score, we the strongest effect from governance disclosure.
only observed the results of the G score at low and middle disclosure
levels. First, the G score presents a strong negative relationship with
5.3 | Corporate efficiency and ESG activities
corporate efficiency at the low disclosure level. Then, the relationship
becomes positive at the lower‐middle disclosure level. At the upper‐ Firms that adopt aggressive CSR strategies outperform their peers for
middle disclosure level, we observed a much stronger positive rela- several reasons, such as higher corporate reputation and favorable
tionship for the G score, with the highest coefficient among all three working conditions that attract productive employees (Turban &
types of disclosures. Furthermore, the coefficient equality test indi- Greening, 1997) or enhanced competitiveness. In this section, we
cates that the two positive coefficients of the ascending trend are sig- attempted to find the relationship between corporate efficiency and
nificantly different. Among all three disclosure types, either for ESG activities. Through comparison with the results of ROA and mar-
positive or negative relationships, the G score has the strongest effect ket value, we also investigated the relationship between ESG activities
on corporate efficiency compared with the E and S scores. One possi- and other financial performance indicators from the perspective of
ble explanation may be that governance information is more tightly profitability and market expectation.
related to corporate operations. Moreover, according to the research
by Eccles, Serafeim, and Krzus (2011), the global interest in gover- 5.3.1 | Environmental activities
nance information is greater than that in environmental and social As shown in Table 4, we found that the verification type is significantly
information. and positively related to corporate efficiency and market value, imply-
Overall, at different disclosure levels, mixed relationships exist ing that companies with their environmental policies subject to an
between ESG disclosure and corporate efficiency. Meanwhile, com- independent assessment tend to be more efficient and highly evalu-
mon features can also be found through the comparison. First, a low ated. The intense market competition encourages firms to employ
disclosure level presents a negative relationship with corporate effi- more strategic environmental policies to enhance competitiveness.
ciency. A lack of disclosure brings about information asymmetry, Because environmental policies have controversial impacts on CFP,
which puts firms at a disadvantaged position in a rational market. the third‐party assessment of environmental policy plays an important
One exception is environmental disclosure. Because the market is pay- role in ensuring the policy's effectiveness and efficiency. This result
ing increasing attention to environmental problems, firms are encour- implies the necessity of an independent assessment for corporate
aged to disclose environmental information, which was seldom done environmental policies.
in the past. Even a low environmental information disclosure level We also observed that a green building policy is positively associ-
enhances corporate reputation, which is associated with higher effi- ated with both corporate efficiency and ROA. The efficiency score of
ciency. Second, a high disclosure level also has a negative effect on firms that employed a green building policy is 0.036 higher than the
corporate efficiency, as observed from the overall ESG score and the score of firms that did not. Similarly, ROA is also higher if a firm takes
E score. Although environmental information disclosure is encouraged, steps towards using environmental technologies or principles in the
a high disclosure level will expose much more unfavorable informa- design and construction of its buildings. In a typical office building,
tion. When a firm attempts to maintain a high disclosure level and 30% of operating expense comes from energy consumption; it is the
good environmental performance in the long term, negative signals largest and most manageable operating expense in the provision of
will destroy the corporate reputation. Corporate efficiency may also office space (Eichholtz, Kok, & Quigley, 2010). In the past decades,
be affected by the additional cost of maintaining a high disclosure the economic return of having more energy efficient and sustainable
level. The third common feature is that for all types of disclosure, properties has stimulated the real estate market. Significant benefits
positive relationships are observed at the lower‐middle disclosure are obtained by using green building properties, which include energy
level. Among these relationships, the coefficient of the G score is conservation, and meeting the shifting preferences of both tenants
10 XIE ET AL.

TABLE 4 Relationships between environmental activities and cor- management system more smoothly and enhance the efficiency of
porate efficiency, ROA, and market value environmental quality management implementation. However, as
Corporate Market one essential strategy of an environmental quality management policy,
Variables efficiency ROA value environmental supply chain management is positively related to cor-
Verification 0.042*** −0.002 0.082** porate efficiency. Many studies have also found a positive link
type (0.012) (0.004) (0.036)
between CFP and green supply chains. As Rao and Holt (2005)
Green building 0.036*** 0.005* 0.015
explained, a green supply chain could be a source of competitiveness,
policy (0.010) (0.003) (0.033)
which can improve corporate efficiency. Unlike green building policy,
Sustainable 0.005 0.006* 0.082**
packaging (0.010) (0.003) (0.033) sustainable packaging is not significantly related to corporate effi-
Environmental quality −0.025*** 0.006 −0.116*** ciency, but firms that choose to use sustainable packaging are more
management policy (0.008) (0.004) (0.028) likely to have higher ROA and market value.
Environmental supply 0.025*** 0.000 0.010 Activities such as emissions reduction initiatives, climate change
chain management (0.009) (0.004) (0.029)
policy, and discussion of climate change opportunities and risks and
Climate change 0.005 0.001 0.028
policy (0.010) (0.004) (0.032) new products related to climate change and energy efficiency policy
Climate change 0.008 0.004 −0.146 have no significant relationship with corporate efficiency, ROA, or
opportunities (0.029) (0.007) (0.107) market value. From the point of view of stakeholder theory, the cli-
discussed
mate change issue is related to the largest number of stakeholders.
Risks of climate 0.003 −0.010 0.017
change discussed (0.014) (0.007) (0.033) However, in terms of the climate change issue, doing good does
Emissions reduction 0.006 −0.008 0.052 not mean doing well. Activities related to climate change are much
initiatives (0.012) (0.008) (0.038) less effective than activities that focus on corporate operational
New products— −0.001 0.002 0.013 problems, such as a green building policy and a green supply chain.
climate change (0.028) (0.009) (0.095)
Henderson (2015) argued that for profit‐maximizing firms, if the sus-
Energy efficiency −0.010 −0.001 0.008
policy (0.013) (0.009) (0.044)
tainable activities simply contribute to a public good, such as a cli-

R&D expenditure 0.005*** 0.000 0.001

mate change policy, this will put the firm at a competitive
(0.001) (0.000) (0.001) disadvantage. Although our result does not show a negative relation-
Firm size 0.015*** 0.008*** −0.034*** ship, it indicates that firms with climate change policies are neither
(0.003) (0.001) (0.010)
more nor less efficient.
Leverage 0.031*** −0.017*** −0.127***
According to the discussion above, as explained by the Porter
(0.005) (0.002) (0.016)
hypothesis, environmental activities that present a positive association
Sector YES YES YES
with CFP have some common features. Not all environmental policies
Country YES YES YES
will improve performance. This improvement occurs only if the policy
# of Obs. 3,604 3,393 3,560
aims to enhance environmental performance with any potential profit
R2 0.440 0.133 0.371
higher than the additional cost (Porter & van der Linde, 1995). In other
Note. The table shows how environmental activities are related to corpo-
words, our results support Hypothesis 2, which states that activities
rate efficiency, ROA, and market value. ROA: return on assets; R&D:
research and development. The significance levels are indicated as follows. that are cost cutting are positively related to corporate efficiency.
*p < 0.1. **p < 0.05. ***p < 0.01.

and investors with respect to environmental issues (Eichholtz, Kok, & 5.3.2 | Social activities
Quigley, 2013). Furthermore, firms offering more updated and envi- As revealed in Table 5, equal opportunity and training policies are both
ronmentally friendly office space are more likely to attract a better positively associated with corporate efficiency and market value.
work force (Turban & Greening, 1997). This may also result in higher Regarding corporate efficiency, the efficiency scores of firms with
efficiency and profitability. equal opportunity policies are 0.043 greater than those of other firms.
The results reveal a significant and negative relationship between The efficiency scores of firms that practice training programs are
corporate efficiency and environmental quality management policy 0.048 greater than those of firms that do not. The results are consis-
and market value. Implementing a comprehensive environmental qual- tent with an experiment‐based study indicating that trained workers
ity management strategy does not come without costs, such as pre- performed better than workers who did not participate in training
vention costs (costs of training employees), appraisal costs (costs of programs (De Grip & Sauermann, 2012). Companies that adopt
monitoring equipment), internal failure costs (costs of separating policies concerning employee career development and that offer equal
wastes), and external failure costs (costs of addressing hazardous work conditions may enhance their corporate reputation and attract
emissions) (Chandrashekar, Dougless, & Avery, 1999). When adopting more productive employees (Turban & Greening, 1997). Furthermore,
an environmental quality management policy, firms will also be many researchers have investigated the effect of on‐the‐job‐training
impeded by the costs of shifting supply chain relationships and coping on productivity and have found a positive relationship between
with moral problems (Miles & Russell, 1997). This is more likely to training programs and employee productivity (De Grip & Sauermann,
result in a negative relationship with corporate efficiency and profit- 2013; Konings & Vanormelingen, 2015), which is also supported by
ability. The challenge is how to adopt an environmental quality our results.
XIE ET AL. 11

TABLE 5 Relationships between social activities and corporate effi- TABLE 6 Relationships between governance activities and corpo-
ciency, ROA, and market value rate efficiency, ROA, and market value

Variables Efficiency ROA Market value Market

Variables Efficiency ROA value
Equal opportunity 0.043*** 0.001 0.050*
policy (0.009) (0.003) (0.029) UN Global Compact 0.029* 0.001 0.044
Signatory (0.015) (0.004) (0.041)
Human rights −0.008 −0.001 0.064**
policy (0.009) (0.004) (0.030) GRI Criteria Compliance 0.014 −0.014*** −0.003
(0.010) (0.004) (0.029)
Training policy 0.048*** −0.001 0.058*
(0.010) (0.005) (0.033) Global Reporting −0.058 0.011 −0.232
Initiatives Checked (0.053) (0.030) (0.293)
Employee −0.004 −0.005** −0.020
CSR training (0.011) (0.002) (0.036) Percentage of 0.118*** 0.003 0.258**
independent directors (0.037) (0.011) (0.113)
Health and −0.031*** −0.001 0.040
safety policy (0.011) (0.003) (0.036) CEO duality −0.015* 0.009*** −0.014
(0.001) (0.000) (0.004)
Fair remuneration 0.006 −0.006 0.141
policy (0.043) (0.018) (0.144) Audit committee 0.001 −0.001*** −0.005
meetings (0.009) (0.003) (0.029)
R&D 0.006*** 0.000 0.001
expenditure (0.001) (0.000) (0.001) Percentage of 0.061 0.051*** 0.597***
Women on Board (0.048) (0.017) (0.172)
Firm size 0.016*** 0.009*** −0.040***
(0.003) (0.001) (0.009) Executive compensation 0.007 −0.025** 0.015
linked to ESG (0.019) (0.011) (0.048)
Leverage 0.030*** −0.017*** −0.117***
(0.004) (0.002) (0.016) Business ethics 0.020 −0.001 −0.031
policy (0.020) (0.006) (0.074)
Sector YES YES YES
R&D expenditure 0.005*** 0.000 0.001
Country YES YES YES
(0.001) (0.000) (0.001)
# of Obs. 3,677 3,451 3,629
Firm size 0.022*** 0.009*** −0.012
2
R 0.435 0.129 0.367 (0.004) (0.001) (0.012)

Note. The table shows how social activities are related to corporate effi- Leverage 0.033*** −0.015*** −0.081***
(0.005) (0.002) (0.019)
ciency, ROA, and market value. CSR: corporate social responsibility;
ROA: return on assets; R&D: research and development. The significance Sector YES YES YES
levels are indicated as follows. Country YES YES YES
*p < 0.1. **p < 0.05. ***p < 0.01. # of Obs. 2,443 2,287 2,426
R2 0.447 0.186 0.345
In addition to equal opportunity and training policies, a human
Note. The table shows how governance activities are related to corporate
rights policy also has a positive relationship with market value, with
efficiency, ROA, and market value. ESG: environmental, social, and gover-
a coefficient of 0.064, which is higher than that for the equal opportu- nance; GRI: Global Reporting Initiative; ROA: return on assets; R&D:
nity and training policies. A human rights policy aims to ensure the research and development. The significance levels are indicated as follows.
protection of the rights of all people involved with a company. Even *p < 0.1. **p < 0.05. ***p < 0.01.
though the benefits are not reflected by corporate efficiency and
ROA, the market value reveals a positive relationship with this policy. investigated global standard corporate governance regulations, includ-
Although a policy that does not focus on operations but aims to create ing the United Nations Global Compact Signatory, Global Reporting
a more ideal corporate culture may have little connection with effi- Initiative (GRI) Criteria Compliance, and Global Reporting Initiatives
ciency and profitability, such a policy would be well evaluated by the Checked, we found that only the United Nations Global Compact Sig-
market. This is because it enhances corporate reputation and may indi- natory is positively related to corporate efficiency. Participants of the
rectly lead to a better workplace. United Nations Global Compact are requested to practice 10 princi-
As with environmental activities, social activities present contro- ples covering four aspects of human rights, labor, environment, and
versial relationships with CFP. Practicing social activities involves anti‐corruption. Most aspects of the 10 principles and ESG activities
costs; adopting a policy where the additional costs cannot be covered overlap, such as human rights and equal opportunity policies and
by the benefits may reduce profits and efficiency. As noted in Table 5, activities concerning environmental issues. Among these, a majority
corporate efficiency is negatively related to having a health and safety of the activities show nonnegative relationships with corporate effi-
policy. Moreover, employee CSR training does not have a positive ciency. Consistently, the relationship between the United Nations
relationship with CFP. These controversial results support Hypothe- Global Compact Signatory and the corporate efficiency is also signifi-
sis 3, which states that social activities that can enhance corporate cant and positive.
reputation and attract able employees with little additional costs are As for the other two global standards, no significant result is
positively related to corporate efficiency. found with either corporate efficiency or market value. Regarding
the GRI Criteria Compliance, companies are required to publish a sus-
tainability report to enhance ESG management and decision making.
5.3.3 | Governance activities This type of nonfinancial reporting may target long‐term management
Governance activities consist of both stakeholder‐ and shareholder‐ strategy and business plans, which is not reflected in our research.
oriented management strategies. As shown in Table 6, among the According to the results, the nonsignificant relationship implies that
12 XIE ET AL.

sustainability reporting has little linkage with financial performance. 5.3.4 | Control variables
However, the negative relationship between GRI Criteria Compliance R&D expenditure is only significantly and positively associated with
and ROA implies an obstacle to promote sustainability reporting in corporate efficiency. For ROA and market value, the coefficients of
the short run. R&D expenditure are not statistically different from 0, which is consis-
In terms of board composition, Table 6 reveals that board indepen- tent with the findings of Lee et al. (2016). Different from accounting‐
dence has the strongest positive relationship with corporate efficiency and market‐based indicators, the corporate efficiency estimated by
and presents a significantly positive relationship with market value as DEA methods is a total factor productivity indicator, reflecting the
well. From the perspective of shareholder theory, boards of directors technology of production and operations. R&D is the source of tech-
take on the responsibility to solve agency problems. In large organiza- nological advancement. Thus, R&D expenditure is supposed to be
tions, the CEO has incentives to keep his job and increase his own ben- directly related to the firm's productivity. Firm size is positively associ-
efits from the position, which may harm the interests of shareholders. ated with corporate efficiency and ROA, while having a nonpositive
The negative relationship between executive compensation linked to relationship with market value. Financial leverage is only positively
ESG and ROA indicates the concern about the agency problem. Com- related to corporate efficiency and negatively related to ROA and mar-
pensation related to nonfinancial objectives may lead to the CEO ket value. From the perspective of accounting, high financial leverage
increasing his own benefits rather than enhancing the value of share- brings about high risk, which may deteriorate profitability and market
holders. Thus, there is an endogenous requirement of the board of value (Opler & Titman, 2009). For the three specifications of the ESG
directors to bargain with the CEO. Moreover, we have also observed activity analyses, the control variables present consistent results.
that CEO duality is negatively related to corporate efficiency. The evi-
dence implies that an independent board can put the board at a more
advantageous bargaining position to replace an inefficient CEO and 6 | CO NC L U SI O N A N D P O LI C Y
enhance corporate efficiency. However, our result also shows that IMPLICATION
CEO duality has a weakly positive relationship with ROA. The empirical
evidence proved that CEO duality is a double‐edged sword regarding In this study, we applied the DEA model to estimate corporate
financial performance (Firth, Wong, & Yang, 2014). Hermalin and efficiency, a comprehensive CFP indicator, and then investigated the
Weisbach (2003) proved that CEO turnover appears to be more sensi- nonlinear relationship between corporate efficiency and ESG
tive to performance when the board is more independent. CEO duality disclosure given different disclosure levels. We further explored the
makes it difficult to replace inefficient CEOs, but it strengthens a valu- relationship of CFP, including corporate efficiency, ROA, and market
able board when the CEO performs well. By comparing the coefficients, value, with ESG activities.
the negative relationship is much stronger, implying that firms should On the basis of these analyses, we found that a moderate
be cautious regarding board composition and refrain from CEO duality disclosure level of ESG information has a significant and positive
when the CEO performs inefficiently. effect on corporate efficiency, in contrast to the high or low disclosure
As for the gender diversity of the board, the percentage of level. To be specific, corporate efficiency and ESG disclosure have a
women on the board has no significant relationship with corporate positive relationship at the middle disclosure level, where most obser-
efficiency. However, we found that the percentage of women on the vations occur. Among the different disclosures, governance disclosure
board has a very strong positive linkage with ROA and market value. has the strongest positive relationship, followed by social disclosure
The coefficients are much greater than all the other activity variables. and environmental disclosure. On the other hand, a low disclosure
The positive relationship between gender diversity and firm perfor- level has a negative association with corporate efficiency, except for
mance has been proven by many other empirical results (Post & Byron, environmental information disclosure, which reveals a weak positive
2015). Our result is also consistent with the findings of Terjesen et al. relationship with corporate efficiency. A high disclosure level also
(2016), who investigated 3,800 global companies. There is little doubt has a negative association with corporate efficiency, especially
that women on the board is the right direction and benefits female environmental disclosure. These results imply that the discretionary‐
employees realizing their potential and eliminating gender discrimina- based disclosure of ESG information has a complex association with
tion in the workplace. corporate efficiency, rather than a simple linear relationship, given
We also observed that audit committee meetings have a signifi- internal and external circumstances.
cant but extremely weak negative relationship with ROA. A business Furthermore, various kinds of CSR activities have been initiated to
ethics policy for nonmanagement or executive employees has no sig- enhance corporate sustainability in the aspects of ESG issues.
nificant relationship with financial performance. Although ethical Evidence from global companies proved that most of the ESG
guidelines for employees are good practice, the effectiveness is limited activities have nonnegative relationships with corporate efficiency,
if the policy is only implemented with nonexecutive employees. ROA, and market value. With respect to environmental activities,
According to the discussion above, several activities regarding global policies that are cost cutting, such as a green building policy, a sustain-
sustainable strategies and board composition are positively related able packaging, an environmental supply chain, or an adoption of
to CFP. The results support Hypothesis 4, which states the positive independent assessment, are positively related to CFP. No significant
effects of an independent and diversified board. Moreover, we also results are found between climate change policies and CFP. For social
conclude that the United Nations Global Compact Signatory has a activities, firms that try to reduce demographic discrimination and
positive effect on corporate efficiency. offer training programs equally tend to outperform their peers. In
XIE ET AL. 13

terms of governance activities, independent directors play an impor- Further research should explore the causal relationship between
tant role in reducing agency costs and maximizing shareholder value, CSR strategies and CFP to clarify what types of efforts firms can
which lead to better financial performance. Including women on the engage in to enhance CFP by doing good. Moreover, the response
board also has a strong positive relationship with financial perfor- to CSR activities from stakeholders is also an important factor that
mance. Moreover, firms that participate in the United Nations Global might influence the CSR strategy choice of individual firms.
Compact Signatory tend to be more efficient.
Overall, though mixed relationships are found in our results, the ACKNOWLEDGEMENTS
evidence shows that there is no doubt about the positive associa- This research was supported by Japan Society for the Promotion of
tion between corporate efficiency and CSR strategy for both ESG Science (JSPS) KAKENHI Grant number JP26000001 and the 4th
information disclosures and ESG activities. The nonlinear Environmental Economics Research Fund of the Ministry of Environ-
regression results clarify the complex relationship between corpo- ment, Japan.
rate efficiency and ESG information disclosure. The ESG activity
analyses indicate common features of activities that are positively ORCID
related to CFP.
Jun Xie http://orcid.org/0000-0001-7511-7869
On the basis of our analyses, we make the following policy recom-
Wataru Nozawa http://orcid.org/0000-0001-8075-4248
mendations. First, firms are encouraged to disclose environmental and
Michiyuki Yagi http://orcid.org/0000-0002-1315-4477
social information at a moderate level. Currently, corporate
Hidemichi Fujii http://orcid.org/0000-0002-3043-1122
governance still plays the most important role in corporate manage-
Shunsuke Managi http://orcid.org/0000-0001-7883-1427
ment, rather than environmental and social issues. To enhance market
competitiveness, disclosing more governance information is strongly
RE FE RE NC ES
recommended. Meanwhile, environmental and social information
Aerts, W., Cormier, D., & Magnan, M. (2008). Corporate environmental dis-
disclosure needs more managerial discretion. Firms that try little to closure, financial markets and the media: An international perspective.
disclose environmental and social information at a low disclosure level Ecological Economics, 64, 643–659. https://doi.org/10.1016/j.
ecolecon.2007.04.012
may face the resistance of poor CFP and fall into the low disclosure
Ambec, S., Cohen, M. A., Elgie, S., & Lanoie, P. (2013). The porter hypoth-
level and low efficiency trap. Our results suggest that one step
esis at 20: Can environmental regulation enhance innovation and
forward will potentially make a difference. A strategic and moderate competitiveness? Review of Environmental Economics and Policy, 7,
disclosure can enhance corporate efficiency, rather than a pro forma 2–22. https://doi.org/10.1093/reep/res016
disclosure with limited transparency or over disclosure with much Aouadi, A., & Marsat, S. (2016). Do ESG controversies matter for firm
unfavorable information. value? Evidence from international data. Journal of Business Ethics,
1–21. https://doi.org/10.1007/s10551‐016‐3213‐8
Second, the effective execution of ESG activities is essential to
Banker, R. D., Charnes, A., & Cooper, W. W. (1984). Some models for esti-
both CSR performance and CFP. According to the guidance on ESG‐
mating technical and scale inefficiencies in data envelopment analysis.
related risks published by the Committee of Sponsoring Organizations Management Science, 30, 1078–1092. https://doi.org/10.1287/
of the Treadway Commission and the World Business Council for mnsc.30.9.1078
Sustainable Development (COSO‐WBCSD, 2018), in the business Barnett, M. L., & Salomon, R. M. (2006). Beyond dichotomy: The curvilinear
relationship between social responsibility and financial performance.
world, the awareness of ESG‐related risks that were once considered
Strategic Management Journal, 27, 1101–1122. https://doi.org/
as uncommon issues is becoming stronger than ever. However, firms 10.1002/smj.557
can only symbolically conform to some environmental policies, such Brammer, S., & Millington, A. (2008). Does it pay to be different? An anal-
as climate change‐related policies, without actual efforts to pursue ysis of the relationship between corporate social and financial
the environmental goals (Haque & Ntim, 2018). Our results suggest performance. Strategic Management Journal, 29, 1324–1343. https://
doi.org/10.1002/smj.714
that such climate change‐related policies are also not beneficial from
Brammer, S., & Pavelin, S. (2008). Factors influencing the quality of corpo-
the perspective of financial performance. In the future, to reduce or
rate environmental disclosure. Business Strategy and the Environment,
remove the impact of ESG‐related risks and capture the potential 17, 120–136. https://doi.org/10.1002/bse.506
profit opportunities, firms need to make more effort and be more Brockett, A., & Rezaee, Z. (2012). Corporate sustainability: Integrating perfor-
effective in execution. mance and reporting. New Jersey: John Wiley & Sons, Ltd.
Third, the success of industry leaders can accelerate the spread of Brower, J., & Rowe, K. (2017). Where the eyes go, the body follows?:
effective CSR strategies. Due to budget constraints, small and Understanding the impact of strategic orientation on corporate social
performance. Journal of Business Research, 79, 134–142. https://doi.
medium‐sized enterprises may spend less R&D expenditure on ESG org/10.1016/J.JBUSRES.2017.06.004
activities, especially on environmental issues, which need a large
Chandrashekar, A., Dougless, T., & Avery, G. C. (1999). The environment is
investment. From this aspect, industry leaders are in an advantageous free: The quality analogy. Journal of Quality Management, 4, 123–143.
position to try and determine effective CSR strategies, and small and https://doi.org/10.1016/S1084‐8568(99)80099‐6
medium‐sized enterprises could benefit from the spillover effect. Clarkson, P. M., Li, Y., Richardson, G. D., & Vasvari, F. P. (2011). Does it
really pay to be green? Determinants and consequences of proactive
Furthermore, government or nongovernment organizations can also
environmental strategies. Journal of Accounting and Public Policy, 30,
help to strengthen industrial associations by holding workshops or 122–144. https://doi.org/10.1016/j.jaccpubpol.2010.09.013
seminars to increase the collaboration between firms that face similar COSO‐WBCSD. 2018. Enterprise risk management. Applying enterprise
ESG issues. risk management to environmental, social and governance‐related
14 XIE ET AL.

risks, Retrieved from WBCSD website: http://docs.wbcsd.org/2018/ Business Strategy and the Environment, 27, 415–435. https://doi.org/
02/COSO_WBCSD_ESGERM.pdf, Jun 12, 2018 10.1002/bse.2007
De Grip, A., & Sauermann, J. (2012). The effects of training on own and co‐ Henderson, R. (2015). Making the business case for environmental sustainability.
worker productivity: Evidence from a field experiment. Economic Journal, ( pp. 22–47). An Organizational Perspective: Leading Sustainable Change.
122, 376–399. https://doi.org/10.1111/j.1468‐0297.2012.02500.x https://doi.org/10.1093/acprof:oso/9780198704072.003.0002
De Grip, A., & Sauermann, J. (2013). The effect of training on productivity: Hermalin, B. E., & Weisbach, M. S. (2003). Boards of directors as an endog-
The transfer of on‐the‐job training from the perspective of economics. enously determined institution: A survey of the economic literature.
Educational Research Review, 8, 28–36. https://doi.org/10.1016/j. Economic Policy Review, 9, 7–26. https://doi.org/10.2139/ssrn.233111
edurev.2012.05.005 Hermalin, B. E., & Weisbach, M. S. (2012). Information disclosure and cor-
Deng, X., Kang, J. K., & Low, B. S. (2013). Corporate social responsibility porate governance. Journal of Finance, 67, 195–234. https://doi.org/
and stakeholder value maximization: Evidence from mergers. Journal 10.1111/j.1540‐6261.2011.01710.x
of Financial Economics, 110, 87–109. https://doi.org/10.1016/j. Higginson, N., Simmons, C., & Warsame, H. (2006). Environmental disclo-
jfineco.2013.04.014 sure and legitimation in the annual report—Evidence from the joint
Dowell, G., Hart, S., & Yeung, B. (2000). Do corporate global environmental solutions project. Journal of Applied Accounting Research, 8, 3–23.
standards create or destroy market value? Management Science, 46, https://doi.org/10.1108/96754260680001048
1059–1074. https://doi.org/10.1287/mnsc.46.8.1059.12030 Hoobler, J. M., Masterson, C. R., Nkomo, S. M., & Michel, E. J. (2016). The
Eccles, R. G., Ioannou, I., & Serafeim, G. (2014). The impact of corpo- business case for women leaders. Journal of Management, 44, 2473,
rate sustainability on organizational processes and performance. 014920631662864–2499. https://doi.org/10.1177/0149206316628643
Management Science, 60, 2835–2857. https://doi.org/10.1287/ Jo, H., & Harjoto, M. A. (2011). Corporate governance and firm value: The
mnsc.2014.1984 impact of corporate social responsibility. Journal of Business Ethics, 103,
Eccles, R. G., Serafeim, G., & Krzus, M. P. (2011). Market interest in nonfi- 351–383. https://doi.org/10.1007/s10551‐011‐0869‐y
nancial information. Journal of Applied Corporate Finance, 23, 113–127. John, C. (1992). Misery's child. Social Semiotics, 2, 73–92. https://doi.org/
https://doi.org/10.1111/j.1745‐6622.2011.00357.x 10.1080/10350339209360349
Eichholtz, P., Kok, N., & Quigley, J. M. (2010). Doing well by doing good? Konings, J., & Vanormelingen, S. (2015). The impact of training on produc-
Green office buildings. American Economic Review, 100, 2492–2509. tivity and wages: Firm‐level evidence. The Review of Economics and
https://doi.org/10.1257/aer.100.5.2492 Statistics, 97, 485–497. https://doi.org/10.1162/REST_a_00460
Eichholtz, P., Kok, N., & Quigley, J. M. (2013). The economics of green Lee, K.‐H., Cin, B. C., & Lee, E. Y. (2016). Environmental responsibility and
building. The Review of Economics and Statistics, 95, 50–63. https:// firm performance: The application of an environmental, social and gov-
doi.org/10.1162/REST_a_00291 ernance model. Business Strategy and the Environment, 25, 40–53.
Erhardt, N. L., Werbel, J. D., & Shrader, C. B. (2003). Board of director https://doi.org/10.1002/bse.1855
diversity and firm financial performance. Corporate Governance, 11, Liang, H., & Renneboog, L. (2017). On the foundations of corporate social
102–111. https://doi.org/10.1111/1467‐8683.00011 responsibility. Journal of Finance, 72, 853–910. https://doi.org/
Esteban‐Sanchez, P., de la Cuesta‐Gonzalez, M., & Paredes‐Gazquez, J. D. 10.1111/jofi.12487
(2017). Corporate social performance and its relation with corporate Lokuwaduge, C. S. D. S., & Heenetigala, K. (2017). Integrating environmen-
financial performance: International evidence in the banking industry. tal, social and governance (ESG) disclosure for a sustainable
Journal of Cleaner Production, 162, 1102–1110. https://doi.org/ development: An Australian study. Business Strategy and the Environ-
10.1016/j.jclepro.2017.06.127 ment, 26, 438–450. https://doi.org/10.1002/bse.1927
Fatemi, A., Glaum, M., & Kaiser, S. (2016). ESG performance and firm value: Lorraine, N. H. J., Collison, D. J., & Power, D. M. (2004). An analysis of the
The moderating role of disclosure. Global Finance Journal. https://doi. stock market impact of environmental performance information. Account-
org/10.1016/j.gfj.2017.03.001 ing Forum, 28, 7–26. https://doi.org/10.1016/j.accfor.2004.04.002
Ferrell, A., Liang, H., & Renneboog, L. (2016). Socially responsible firms. Luan, C. J., & Tang, M. J. (2007). Where is independent director efficacy?
Journal of Financial Economics, 122, 585–606. https://doi.org/ Corporate Governance, 15, 636–643. https://doi.org/10.1111/j.1467‐
10.1016/j.jfineco.2015.12.003 8683.2007.00593.x
Firth, M., Wong, S. M. L., & Yang, Y. (2014). The double‐edged sword of Margolis, J. D., Elfenbein, H. A., & Walsh, J. P. (2009). Does it pay to be
CEO/chairperson duality in corporatized state‐owned firms: Evidence good … and does it matter? A meta‐analysis of the relationship
from top management turnover in China. Journal of Management and between corporate social and financial performance. SSRN Electronic
Governance, 18, 207–244. https://doi.org/10.1007/s10997‐012‐ Journal, 1001, 41234–48109. https://doi.org/10.2139/ssrn.1866371
9225‐6
Margolis, J. D., & Walsh, J. P. (2003). Misery loves companies: Rethinking
Freeman, E., & McVea, J. (2001). A stakeholder approach to strategic man- social initiatives by business. Administrative Science Quarterly, 48, 268.
agement. SSRN Electronic Journal, 1, 276. https://doi.org/10.2139/ https://doi.org/10.2307/3556659
ssrn.263511
McWilliams, A., & Siegel, D. (2001). Corporate social responsibility: A the-
Friedman, M. (1970). The social responsibility of business is to increase its ory of the firm perspective. Academy of Management Review, 26,
profits. Corporate Ethics and Corporate Governance, 6, 173–178. 117–127. https://doi.org/10.5465/AMR.2001.4011987
https://doi.org/10.1007/978‐3‐540‐70818‐6_14
Mervelskemper, L., & Streit, D. (2017). Enhancing market valuation of ESG
Fujii, H., Iwata, K., Kaneko, S., & Managi, S. (2013). Corporate environmen- performance: Is integrated reporting keeping its promise? Business
tal and economic performance of Japanese manufacturing firms: Strategy and the Environment, 26, 536–549. https://doi.org/10.1002/
Empirical study for sustainable development. Business Strategy and bse.1935
the Environment, 22, 187–201. https://doi.org/10.1002/bse.1747
Michael, V., & Paul, A. (1997). A resource‐based perspective on corporate
Garcia, A. S., Mendes‐Da‐Silva, W., & Orsato, R. (2017). Sensitive indus- environmental performance and profitability reproduced with permis-
tries produce better ESG performance: Evidence from emerging sion of the copyright owner. Further Reproduction Prohibited without
markets. Journal of Cleaner Production, 150, 135–147. https://doi.org/ Permission., 40, 534–559. https://doi.org/10.2307/257052
10.1016/j.jclepro.2017.02.180
Miles, M. P., & Russell, G. R. (1997). ISO 14000 total quality environ-
Haque, F., & Ntim, C. G. (2018). Environmental policy, sustainable develop- mental management: The integration of environmental marketing,
ment, governance mechanisms and environmental performance. total quality management, and corporate environmental policy.
XIE ET AL. 15

Journal of Quality Management, 2, 151–168. https://doi.org/ Terjesen, S., Couto, E. B., & Francisco, P. M. (2016). Does the presence of
10.1016/S1084‐8568(97)90026‐2 independent and female directors impact firm performance? A multi‐
Moore, G. (2001). Corporate social and financial performance: An investi- country study of board diversity. Journal of Management and Gover-
gation in the U.K. supermarket industry. Journal of Business Ethics, 34, nance, 20, 447–483. https://doi.org/10.1007/s10997‐014‐9307‐8
299–315. https://doi.org/10.1023/A:1012537016969 Trumpp, C., & Guenther, T. (2017). Too little or too much? Exploring U‐
Nollet, J., Filis, G., & Mitrokostas, E. (2016). Corporate social responsibil- shaped relationships between corporate environmental performance
ity and financial performance: A non‐linear and disaggregated and corporate financial performance. Business Strategy and the Environ-
approach. Economic Modelling, 52, 400–407. https://doi.org/ ment, 26, 49–68. https://doi.org/10.1002/bse.1900
10.1016/j.econmod.2015.09.019 Turban, D. B., & Greening, D. W. (1997). Corporate social performance and
Opler, T. C., & Titman, S. (2009). Financial distress and corporate perfor- organizational attractiveness to prospective employees. Academy of
mance. Journal of Finance, 49, 1015–1040. https://doi.org/10.2307/ Management Journal, 40, 658–672. https://doi.org/10.2307/257057
2329214 Verrecchia, R. E. (1990). Information quality and discretionary disclosure.
Orlitzky, M., Schmidt, F. L., & Rynes, S. L. (2003). Corporate social and Journal of Accounting and Economics, 12, 365–380. https://doi.org/
financial performance: A meta‐analysis. Organization Studies, 24, 10.1016/0165‐4101(90)90021‐U
403–441. https://doi.org/10.1177/0170840603024003910 Verrecchia, R. E. (2001). Essays on disclosure. Journal of Accounting and Eco-
Patten, D. M. (1992). Intra‐industry environmental disclosures in response nomics, 32, 97–180. https://doi.org/10.1016/S0165‐4101(01)00025‐8
to the Alaskan oil spill: A note on legitimacy theory. Accounting, Organi- Waddock, S. A., & Graves, S. B. (1997). The corporate social performance‐
zations and Society, 17, 471–475. https://doi.org/10.1016/0361‐ financial performance link. Strategic Management Journal, 18, 303–319.
3682(92)90042‐Q https://doi.org/10.1002/(SICI)1097‐0266(199704)18:4<303::AID‐
Peloza, J. (2009). The challenge of measuring financial impacts from invest- SMJ869>3.0.CO;2‐G
ments in corporate social performance. Journal of Management, 35, Wang, S.‐Y. (2012). Credit constraints, job mobility, and entrepreneurship:
1518–1541. https://doi.org/10.1177/0149206309335188 Evidence from a property reform in China. The Review of Economics and
Porter, M. E., & van der Linde, C. (1995). Toward a new conception of the Statistics, 94, 532–551. https://doi.org/10.1162/REST_a_00160
environment‐competitiveness relationship. Journal of Economic Per- Wang, Z., & Sarkis, J. (2017). Corporate social responsibility governance,
spectives, 9, 97–118. https://doi.org/10.1257/jep.9.4.97 outcomes, and financial performance. Journal of Cleaner Production,
Post, C., & Byron, K. (2015). Women on boards and firm financial perfor- 162, 1607–1616. https://doi.org/10.1016/j.jclepro.2017.06.142
mance: A meta‐analysis. Academy of Management Journal, 58, Weber, O. (2014). Environmental, social and governance reporting in
1546–1571. https://doi.org/10.5465/amj.2013.0319 China. Business Strategy and the Environment, 23, 303–317. https://
Qiu, Y., Shaukat, A., & Tharyan, R. (2016). Environmental and social disclo- doi.org/10.1002/bse.1785
sures: Link with corporate financial performance. The British Accounting Yu, E. P., Guo, C. Q., & van Luu, B. (2018). Environmental, social and gov-
Review, 48, 102–116. https://doi.org/10.1016/j.bar.2014.10.007 ernance transparency and firm value. Business Strategy and the
Rao, P., & Holt, D. (2005). Do green supply chains lead to competitiveness Environment. https://doi.org/10.1002/bse.2047
and economic performance? International Journal of Operations & Zhu, J., Ye, K., Tucker, J. W., & Chan, K. (. J.). (2016). Board hierarchy, inde-
Production Management, 25, 898–916. https://doi.org/10.1108/ pendent directors, and firm value: Evidence from China. Journal of
01443570510613956 Corporate Finance, 41, 262–279. https://doi.org/10.1016/j.
Riedl, A., & Smeets, P. (2017). Why do investors hold socially responsible jcorpfin.2016.09.009
mutual funds? Journal of Finance, 72, 2505–2550. https://doi.org/
10.1111/jofi.12547 SUPPORTI NG INFORMATION
Santos, L., & Escanciano, C. (2002). Benefits of the ISO 9000: 1994 system.
International Journal of Quality & Reliability Management, 19, 321–344.
Additional supporting information may be found online in the
https://doi.org/10.1108/02656710210415703 Supporting Information section at the end of the article.
Shane, P. B., Spicer, B. H., Shane, P. B., & Spicer, B. H. (1983). Market
response to environmental information produced outside the firm.
The Accounting Review, 58, 521–538. https://doi.org/10.2307/246532
Siew, R. Y. J. (2015). A review of corporate sustainability reporting tools How to cite this article: Xie J, Nozawa W, Yagi M, Fujii H,
(SRTs). Journal of Environmental Management, 164, 180–195. https:// Managi S. Do environmental, social, and governance activities
doi.org/10.1016/j.jenvman.2015.09.010 improve corporate financial performance?. Bus Strat Env.
Sternberg, E. (1997). The defects of stakeholder theory. Corporate 2018. https://doi.org/10.1002/bse.2224
Governance: An International Review, 5, 3–10. https://doi.org/
10.1111/1467‐8683.00034