Energy Policy 153 (2021) 112252

Contents lists available at ScienceDirect

Energy Policy
journal homepage: http://www.elsevier.com/locate/enpol

Does green investment improve energy firm performance?

Yufeng Chen *, Yanbai Ma
College of Business Administration, Capital University of Economics and Business, Beijing, 100070, China

A R T I C L E I N F O A B S T R A C T

Keywords: This paper aims to provide new evidence on the relationship between green investment and firm performance
Green investment through micro-level data. Data of energy listed firms in China from 2008 to 2017 are used here to explore this
Financial performance relationship. The research results show that green investment has a significant and positive correlation with
Environmental policy
financial performance, that is, increasing green investment helps improve financial performance. In the third
Energy conservation and emission reduction
Energy listed firms
year after investment in energy conservation and emission reduction, financial performance has been signifi­
cantly improved. Additionally, environmental tax, government subsidies, and technological innovation have
different positive moderating effects on green investment in promoting financial performance, and this result is
more obvious in the long-term performance. Lastly, this paper finds that green investment helps reduce envi­
ronmental violations and promote environmental performance, and environmental performance can strengthen
the impact of green investment in improving the long-term performance of firms. This conclusion implies that
firms should take environmental investment as its long-term strategy.

1. Introduction reduction (Wang et al., 2018). According to the data from ‘China
Eco-Environment Status Bulletin’, from 2017 to 2019, the average
Over the past 40 years of reform and opening-up policy, China’s number of EIA report projects approved by the Chinese government was
economy has made remarkable achievements (Cui and Huang, 2018). 208,000 per year, and the average total investment was 24.5 trillion
However, China’s economic development has led to serious environ­ (unit: RMB) per year. From 2008 to 2017, China’s total investment in
mental pollution due to low-cost manufacturing and less stringent environmental pollution control has nearly doubled, showing a contin­
environmental regulations (Duanmu et al., 2018). China’s air quality uous upward trend. Moreover, the proportion of this total investment in
ranks fourth from the bottom according to ‘the Report of Global Envi­ GDP has also been maintained at about 2% (see Fig. 1). Correspondingly,
ronmental Performance Index (2018)’ announced by Yale University. China has also made outstanding achievements in environmental
Besides, in the ‘China Eco-Environment Status Bulletin (2018)’ issued by pollution governance. In 2018, China’s carbon emissions per unit of GDP
the Ministry of Ecology and Environment of China, the proportion of air dropped by 45.8% compared to 2005, basically reversing the rapid
quality substandard in 338 cities at the prefecture-level and above in growth of greenhouse gas emissions. At the same time, China’s renew­
China will reach 64.2% in 2018. It can be seen from the above data that able energy investment ranks first in the world, and the cumulative
it is urgent to strengthen China’s environmental governance. Firms are reduction in carbon dioxide emissions also ranks first in the world.
the main body of the economy, the main users of energy, and the main However, China’s environmental problems are still not optimistic. It is
producers of environmental pollution (Huang and Lei, 2021). Improving estimated that China needs an additional 40.3–123.4 trillion (unit: RMB)
the pollution situation through green investment has gradually become to fund the transition to a green economy.
an important business decision of firms (Wang et al., 2018), and one Green investment is a complex management process and economic
kind of important manifestation of firm social responsibility (Wang behavior. It is not easy to achieve the “win-win” of its ecological goals
et al., 2015). Enlarging the scale of green investment by firms is an and economic benefits. The literature on the economic consequences of
important idea to realize green transformation of China’s economy. green investment finds that whether firms invest in green technology
Indeed, the Chinese government has paid great attention to envi­ depends on the size between the investment cost and the benefits of
ronmental pollution in recent years, such as actively encouraging and emission reduction (Yang et al., 2018), and firms only invest in green
supporting firms’ green practices in energy conservation and emission technology when they are profitable (Stucki, 2018). For firms, the more

* Corresponding author. No. 121 Zhangjia intersection, Huaxiang, Fengtai District, Beijing, 100070, China.
E-mail address: [email protected] (Y. Chen).

https://doi.org/10.1016/j.enpol.2021.112252
Received 23 September 2020; Received in revised form 8 March 2021; Accepted 13 March 2021
Available online 30 March 2021
0301-4215/© 2021 Elsevier Ltd. All rights reserved.
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

Unfortunately, the direct impact of green investment on firm perfor­

mance has been ignored, which is what this paper will discuss. Besides,
most studies focus on general firm investment, less on green investment
in energy industry (Yang et al., 2019; Zhou, 2019). Green investment has
become a key driver of the energy industry, and its rapid growth is
currently mainly driven by China (Eyraud et al., 2013). This paper can
fill this gap by providing evidence on green investment from energy
firms. In addition, limited by the differences in sample selection, the
method of analysis and empirical design (Zhang et al., 2019), the impact
of green investment on firm financial performance is uncertain, and
there is a lack of consensus on the relationship between them (Trumpp
and Guenther, 2017). In view of these problems, this paper aims to
provide new evidence on the relationship between green investment and
firm performance to open the black box of green investment.
Compared with previous literature, the possible contributions of this
Fig. 1. Environmental pollution control investment in China. study are as follows. First, by directly discussing the impact of green
investment on firm financial performance, it provides micro evidence of
concerned issue is not whether the organization is worth greening, but China’s energy industry in the dispute over whether green investment
when the environmental protection investment will be rewarded (King decisions and firm performance are coordinated or conflicted. This
and Lenox, 2001). China’s green development started late, and there is a paper further interprets the performance of green investment from the
certain gap between China and the international level in policy system, perspective of environmental regulation, government subsidies and
technical strength and practical application. Whether the contradiction technological innovation, which enriches the research scope of green
between economic growth and environmental protection can be coor­ development theory. It also provides a practical basis for encouraging
dinated through green investment is the key for China to promote the firms to actively carry out green investment activities. Second, this
construction of ecological civilization and development transformation. paper demonstrates the positive impact of green investment on envi­
Based on this, exploring the impact mechanism of green investment ronmental performance, and then analyzes the impact of environmental
behavior on economic performance has important theoretical value and performance on the relationship between green investment and finan­
practical significance for China’s environmental policies in the imple­ cial performance. This conclusion is helpful to overcome the limitation
mentation stage and for enhancing the green competitiveness of firms. of some literature on the correlation between environmental and
Previous literature on the impact of environmental governance on financial performance. Third, this conclusion has important policy im­
the choice of firm behavior, a series of studies around the “Porter Hy­ plications. The negative externalities of economic activities must
pothesis” is a main thread in this field. At the micro firm level, research establish an environmental governance pattern with the state-
mainly focuses on the impact of environmental regulation on firm per­ dominated and firms as the main body, and give play to the guiding
formance (e.g., firm cost and production efficiency) (Jaffe et al., 1995; function of the government in resource allocation. Firms should take the
Fleishman et al., 2009) and firm innovation (Ambec et al., 2013; Jiang environmental investment as its long-term strategy.
et al., 2018; You et al., 2019). Environmental decision of firms has a The rest of this paper proceeds as follows: Section 2 provides a
great impact on environment and firm value, and the most direct way to literature review, Section 3 displays hypothesis development, Section 4
understand the results of environmental behavior of firms is to test the explains the research design, Section 5 presents the empirical results and
relationship between environmental decision and firm performance conclusions, and policy implications are given in Section 6.
(Anderson-Weir, 2010). Firm financial performance will affect man­
agers’ investment decisions in the environment, which makes them have 2. Literature review
different investments in coping with social and consumer pressure
(Schaltenbrand et al., 2018). On one hand, environmental regulation Green investment refers to the investment necessary to reduce
drives firms to internalize environmental costs, which increases the cost greenhouse gas and air pollutant emissions, without significantly
burden of firms. On the other hand, “innovation compensation” effect reducing the production and consumption of non-energy goods (Eyraud
will attract firms to participate in green practice (Porter and et al., 2013). As a new way for firms to allocate resources, green in­
Van-der-Linde, 1995; Tang et al., 2018). Therefore, the green invest­ vestment puts limited resources into the development of green tech­
ment behavior of firms needs to take into account the “cost focus” path nology and renewable resources. By reducing energy consumption,
((Hassel et al., 2005); Popp et al., 2010) and the “value creation” path improving resource utilization efficiency and seeking renewable energy
(Saunila et al., 2018; Tang et al., 2018). According to the above analysis, as substitutes, it can reduce pollutant emissions and improve environ­
this paper puts forward the question: can green investment promote the mental quality. Therefore, under strict environmental regulations, the
growth of Chinese listed firms while achieving its environmental environmental practice of firms will significantly reduce the probability
benefits? of chemical oxygen demand (COD) emissions (Fan et al., 2019). This
The existing literature provides abundant evidence on the relation­ finding just confirms the results of Long et al. (2017). They found that
ship between green practice and firm performance. However, previous the positive effect of environmental innovation behavior on environ­
studies mainly focused on the impact of environmental regulation on mental performance is significantly better than economic performance,
firm performance, few studies focused on the direct relationship be­ which shows the advantages of environmental protection practice in
tween green investment and firm performance. The impact of environ­ reducing pollution. Luo et al. (2020) also shows that the general impact
mental regulation on firm performance is partly achieved through of FDI and foreign trade on carbon dioxide emissions is positive. How­
pollution charges and administrative punishment. These environmental ever, reducing pollution is the cost burden of firms, which is not
regulation tools will directly lead to the decline of financial perfor­ conducive to competitiveness; and reducing emissions can improve ef­
mance. Correspondingly, in order to reduce the regulatory cost, ficiency and save money, which makes firms have cost advantages (Hart
expanding the scale of green investment is a feasible path for firms. This and Ahuja, 1996). As noted above, how do green investment affect
is another key function of environmental regulation, that is, to realize firm’s production operations and economic consequences?
the internalization of the externality of environmental pollution. On one hand, green investment will occupy the resources used by
firms for normal production and operation, which will affect their

2
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

existing production and sales, and finally damage their financial per­ 3. Hypothesis development
formance. For example, the purchase of environmental protection
equipment and investment in R&D of green technology cost a large It can be seen from the above results that as an important form of
amount of money, but the income is uncertain, which leads firms to face firm investment, green investment will become a key measure for firms
greater investment risk (Hart and Ahuja, 1996). Moreover, green tech­ to gain competitive advantage. Although firms’ investment in environ­
nology investment usually creates benefits for others, but investors must mental protection is risky (Hart and Ahuja, 1996), proactive environ­
bear all costs, which causes the firm lack motivation to invest in new mental management strategies can improve firm performance through
technology to create public interests. Due to the public goods attribute of process innovation and product differentiation (Porter and Van der
ecological environment, the negative environmental externalities Linde, 1995; Hart, 1995; Oltra and Saint-Jean, 2009). He et al. (2019)
caused by economic activities can only be solved by environmental shows that increasing expenditure on environmental pollution control
policies adopted by the government (Popp et al., 2010). Therefore, green and adjusting industrial structure is conducive to improving green
investment is a passive behavior of firms catering to the government’s economic development indicators. Therefore, firms can obtain sustain­
environmental supervision (Lee, 2010; Berrone et al., 2013; Liao and able competitive advantages by reducing the adverse effects of business
Shi, 2018). activities on the natural environment (Clarkson et al., 2011). There are
On the other hand, green investment can release the signal that firms two possible reasons: first, environmental degradation will eventually
are keen on undertaking social responsibility, and bring good external damage sustained economic growth and social welfare brought by it,
reputation to firms, thus contribute to the improvement of economic and finally force firms to take an active part in environmental protection
performance (Hart, 1995; Tang et al., 2018). Firms will actively increase work. Thus, forming a new market competition pattern that automati­
the scale of green investment to reduce the cost of complying with cally restricts firms’ production behavior, gradually improving envi­
current environmental laws and regulations, so firms also show an active ronmental quality, and making green investment as indispensable part
behavior in green investment (Maxwell and Decker, 2006). This is also of a firm investment; second, firms that take the lead in carrying out
reflected in the fact that firms pay more attention to economic, insti­ green investment may have a greater market competitive advantage
tutional and social sustainability, and are more willing to invest in green than other firms. They will master more green investment technology
innovation (Saunila et al., 2018). Hence, although “pay to be green” is and related information, so that green investment becomes an important
related to improving firm performance, but there are considerable dif­ driving force for their sustainable development, and ultimately bring
ferences in green investment between different firms, such a strategy about the improvement of financial performance. To sum up, the first
cannot be easily imitated by all firms (Hart and Ahuja, 1996; Clarkson hypothesis is put forward:
et al., 2011).
Hypothesis 1. Green investment has a significant and positive corre­
In summary, “good behavior” of firms does not necessarily bring
lation with firm financial performance.
“good results”. There are two categories of literature about the impact of
firm’s “good behavior” on financial performance: the first type of liter­ The initial green investment of firms is mainly constrained by the
ature shows that green investment has a positive effect on financial government’s environmental policies. Therefore, the choice of envi­
performance. For example, Banerjee (2001) finds that green innovation ronmental policy tools will influence the effect of green investment on
strategy can improve the operation mode, reduce production costs, and firm financial performance. According to Porter Hypothesis, appropriate
improve corporate reputation, thus bringing significant improvement in environmental regulations can encourage innovation to compensate for
social and financial performance. Marchi (2010) also shows that environmental costs, to improve firm resource productivity (Porter and
through the implementation of green innovation, integrating environ­ Van-der-Linde, 1995). Thus, greater regulatory and norm pressures on
mental concerns into the strategic formulation of firms is conducive to environmental issues have a positive impact on the tendency of firms to
consolidating the competitive advantage. Guenster et al. (2011) find engage in environmental innovation (Berrone et al., 2013). Because the
that the environmental innovation practice of firms can achieve the dual looseness or strictness of environmental regulations and policies will
goals of environmental performance and economic performance. Chariri affect green investment decision-making of firms. The loose environ­
et al. (2018) shows that green investment significantly promotes firm mental regulation can’t provide opportunities for the current investment
growth of financial performance. Furthermore, firm characteristics such of firms in green, because it is unlikely that firms with more ability to use
as firm size, foreign ownership, industry profile, and frequency of audit fossil energy will enter the renewable energy market at present (Kim,
committee meeting will affect the effect of green investment. In the 2013). Popp’s (2006a) study also proves that environmental regulations
second kind of literature, green investment has a negative impact on promote the number of emission reduction patents of SO2 and NOx in­
financial performance. Most firms carry out green investment under the crease significantly, to reduce emissions of pollutants and improve
pressure of environmental regulation, which leads to the high private environmental performance. In addition, Lee (2010) finds that envi­
cost of firms, and the obvious lack of resource investment in profitable ronmental regulations lead to the emergence and diffusion of new and
commercial projects (Palmer et al., 1995; Ambec and Lanoie, 2008; more environmentally friendly technologies, enabling firms with
Weche, 2018), and a sharp decline in firm profits and capital (Fan et al., advanced environmental technologies to benefit from strict environ­
2019). mental policies and gain competitive advantages. Therefore, environ­
According to the above research, the impact mechanism of green mental regulation will promote firms to increase green investment,
investment on firm performance has dual character. Green governance develop energy conservation and environmental protection technology,
does not mean that firms give up pursuing economic benefits, but pro­ and ensure the dual benefits of economy and environment. The second
mote sustainable economic development by investing in green innova­ hypothesis is:
tion. With green growth gradually becoming the core of economic
Hypothesis 2. Environmental regulation positively moderates the
development, whether these results will show new changes remains to
relationship between green investment and financial performance.
be determined. Based on clarifying the motivation of green investment,
this part reviews the economic consequences of green investment on the In fact, except for the impact of command-oriented environmental
production and operation of firms under the influence of above factors. policy tools on firm performance, market-oriented environmental policy
Combined with the evolution and discussion of previous studies in the tools play a similar role (Carraro and Siniscalaco, 1994; Popp, 2006b;
introduction part, the topic of this study is: how does the green invest­ Acemoglu et al., 2016). Carraro and Siniscalaco’s (1994), who study on
ment of China’s energy listed firms affect their financial performance? environmental policy tools, found that innovation subsidies, such as
pollution taxes, have the effect of promoting emission reduction. Popp’s
(2006b) conclusion shows that the combination of reasonable carbon

3
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

tax and R&D subsidy has the best effect on the implementation of (2) The data of financial performance, environmental taxes, government
climate policy. This result is similar to the evidence provided by Ace­ subsidies, technological innovation and other research variables are all
moglu et al. (2016). Their research shows that carbon taxes and R&D from the China Stock Market and Accounting Research Database
subsidies may encourage clean technology production and innovation. (CSMAR).
In addition, government subsidies are also the main force supporting the
development of small and medium-sized renewable energy firms (Yang 4.2. Variables
et al., 2019). Meanwhile, government subsidies are an effective means to
alleviate financing constraints and encourage firm innovation (Howell, (1) Green investment. According to definition of green investment
2017; Yang et al., 2019). (Eyraud et al., 2013), all kinds of capital expenditure related to
In the open economic circumstance, firms should abandon the pro­ environmental protection should be included in firm green in­
duction strategy of high pollution and high energy consumption when vestment, for example, financial investment in renewable tech­
the market economy is moving towards green and sustainable devel­ nology, selected energy-saving technology and R&D in green
opment as a whole. Firms should take the initiative to carry out green technology. In order to control the impact of firm size differences,
investment and provide innovative environmental protection products firms’ total assets at the end of the year are used to offset green
for the market. They also need to attract new customers through tech­ investment. According to the contents of firm social responsibility
nological innovation, improve market adaptability and enhance sus­ report and other documents, the following expenditures are
tainable competitive advantage. For example, Zhang et al. (2019) proves defined as a firm’s green investment: expenditures and expenses
that there is a positive correlation between green innovation and firm related to technological transformation and R&D, industrial
performance. By and large, government subsidies and technological waste treatment and other pollutants treatment, purchase and
innovation can help improve the relationship between green investment construction of desulfurization and denitrification equipment
and firm financial performance. To sum up, Hypothesis 3-4 is proposed: and boiler reconstruction, clean and renewable project con­
struction, mine ecological environment restoration management
Hypothesis 3. Government subsidy will positively moderate the effect
and greening.
of green investment in promoting financial performance.
(2) Financial performance. The measurement of firm performance
Hypothesis 4. Technological innovation will also moderate the effect should reflect its profitability and growth, and net profits can
of green investment in promoting financial performance. effectively reflect the profitability of firms (Zhang et al., 2019).
Return on equity (ROE) can not only reflect the operating effi­
4. Research design ciency of the firm, but also its capital structure, which is condu­
cive to measuring financial performance (Hart and Ahuja, 1996).
4.1. Sample and data Choi et al. (2010) also choices return on equity (ROE) and
Tobin’Q to measure firm financial performance. Surroca et al.
China’s Stock Exchange has issued ‘Guide on Social Responsibility’ (2010) shows that the reason why Tobin’Q is used to measure
since 2006, requiring listed firms to disclose environmental information, financial performance is that it can capture the value of long-term
and gradually expanding firm types that are enforced to disclose social investment (such as intangible assets). On the whole, this paper
responsibility information. It can be found that environmental protec­ mainly uses Net profit, ROE and Tobin’Q to measure firm finan­
tion investment and other environmental governance information is its cial performance.
important contents from the social responsibility report. With the Chi­ (3) Environmental policy. According to the research results of the
nese government paying more attention to the problem of ecological existing literature, environmental policies mainly include the
environment, the environmental information disclosure of energy firms government’s restriction, supervision and incentive schemes for
and heavy polluting firms has become a focus of stakeholders. The firm pollutant emissions through environmental regulations,
development of China’s economy mainly depends on energy consump­ emission standards and permits, environmental R&D subsidies,
tion dominated by fossil fuel energy (Ji and Zhang, 2019). Moreover, etc., so as to achieve environmental protection purpose of energy
green investment has become a key driver of the energy industry, and its conservation and emission reduction (e.g., Carraro and Sinis­
rapid growth is currently mainly driven by China (Eyraud et al., 2013). caico, 1994; Jaffe et al., 1995; Leiter et al., 2011; Berrone et al.,
Therefore, this paper selects energy listed firms in China from 2008 to 2013). Davidovic et al. (2020) says that environmental taxes are
20171 as the research sample. According to the “Industry Classification often regarded as the key to more effective environmental pro­
Guidelines” issued by the China Securities Regulatory Commission in tection. The Chinese government has started to implement the
2012, and He et al. (2019), the definition of listed energy firm is as environmental protection tax law nationwide since the beginning
follows: industries of oil, natural gas and coal mining (code B06 and of 2018. The pollution charge policy has been changed from
B07), industries of oil coking (code C25), industries of electricity, heat, pollution charge to environmental tax to cope with the increas­
gas, and water production and supply (codes D44, D45, and D46). This ingly serious climate change and air pollution. Therefore, quasi
paper excluded ST firms, firms with incomplete data and missing values, environmental tax is used to replace environmental tax in this
and obtained 378 observations of 90 energy listed firms. This paper paper, mainly including resource tax, urban construction tax,
deals with winsorize of all continuous variables at the level of 1% to water conservancy construction fund, land use tax, mineral
avoid the influence of outliers. resource tax, consumption tax, vehicle and ship use tax, river
Data sources are as follows: (1) green investment. Using the firm management fee, etc. Referring to Howell (2017), subsidies
social responsibility report, environmental report and sustainable received by firms from the government are used to measure
development report disclosed by Cninfo2 to manually collect and sort. government subsidies.
(4) Technological innovation. According to Zhang et al. (2019), the
number of patent applications of firms is used as the proxy var­
1 iable of technological innovation.
Since 2006 and 2007 are the experimental years for public disclosure of
environmental information, only a small number of listed firms disclose social (5) Control variables. According to the idea of the literature quoted
responsibility reports, so it is difficult to obtain environmental protection in­ in this paper, control variables are mainly divided into two cat­
vestment data. Hence, the sample period began in 2008.2017 is the most recent egories. First, control some variables about the characteristics of
year to obtain data. the firms, including firm size (Firmsize, measured as the loga­
2
Website: http://www.cninfo.com.cn/new/index. rithm of total assets), leverage (Lev, measured as liabilities to

4
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

total assets), and asset structure (Assetstru, measured as fixed summary statistics of financial performance indicators, the mean value
assets to total assets). Second, control the variables of firm of Tobin’Q is 0.963. According to the theory of Tobin’s Q, firms may not
governance, including the proportion of independent directors buy new investment products, and investment expenditure may
(Pid, measured as the number of independent directors to the decrease when Q < 1. The mean value (0.065) of ROE is less than the
number of board members), equity concentration (H5, measured median (0.076), indicating that more than half of the listed energy firms’
as the sum of the square of top five major shareholders’ share­ return on equity is higher than the industry average level. Similarly,
holding ratio), and the shareholding ratio of the largest share­ descriptive statistics of Netprofit also show that more than half of the
holder (Top1, measured as the number of shares of the first listed energy firms have a higher net profit than the industry average
largest shareholder divided by the total number of shares). In level. The mean value of Greeninvest is 0.018, which is greater than the
addition, this paper controls the year and industry fixed effect. median (0.006), and the maximum value is 0.17, which shows that most
of China’s energy listed firms only invest a small part of their total assets
in green practice. This result is consistent with Wang et al. (2018)’s
4.3. Models
summary statistics on green investment of Chinese manufacturing listed
firms. From the mean, median and maximum value of environmental
This paper mainly studies the impact of green investment on firm
tax, government subsidies and patent application numbers, the sample
financial performance. In order to test Hypothesis 1, the following model
firms show great differences in environmental policies and technological
is established:
innovation. Among the control variables, the descriptive statistical re­

∑ ∑ ∑
CFPi,t+3 = β0 + β1 Greeninvesti,t + γ ControlVari,t + Industry + Year + ε (1)

where the dependent variable CFPi,t+3 represents the financial perfor­ sults of firm characteristics and firm governance conform to the char­
mance of firm i in year t+3. Independent variable Greeninvesti,t repre­ acteristics of Chinese listed firms.
∑
sents green investment of firm i in year t. ControlVari,t represents
∑ ∑
control variable. Industry and Year represents a vector of industry 5.2. The impact of green investment on financial performance
and year dummies. ε is the error term.
In order to discuss the influence of moderating variables on the Table 2 is the regression result of Hypothesis 1, which is used to test
relationship between green investment and financial performance, this the Model (1). In columns (1)–(3), the dependent variables Tobin’Q,
paper introduces three variables, environmental tax, government sub­ ROE and Netprofit all adopt the data in year t+3. The reason is that there
sidies and technological innovation, and their interaction term with is a time lag between the start of emission reduction and the realization
green investment. Models (2)–(4) are used to test Hypothesis 2-4, of a guaranteed return due to the sluggish characteristics of investment,
respectively. so the impact of emission reduction costs on firm performance is time
lagging (Hart and Ahuja, 1996). Zhang et al. (2018) also adopts a similar

CFPi,t+3 = β0 + β1 Greeninvesti,t + β2 environtaxi,t + β3 Greeninvesti,t × environtaxi,t +

∑ ∑ ∑ (2)
γ ControlVari,t + Industry + Year + ε

idea, taking firm performance in year t+3 as the dependent variables.

From the regression results in Table 2, it can be found that the co­
CFPi,t+3 = β0 + β1 Greeninvesti,t + β2 subsidyi,t + β3 Greeninvesti,t × subsidyi,t +
∑ ∑ ∑ efficients of Greeninvest (β1 = 1.7435, 0.7477 and 7.2431, respectively)
γ ControlVari,t + Industry + Year + ε are significant and positive at the 10% level, indicating that green in­
(3) vestment has a significant and positive correlation with Tobin’Q, ROE
and Netprofit. It means that increasing green investment can improve
CFPi,t+3 = β0 + β1 Greeninvesti,t + β2 patenti,t + β3 Greeninvesti,t × patenti,t +
∑ ∑ ∑
γ ControlVari,t + Industry + Year + ε
(4) Table 1
Descriptive statistics.
where the environtaxi,t, subsidyi,t, and patenti,t represents the environ­ Variables N mean sd min p50 max
mental tax, government subsidy, and innovation patents of firm i in year Tobin’Q 378 0.963 0.836 0.149 0.672 4.355
t, respectively. The Greeninvesti,t × environtaxi,t, Greeninvesti,t × sub­ ROE 378 0.065 0.104 − 0.418 0.076 0.285
sidyi,t, Greeninvesti,t × patenti,t is the interaction term of green invest­ Netprofit 335 20.20 1.954 15.64 20.29 25.50
ment and environmental tax, government subsidy, and innovation Greeninvest 378 0.018 0.031 0 0.006 0.170
environtax 365 15.12 3.091 0 15.41 21.38
patents, respectively.
subsidy 332 17.27 1.784 13.34 17.17 22.23
patent 378 2.629 1.678 0 2.565 8.103
5. Empirical results Firmsize 378 23.77 1.463 20.41 23.74 28.28
Lev 378 0.554 0.173 0.077 0.564 0.901
5.1. Descriptive statistics Assetstru 378 0.464 0.168 0.127 0.460 0.846
Pid 378 0.357 0.043 0.250 0.333 0.556
H5 378 0.268 0.156 0 0.259 0.759
Table 1 is descriptive statistics of the main variables. From the Top1 378 0.466 0.175 0.005 0.494 0.863

5
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

Table 2 government subsidies and technological innovation, respectively. In

The impact of green investment on financial performance (benchmark columns (1)–(3), the coefficients of Greeninvest × environtax (β3 =
regression). 0.3008 and 0.0362) is significantly positive when the dependent vari­
Dependent (1) (2) ROE (t+3) (3) ables are Tobin’Q and ROE, but the interaction had no significant effect
variable Tobin’Q (t+3) Netprofit (t+3) on promoting Netprofit; environtax has a significant and negative
Greeninvest 1.7435* (1.95) 0.7477* (1.96) 7.2431* (1.77) correlated with Tobin’Q, while the correlation with ROE and Netprofit is
Firmsize − 0.2845*** 0.0242** (2.33) 1.1475*** (10.44) not significant. In columns (4)–(6), the coefficient of Greeninvest ×
(− 6.45) subsidy (β3 = 0.1739 and 0.0386) is also significantly positive when the
Lev − 1.5859*** − 0.3225*** − 4.0347***
dependent variables are Tobin’Q and ROE, but the negative moderating
(− 4.35) (− 3.27) (− 4.55)
Assetstru − 0.4641* (− 1.79) 0.0541 (1.05) 0.2986 (0.37) effect of government subsidies on green investment to improve Netprofit
Pid − 0.4626 (− 0.59) − 0.1286 (− 0.92) − 1.2780 (− 0.71) is not significant; subsidy has a significant and negative correlated with
H5 3.2836*** (3.18) 0.4092* (1.73) 5.9504** (2.13) Tobin’Q, ROE and Netprofit. In columns (7)–(9), the coefficient (β3 =
Top1 − 2.9374*** − 0.3941** − 4.6495* (− 1.98) 1.2908) of Greeninvest × patent is significantly positive at the 5% level,
(− 3.31) (− 2.07)
_cons 9.4112*** (7.75) − 0.2949 (− 1.13) − 4.5271 (− 1.55)
patent is only significant and positive correlated with Tobin’Q, but the
Year/Industry yes yes yes interaction has no significant effect on ROE and Netprofit.
Obs 118 118 102 Compared with the coefficients of the variables in Table 4, envi­
Adj_R2 0.6914 0.4017 0.7637 ronmental tax, government subsidies and technological innovation have
Note: t is the value in parentheses, *** means 1% level is significant, ** means a positive moderating effect on green investment in promoting financial
5% level is significant, * means 10% level is significant, and the same below. performance. However, there are significant differences in the effects of
the three moderating variables on this relationship. First, as a whole,
firm financial performance. Therefore, Hypothesis 1 is proved. Hart and technological innovation has the greatest contribution to the promotion
Ahuja (1996) point out that emission reduction leads to higher returns of Tobin’Q, and it will also strengthen the positive role of green in­
on equity two or three years later. This result shows that green invest­ vestment in Tobin’Q. Tobin’Q can well reflect the value of long-term
ment has a long-time lag in promoting firm financial performance. It can investment of firms (Surroca et al., 2010), the result shows that green
be seen that the conclusion is also strong evidence for the results of Hart investment has brought “innovation compensation” effect for firms in
and Ahuja (1996). the long term, so it has a significant effect in improving Tobin’Q. The
In the benchmark regression of Table 2, considering the investment second is the moderating effect of environtax, and the last is subsidy.
lagging, financial performance indicators in year t+3 is taken as the Second, although the coefficients of greeninvest × environtax and
dependent variable. However, once the investment occurs, it is bound to greeninvest × subsidy are significantly positive, environtax and subsidy
have an impact on firm financial performance in the current and future. have a negative effect on firm performance. On one hand, under the
The results have shown that green investment will improve financial influence of market forces, the primary task of firms that choose
performance after two years. However, how will these results change in short-sighted behavior is to survive. As a market-oriented policy tool,
the current period and next two years after investment? In China, the environmental tax may not be conducive to the realization of long-term
length of the lag period may be different (Zhang et al., 2019). Therefore, value in the short term. On the other hand, government subsidies “crowd
the data of Tobin’Q, ROE and Netprofit in year t, t+1 and t+2 are used as out” the resources and motivation of firms to engage in green innovation
dependent variables to further explore the dynamic impact of green (Shleifer and Vishny, 1994). After receiving government subsidies, it is
investment on firm financial performance in Table 3. In column (2)–(3), possible to reduce green investment in its resources, which makes the
(5)–(6), (8)–(9), there is no significant and positive relationship between government subsidies and green investment appear a “substitution”
green investment and financial performance, while Greeninvest with phenomenon. Third, from the above results, environmental taxes, gov­
Netprofit and ROE show the negative correlation in year t and t+2 ernment subsidies, and technological innovations have no significant
respectively. The reason may lie in that the motive for maximizing effect on moderating green investment to boost net profits, but have a
shareholders’ benefits may induce firms to pay more attention to significant positive effect on Tobin’Q and ROE (especially Tobin’Q),
changes in short-term performance such as the growth of net profit, indicating that environmental taxes and government subsidies have a
resulting in green investment’s effect in improving financial perfor­ significant positive moderating effect on green investments in promot­
mance is not significant, and even negatively affect financial perfor­ ing Tobin’Q and ROE. Technological innovation only has a significant
mance. In columns (1), (4), and (7), the coefficients (β1 = 1.601, 2.3143 positive moderating effect on green investment in promoting Tobin’Q.
and 2.6439, respectively) of Greeninvest are positive and significant, On the whole, environmental taxes, government subsidies and techno­
indicating that Greeninvest has a significantly positive correlation with logical innovation have different positive moderating effect on green
Tobin’Q. When Tobin’Q is used to measure firm financial performance, investment in promoting financial performance, and this result is more
the indicator can effectively describe the value of long-term investments obvious in the long-term performance, but not for the short-term plan
(Surroca et al., 2010). Therefore, according to Tables 2 and 3, it can be that pays more attention to net profit. The Hypothesis 2-4 is supported.
found that when firms use green investment as a long-term development
strategy, the impact of short-term profit and loss on firms will be 5.3. Robustness test
weakened, and green governance will help increase the long-term value
of firms. Therefore, green investment and Tobin’Q were significantly In order to verify the reliability of the empirical results, the following
positive in year t = 0–3, while green investment showed a significant robustness tests are carried out. First, this paper uses the natural loga­
and positive correlation with ROE and Netprofit in year t > 2. rithm of green investment (Ln (greeninvest)) instead of the variable
Table 4 is the regression result of Hypothesis (2)–(4), which is used to Greeninvest to regress the Model (1), and the corresponding results are
test the Model (2)–(4). Columns (1)–(3), (4)–(6), (7)–(9) show the shown in columns (1)–(3) in Table 5. As shown in the table, green in­
empirical results of moderating effects3 of environmental taxes, vestment has a significant and positive effect on Tobin’Q and Netprofit,
but not on ROE. Since each firm does not disclose green investment in
every year, the distribution of green investment in each firm in the 10-
3
According to Cohen et al. (2003) and the methods of most existing studies, year research period is not uniform. Excluding the samples with green
This paper conducted a centralized treatment of both the moderating variables investment less than 3 years and repeat the regression of Model (1). The
and the independent variables in order to make the coefficients of regression test results are shown in columns (4)–(6) of Table 5. It can be seen from
equations more explanatory. the coefficient of green investment that there is a significant and positive

6
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

Table 3
The impact of green investment on financial performance (dynamic analysis).
Dependent (1) (2) ROE (t) (3) (4) (5) ROE (t+1) (6) (7) (8) ROE (t+2) (9)
variable Tobin’Q (t) Netprofit (t) Tobin’Q (t+1) Netprofit Tobin’Q (t+2) Netprofit
(t+1) (t+2)

Greeninvest 1.6010* 0.0721 (0.54) − 1.1960 2.3143*** 0.1477 (0.82) 1.5376 (0.66) 2.6439*** − 0.0323 4.3636 (1.46)
(1.89) (− 0.65) (2.32) (3.65) (− 0.10)
Firmsize − 0.3013*** 0.0082 (1.59) 1.0571*** − 0.3121*** 0.0116* 1.0994*** − 0.3039*** 0.0150* 1.1350***
(− 7.92) (17.66) (− 8.99) (1.73) (14.73) (− 7.52) (1.90) (12.06)
Lev − 2.1242*** − 0.2279*** − 3.7730*** − 2.0990*** − 0.2781*** − 3.9928*** − 1.7557*** − 0.3164*** − 4.1815***
(− 8.61) (− 5.23) (− 8.31) (− 8.27) (− 4.46) (− 6.50) (− 6.21) (− 4.25) (− 6.51)
Assetstru − 0.3405* 0.0669** 0.5543 (1.29) − 0.2360 0.0791* 0.9131 (1.62) − 0.4080* 0.0458 (1.00) 0.7272 (1.08)
(− 1.90) (2.05) (− 1.18) (1.90) (− 1.77)
Pid 0.2924 (0.43) − 0.0758 − 2.6132* − 0.1793 − 0.1646 − 2.3507 − 0.3353 − 0.2000 − 1.6782
(− 0.77) (− 1.94) (− 0.25) (− 1.52) (− 1.53) (− 0.51) (− 1.59) (− 1.03)
H5 2.9620*** 0.1733 (1.43) 2.8362* 4.3025*** 0.2218 (1.39) 4.2709** 4.0992*** 0.4361** 4.3138**
(4.31) (1.87) (5.00) (2.61) (4.11) (2.35) (2.00)
Top1 − 2.3876*** − 0.1337 − 1.7824 − 3.7844*** − 0.1457 − 3.1112** − 3.6031*** − 0.3542** − 3.3455*
(− 4.26) (− 1.34) (− 1.32) (− 5.22) (− 1.07) (− 2.23) (− 4.26) (− 2.35) (− 1.86)
_cons 9.4213*** 0.0585 (0.43) − 1.5602 10.6339*** − 0.0168 − 2.6847 10.279*** − 0.0268 − 4.1129*
(10.72) (− 1.06) (11.64) (− 0.10) (− 1.45) (9.30) (− 0.13) (− 1.83)
Year/ yes yes yes yes yes yes yes yes yes
Industry
Obs 378 378 335 246 246 219 172 172 151
Adj_R2 0.6902 0.2979 0.7488 0.7412 0.2995 0.7596 0.7169 0.3658 0.7581

Table 4
The results of moderating effects.
Dependent (1) (2) ROE (t+3) (3) (4) (5) ROE (t+3) (6) (7) (8) ROE (t+3) (9)
variable Tobin’Q (t+3) Netprofit Tobin’Q (t+3) Netprofit Tobin’Q (t+3) Netprofit
(t+3) (t+3) (t+3)

Greeninvest − 0.1369 0.5469 (1.36) 7.0578 (1.55) − 0.1828 0.5481 (1.34) 4.2022 (1.27) − 0.0621 0.5762 (1.44) 7.8851*
(− 0.23) (− 0.16) (− 0.05) (1.67)
Greeninvest × 0.3008** 0.0362* 0.0218 (0.08)
environtax (2.33) (1.77)
Greeninvest × 0.1739** 0.0386* − 0.0546
subsidy (2.00) (1.86) (− 0.29)
Greeninvest × 1.2908** 0.1493 (1.35) − 0.5382
patent (2.46) (− 0.39)
environtax − 0.0481*** 0.0014 (0.25) − 0.0066
(− 3.85) (− 0.12)
subsidy − 0.1403*** − 0.0180*** − 0.3426***
(− 4.34) (− 2.96) (− 3.17)
patent 0.0789** − 0.0167 0.0431 (0.47)
(2.52) (− 1.53)
Firmsize − 0.1950*** 0.0245* 1.1592*** − 0.1557*** 0.0526*** 1.5696*** − 0.2949*** 0.0339** 1.1148***
(− 3.96) (1.76) (7.99) (− 2.83) (4.15) (9.46) (− 6.48) (2.49) (7.93)
Lev − 1.6225*** − 0.3381*** − 4.0368*** − 1.4993*** − 0.3701*** − 4.7181*** − 1.4388*** − 0.3649*** − 3.8671***
(− 4.85) (− 3.30) (− 4.28) (− 3.91) (− 3.74) (− 4.33) (− 4.50) (− 3.14) (− 3.94)
Assetstru − 0.3376 0.0647 (1.22) 0.3109 (0.37) − 0.3499 0.0895* 0.0985 (0.09) − 0.2637 0.0508 (0.96) 0.2963 (0.36)
(− 1.41) (− 1.52) (1.76) (− 1.06)
Pid − 0.5769 − 0.1195 − 1.2992 − 0.8269 − 0.1056 − 2.7830 − 0.8374 − 0.0680 − 1.4334
(− 0.78) (− 0.85) (− 0.72) (− 1.04) (− 0.64) (− 1.32) (− 1.21) (− 0.44) (− 0.79)
H5 2.7434*** 0.3903 (1.57) 5.9065** 4.0974*** 0.2261 (0.80) 5.9406** 2.4515** 0.4728* 5.8214*
(2.90) (2.10) (3.85) (2.23) (2.53) (1.96) (1.98)
Top1 − 2.5313*** − 0.3734* − 4.6208* − 4.2109*** − 0.2200 − 5.5290** − 2.1857*** − 0.4424** − 4.5416*
(− 3.18) (− 1.89) (− 1.96) (− 4.67) (− 0.87) (− 2.09) (− 2.82) (− 2.19) (− 1.84)
_cons 7.9887*** − 0.3375 − 4.7024 9.0592 (7.87) − 0.7309** − 7.4313** 9.2780 (7.53) − 0.4791 − 3.8822
(6.46) (− 1.11) (− 1.40) (− 2.34) (− 2.02) (− 1.56) (− 1.14)
Year/Industry yes yes yes yes yes yes yes yes yes
Obs 118 118 102 94 94 79 118 118 102
Adj_R2 0.7292 0.4129 0.7638 0.7817 0.5279 0.8236 0.7339 0.4237 0.7643

correlation between green investment and the three indicators of firm 5.4. Endogenous test
performance. Compared with the coefficients of green investment in
columns (1)–(3), the coefficients in columns (4)–(6) are significantly The results show that green investment can promote firm financial
higher. The reason is that the measurement methods of independent performance. On the other hand, the better the financial performance,
variables in columns (1)–(3) and (4)–(6) are different, and the number of the more green investment possible. There are endogenous problems
regression samples is also different. However, the test results are not caused by the two-way causal relationship between them. In addition,
substantially different from the above conclusions. Hence, the research because this paper only selects 378 listed energy firms as the research
conclusion of this article is more reliable. In other words, green invest­ samples, there is a large sample selection problem, which is easy to
ment helps improve firm financial performance. produce sample selection bias. This paper uses instrumental variable
method and Heckman two-stage regression to overcome the above

7
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

Table 5
The result of robustness test.
Dependent variable Change measurement method Exclude Greeninvest less than 3 years

(1) (2) ROE (t+3) (3) (4) (5) ROE (t+3) (6)
Tobin’Q (t+3) Netprofit (t+3) Tobin’Q (t+3) Netprofit (t+3)

Ln (greeninvest) 0.0503** (2.32) 0.0012 (0.32) 0.0721* (1.68)

Greeninvest 1.7435* (1.95) 0.7477* (1.96) 7.2431* (1.77)
Firmsize − 0.3811*** (− 4.63) 0.0136 (1.33) 0.9746*** (9.04) − 0.2845*** (− 6.45) 0.0242** (2.33) 1.1475*** (10.44)
Lev − 1.3349*** (− 3.15) − 0.2414*** (− 3.08) − 3.6499*** (− 4.94) − 1.5859*** (− 4.35) − 0.3225*** (− 3.27) − 4.0347*** (− 4.55)
Assetstru − 0.5131** (− 2.00) 0.0517 (1.10) 0.8329 (1.13) − 0.4641* (− 1.79) 0.0541 (1.05) 0.2986 (0.37)
Pid 1.1154 (0.94) − 0.1112 (− 0.80) − 0.3495 (− 0.21) − 0.4626 (− 0.59) − 0.1286 (− 0.92) − 1.2780 (− 0.71)
H5 4.3032 (3.55) 0.2981 (1.30) 5.8931** (2.34) 3.2836*** (3.18) 0.4092* (1.73) 5.9504** (2.13)
Top1 − 3.4830*** (− 4.05) − 0.3104* (− 1.68) − 4.8473** (− 2.25) − 2.9374*** (− 3.31) − 0.3941** (− 2.07) − 4.6495* (− 1.98)
_cons 10.5386*** (6.99) − 0.0542 (− 0.21) − 1.1633 (− 0.43) 9.4112*** (7.75) − 0.2949 (− 1.13) − 4.5271 (− 1.55)
Year/Industry yes yes yes yes yes yes
Obs 150 150 130 118 118 102
Adj_R2 0.6477 0.2891 0.7887 0.6914 0.4017 0.7637

Table 6
The result of endogenous test - IV-GMM estimation.
Dependent variable (1) (2) (3)

Greeninvest Tobin’Q (t+3) Greeninvest ROE (t+3) Greeninvest (t+3) Netprofit (t+3)

Greeninvest 9.4185* (1.57) 0.4329* (1.98) 2.9381* (1.35)

GreenIV 0.0000** (2.42) 0.0000** (2.42) 0.0000** (2.08)
Firmsize − 0.0057* (− 1.78) − 0.2640*** (− 6.68) − 0.0057* (− 1.78) 0.0246*** (2.72) − 0.0051 (− 1.50) 1.1579*** (13.63)
Lev 0.0156 (1.12) − 1.6397*** (− 4.59) 0.0156 (1.12) − 0.3146*** (− 3.56) 0.0194 (1.24) − 4.0241*** (− 4.96)
Assetstru − 0.0071 (− 0.52) − 0.4012 (− 1.58) − 0.0071 (− 0.52) 0.0466 (1.02) − 0.0190 (− 1.31) 0.0516 (0.08)
Pid − 0.0450 (− 1.12) 0.0809 (0.09) − 0.0450 (− 1.12) − 0.1525 (− 1.20) − 0.0484 (− 1.19) − 1.5725 (− 0.95)
H5 − 0.0563 (− 1.01) 3.2851*** (3.62) − 0.0563 (− 1.01) 0.3764* (1.79) − 0.0471 (− 0.80) 5.7052** (2.30)
Top1 0.0822 (1.65) − 3.0727*** (− 4.09) 0.0822 (1.65) − 0.3610** (− 2.18) 0.0696 (1.31) − 4.4148** (− 2.16)
_cons 0.1200* (1.85) 8.4796*** (8.62) 0.1200* (1.85) − 0.3078 (− 1.42) 0.1136 (1.65) − 4.6653** (− 2.17)
Year/Industry yes yes yes yes yes yes
Obs 118 118 118 118 102 102
Adj_R2 0.2920 0.6061 0.2920 0.3946 0.4320 0.7601

endogenous problems. ∑ ∑ ∑
Instrumental variable requires that they must be independent of GIi,t = β0 + β1 SOEsi,t + γ ControlVari,t + Industry + Year + ε
error term and related to independent variables, so it is difficult to find a (5)
suitable instrumental variable. Based on the consideration of operability
The results of the first stage show that SOEs, Firmsize and Pid are
and applicability, this paper takes the average value of green investment
positively correlated with GI, while Lev is negatively correlated with GI.
in the industry and city where the firm is located as the instrumental
The inverse mills ratio (IMR) calculated by the first stage model was
variable of green investment (GreenIV), and adopts the two-stage IV-
incorporated into the Model (1), and the regression of the model was
GMM estimation method to test. Kaustia and Rantala (2015) point out
repeated. After controlling the sample selection bias, the regression re­
that the investment decisions and other economic activities of firms are
sults in Table 7 show that the coefficient of green investment is signif­
easily affected by the same activities of other firms in the same industry
icantly positive, which is consistent with the results in this paper. By and
and region. Therefore, this instrumental variable generally meets the
large, the conclusions of this study are reliable.
requirements of correlation and exogenous. The estimated results are
shown in Table 6. From the results of the first stage, the coefficients of
GreenIV are significantly positive at the 5% level. In the second stage of 5.5. The effect of environmental performance
regression results, the coefficient of green investment is significantly
positive. This result is consistent with the conclusion of this paper. The results show that green investment can significantly improve
The advantage of Heckman two-stage model is to correct the financial performance in the long run. This also proves that firms only
endogenous problems caused by self-selection bias and sample selection invest in green technology when they are profitable (Stucki, 2018). In
bias. The model is generally divided into two stages: the first stage is fact, the impact of green behavior on environmental performance is
probit model, which is used to estimate the probability of the occurrence significantly greater than that on economic performance (Long et al.,
of selective bias variables and obtain the inverse Mills ratio. The second 2017), and it is easier for firms to improve environmental performance
stage is to regress the inverse Mills ratio together with other variables to in the short term. Neely et al. (1995) points out that environmental
correct the endogenous problem caused by selection bias. Therefore, it is performance refers to the efficiency and effect of firms’ compliance with
reasonable to adopt this model here. In this paper, the dependent vari­ a series of environmental regulations. Firms can obtain financial per­
able in the first stage is the dummy variable (GI) of the level of green formance and sustainable competitive advantage by reducing the
investment. When green investment exceeds the average value of Year- adverse impact of business activities on the natural environment
Industry, the value is 1; otherwise, the value is 0. The control variables (Clarkson et al., 2011; Wang et al., 2015). Therefore, the primary
remain unchanged. In addition, the dummy variable of property right function of green investment is to reduce pollutant emissions, improve
(SOEs) is added. When SOEs are state-owned firms, the value is 1; environmental quality and improve environmental performance. By
otherwise, the value is 0. The Probit model of the first stage is as follows: establishing a green image to attract green investors and green con­
sumers, increase market share, and reduce costs through resource uti­
lization efficiency, reduce pollution avoidance costs, and bring

8
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

Table 7 result. In the first two years after green investment, environmental
The result of endogenous test - Heckman two stage model. performance has been significantly improved. The reason may be that
Dependent (1) Probit (2) (3) (4) there is a certain time lag in green investment. Based on this, discussing
variable model the role of environmental performance between green investment and
Greeninvest ROE (t+3) Netprofit Tobin’Q financial performance. Columns (5)–(7) introduce the environmental
(t+3) (t+3) performance and its interaction item with green investment. It can be
Greeninvest 1.9787* 0.1004* 8.6100*
seen from the coefficient of Greeninvest that the result of green invest­
(1.87) (1.15) (1.12) ment promoting financial performance has not changed. When the
Firmsize 0.1234* 0.0812 (1.00) 0.0206 (1.14) 1.2301*** dependent variables are ROE and Netprofit, the coefficient of EP is
(1.73) (4.81) significantly positive, while the coefficient of Greeninvest × EP is
Lev − 1.9275*** − 4.5128*** − 0.4026*** − 5.5365***
significantly negative. This result is opposite when the dependent vari­
(− 3.47) (− 4.07) (− 3.33) (− 3.15)
Assetstru 0.6310 (1.22) 1.3005*** 0.0951 (1.13) 1.6211 (1.24) able is Tobin’Q, but it is not statistically significant. It shows that green
(3.39) investment is conducive to the improvement of environmental perfor­
Pid 3.5472* 5.0731*** − 0.0545 2.4085 (0.54) mance, and environmental performance can strengthen the positive
(1.80) (2.78) (− 0.21) impact of green investment on financial performance. Hence, environ­
H5 − 1.6200 − 0.7826 0.3144 (0.87) 3.6081 (0.92)
(− 0.85) (− 0.66)
mental performance positively moderates the relationship between
Top1 0.7224 (0.45) − 0.0680 − 0.3316 − 3.1592 green investment and financial performance.
(− 0.10) (− 1.16) (− 0.99) It can be found that green investment is conducive to the improve­
SOEs 0.5190** ment of environmental performance in the short term, but conducive to
(2.09)
the improvement of financial performance in the long term. Facing the
IMR 3.9390*** 0.0333 (0.17) 1.8474 (0.63)
(2.97) pressure of environmental regulation, green investment behavior helps
_cons − 3.0594* − 2.3543 − 0.1798 − 8.2541 evade government regulation, so the environmental performance is
(− 1.75) (− 0.93) (− 0.31) (− 0.98) improved in the short term. Meanwhile, the environmental governance
Year/ yes yes yes yes behaviors of firms are easy to win the government and social support,
Industry
Obs 372 92 92 77
which is conducive to the production and operation of firms. Hence,
Pseudo R2/ 0.0510 0.7697 0.5031 0.7856 after the improvement of environmental quality, financial performance
Adj_R2 has been improved accordingly. This result supports Porter Hypothesis,
that is, the cost of environmental governance compresses the profit
space of firms, which can’t effectively stimulate firms to carry out
performance improvements (Klassen and McLaughlin, 1996). For
technological innovation in the short term; but in the long run, the
instance, the prosocial attitude of firms can promote green innovation
positive effect of green investment will gradually manifest, realizing the
and improve market competitiveness (Aghion et al., 2020). Thus, envi­
innovation compensation effect (see the result in column (7) of Table 4).
ronmental performance will affect the effect of green investment on
It is worth noting that China’s environmental technology and technol­
financial performance. This paper further discusses the role of envi­
ogy gap rate (TGR) has great heterogeneity in region and space (Long
ronmental performance in this relationship. According to Klassen and
et al., 2018a, 2018b). This fact is not conducive to the improvement of
McLaughlin (1996), and Campos et al. (2015), environmental violation
environmental performance and economic performance of Chinese firms
records of firms are used as proxy variables of environmental perfor­
at the general level, and is a key factor for China’s green development
mance (EP).4 This paper insists that environmental performance is a
lagging. Therefore, firms should take green investment as a long-term
centralized reflection of firm environmental governance, and firms with
development strategy, in line with the concept of green growth and
effective environmental governance should have fewer environmental
sustainable development.
violations than those with polluting firms. However, environmental vi­
olations and performance are two opposite indicators. The better the
6. Conclusions and policy implications
environmental performance is, the less the probability of environmental
violations and the less the behavior of environmental violations will be.
Green investment has become an important task for economic soci­
As discussed, green investment in the current period and lag 1–3
ety to achieve sustainable development in contemporary era. Because of
periods are used as independent variables, respectively. Environmental
the positive externalities of green investment and the public goods at­
performance is used as the dependent variable. Testing the impact of
tributes of environmental governance, the impact of green investment
green investment on environmental performance, and the results are
on environmental and financial performance is uncertain. In order to
shown in columns (1)–(4) of Table 8. The results show that the coeffi­
open the black box of green investment behavior of firms, data of energy
cient of Greeninvest is significantly negative in columns (2) and (3),
listed firms in China from 2008 to 2017 are used here to explore the
indicating that the more green investment, the less probability of envi­
impact of green investment on firm performance. The results show that
ronmental violations, that is, green investment helps promote environ­
green investment is significant and positively correlated with Tobin’Q,
mental performance. However, there are obvious time limits to this
ROE and Netprofit, that is, increasing green investment can improve
firm financial performance. The dynamic analysis shows that the effect
of green investment in improving financial performance becomes more
4
The data of firm environmental violations are from the website of the significant with the extension of the investment period. Financial per­
Institute of Public and Environmental Affairs (IPE, http://www.ipe.org.cn). formance is significantly improved in the third year after green invest­
Klassen and McLaughlin (1996) find that rewards or punishments obtained by ment. It means that when green investment is regarded as a long-term
firms in environmental aspects could better reflect environmental performance. strategy, the improvement of financial performance is significantly
Campos et al. (2015) research exemplifies this view well. They characterize
better than the investment behavior motivated by short-term profit.
environmental performance by the number of firms that do not meet national or
Additionally, environmental tax, government subsidies, and techno­
international standards. Therefore, it is reasonable to represent environmental
performance use environmental violation records. If the firm has no environ­ logical innovation have different positive moderating effects on green
mental violations in a certain year, the environmental performance (EP) value investment in promoting financial performance, especially the long-
is 0, otherwise, it is 1. Because the index is a binary choice variable, the Probit term performance. Lastly, the result shows that green investment can
model is used to test the impact of green investment on environmental per­ promote environmental performance, and environmental performance
formance. The results correspond to columns (1)–(4) of Table 8. can strengthen the impact of green investment in improving its long-

9
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

Table 8
The effect of environmental performance.
Dependent (1) EP (t) (2) EP (t+1) (3) EP (t+2) (4) EP (t+3) (5) (6) ROE (t+3) (7)
variable Tobin’Q (t+3) Netprofit (t+3)

Greeninvest 2.6997 (0.84) − 12.1850*** − 9.4572* − 19.7374 4.0562* (1.87) 0.7526** (2.14) 7.9131** (2.17)
(− 2.74) (− 1.70) (− 1.51)
EP − 0.0084 (− 0.97) 0.0037* (1.94) 0.0480*** (2.66)
Greeninvest × EP 0.7894 (1.59) − 0.2184** (− 2.22) − 3.4030***
(− 3.51)
Firmsize 0.2911*** (2.76) 0.2439 (1.58) 0.1805 (1.17) 0.1998 (1.20) − 0.3203*** 0.0165 (1.65) 1.0608*** (10.44)
(− 5.08)
Lev 0.0049 (0.01) − 0.6273 (− 0.58) 0.6323 (0.47) − 0.1858 (− 0.15) − 1.3589*** − 0.2726*** − 4.0695***
(− 3.35) (− 3.33) (− 5.63)
Assetstru − 0.6432 (− 0.80) − 0.5989 (− 0.63) 0.4335 (0.39) − 0.3738 (− 0.26) − 0.4785* (− 1.87) 0.0532 (1.09) 0.8275 (1.15)
Pid − 1.2065 (− 0.53) − 9.8691** (− 2.34) − 1.5696 (− 0.94) − 5.0359 (− 1.00) 1.1753 (0.95) − 0.0778 (− 0.54) − 0.4997 (− 0.29)
H5 − 1.7475 (− 0.66) − 3.2243 (− 1.06) − 0.3007 (− 0.09) − 5.5868 (− 1.39) 3.7875*** (3.54) 0.3096 (1.31) 5.3557** (2.25)
Top1 − 1.7496 (− 0.88) − 0.8736 (− 0.38) − 2.7863 (− 1.12) 2.0137 (0.70) − 3.0617*** − 0.3467* (− 1.81) − 4.7576** (− 2.33)
(− 3.94)
_cons − 9.6755*** − 1.2481 (− 0.34) − 3.7995 (− 1.06) − 3.2436 (− 0.83) 9.3596*** (7.47) − 0.1179 (− 0.45) − 2.3974 (− 0.91)
(− 3.75)
Year/Industry yes yes yes yes yes yes yes
Obs 378 179 172 115 150 150 130
Pseudo R2/Adj_R2 0.1633 0.2242 0.1759 0.1585 0.6506 0.3219 0.7973

term performance. Declaration of competing interest

The empirical results have the following policy implications: First,
the government should strengthen the enforcement of environmental The authors declare that they have no known competing financial
taxes, and appropriately increase the environmental taxes and fees on interests or personal relationships that could have appeared to influence
firm pollutant emissions. Exploring a reasonable basis for environmental the work reported in this paper.
charges in the implementation of environmental policies, and give play
to the positive role of environmental taxes on green investment in Acknowledgements
improving firm performance. Second, due to the inertia of organization
and the malpractice of some firms obtaining subsidy resources through This study was financially supported by the National Natural Science
rent-seeking activities, the government should establish a scientific Foundation of China (No. 71673250); Zhejiang Provincial Natural Sci­
evaluation mechanism of subsidy, strengthen the supervision and ence Foundation of China Outstanding Youth Foundation (No.
assessment of supporting firms, and improve the incentive effect of LR18G030001); MOE Major Project of Key Research Institute of Hu­
subsidies. Third, the government should vigorously develop green credit manities and Social Sciences in Universities, China (No. 14JJD770019);
and green direct financing, and strengthen financial support for R&D Zhejiang Provincial Philosophy and Social Science Foundation, China
and application of green technology. The results show that green in­ (No. 18NDJC184YB). We would be grateful for any helpful comments
vestment brings innovation compensation effect. Therefore, it is neces­ and any remaining errors are the joint responsibility of the authors.
sary to use green finance to improve the independent innovation ability
and knowledge spillover rate of China’s local firms (Long et al., 2018a, References
2018b, ), and give priority to supporting firm financing that is conducive
to green transformation. In addition, firms should actively participate in Acemoglu, D., Akcigit, U., Hanley, D., Kerr, W., 2016. Transition to clean technology.
J. Polit. Econ. 124 (1), 52–104.
green practice, incorporate environmental issues into the production Aghion, P., Bénabou, R., Martin, R., Roulet, A., 2020. Environmental preferences and
function and strategy formulation. Through improving environmental technological choices: is market competition clean or dirty?. In: NBER Working
performance to establish the green image and reputation of firms, to Paper, p. 26921.
Ambec, S., Cohen, M.A., Elgie, S., Lanoie, P., 2013. The Porter Hypothesis at 20: can
achieve the promotion of financial performance and competitive environmental regulation enhance innovation and competitiveness? Rev. Environ.
advantage. Econ. Pol. 7 (1), 2–22.
There are some limitations in this study. First, this paper only dis­ Anderson-Weir, C.H., 2010. How does the Stock market react to corporate environmental
news? Und. Econ. Rev. 6 (1), 1–29.
cusses the performance of green investment with listed energy firms as Banerjee, S.B., 2001. Managerial perceptions of corporate environmentalism:
research samples, and does not extend this issue to a more general interpretations from industry and strategic implications for organizations. J. Manag.
micro-level. Second, this paper explores the impact of green investment Stud. 38 (4), 489–513.
Berrone, P., Fosfuri, A., Gelabert, L., et al., 2013. Necessity as the mother of ‘green’
on firm performance from external factors such as environmental
inventions: institutional pressures and environmental innovations. Strat. Manag. J.
regulation, without considering the impact of internal factors such as 34 (8), 891–909.
firm characteristics and firm governance. Hence, it will be a beneficial Campos, L.M.S., de Melo Heizen, D.A., Verdinelli, M.A., Cauchick Miguel, P.A., 2015.
direction to further compare the differences of the above research in Environmental performance indicators: a study on ISO 14001 certified companies.
J. Clean. Prod. 99, 286–296.
firm size and industry attributes, and to incorporate firm governance Carraro, C., Siniscaico, D., 1994. Environmental policy reconsidered: the role of
into the research framework of this paper, such as ownership structure technological innovation. Eur. Econ. Rev. 38 (3–4), 545–554.
and board governance mechanism. Chariri, A., Bukit, G., Eklesia, O., Christi, B., Tarigan, D., 2018. Does green investment
increase financial performance? Empirical evidence from Indonesian companies. E3S
Web of Conferences 31 1–7.
CRediT authorship contribution statement Choi, J.S., Kwak, Y.M., Choe, C., 2010. Corporate social responsibility and corporate
financial performance: evidence from Korea. Aust. J. Manag. 35, 291–311.
Clarkson, P.M., Li, Y., Richardson, G.D., Vasvari, F.P., 2011. Does it really pay to be
Yufeng Chen: Conceptualization, Supervision, Writing – review & green? Determinants and consequences of proactive environmental strategies.
editing, Funding acquisition. Yanbai Ma: Data curation, Software, J. Account. Publ. Pol. 30 (2), 122–144.
Methodology, Writing – original draft. Cui, L., Huang, Y., 2018. Exploring the schemes for green climate fund financing:
international lessons. World Dev. 101, 173–187.

10
Y. Chen and Y. Ma Energy Policy 153 (2021) 112252

Davidovic, D., Harring, N., Jagers, S.C., 2020. The contingent effects of environmental metafrontier directional slacks-based measure approach. Renew. Sustain. Energy
concern and ideology: institutional context and people’s willingness to pay Rev. 82, 3962–3971.
environmental taxes. Environ. Polit. 29 (4), 674–696. Luo, Y.S., Lu, Z.N., Long, X.L., 2020. Heterogeneous effects of endogenous and foreign
Duanmu, J.L., Bu, M.L., Pittman, R., 2018. Does market competition dampen innovation on CO2 emissions stochastic convergence across China. Energy Econ. 91,
environmental performance: evidence from China. Strat. Manag. J. 39 (11), 104893.
3006–3030. Marchi, V.D., 2010. Cooperation toward green innovation: an empirical investigation.
Eyraud, L., Clements, B., Wane, A., 2013. Green investment: trends and determinants. Marco Fanno Working Paper, #119.
Energy Pol. 60, 852–865. Maxwell, J.W., Decker, C.S., 2006. Voluntary environmental investment and responsive
Fan, H., Zivin, J., Kou, Z., Liu, X., Wang, H., 2019. Going green in China: firms’ responses regulation. Environ. Resour. Econ. 33, 425–439.
to stricter environmental regulations. In: NBER Working Paper, p. 26540. Neely, A., Gregory, M., Plattsk, K., 1995. Performance measurement system design: a
Fleishman, R., Alexander, R., Bretschneider, S., Popp, D., 2009. Does regulation stimulate literature review and research agenda. Int. J. Oper. Prod. Manag. 15 (4), 80–116.
productivity? The effect of air quality policies on the efficiency of US power plants. Oltra, V., Saint Jean, M., 2009. Sectoral systems of environmental innovation: an
Energy Pol. 37 (11), 4574–4582. application to the French automotive industry. Technol. Forecast. Soc. Change 76
Guenster, N., Derwall, J., Bauer, R., Koedijk, C., 2011. The economic value of corporate (4), 567–583.
eco-efficiency. Eur. Financ. Manag. 17 (4), 679–704. Popp, D., 2006a. International innovation and diffusion of air pollution control
Hart, S.L., 1995. A natural-resource-based view of the firm. Acad. Manag. Rev. 20 (4), technologies: the effects of NOx and SO2 regulation in the US, Japan, and Germany.
996–1014. J. Environ. Econ. Manag. 51 (1), 46–71.
Hart, S.L., Ahuja, G., 1996. Does it pay to be green? An empirical examination of the Popp, D., 2006b. R&D subsidies and climate policy: is there a “free lunch”? Climatic
relationship between emission reduction and firm performance. Bus. Strat. Environ. Change 77 (3–4), 311–341.
5 (1), 30–37. Popp, D., Newell, R.G., Jaffe, A.B., 2010. Energy, the environment and technological
Hassel, L.G., Nilsson, H., Nyquist, S., et al., 2005. The value relevance of environmental change. In: Hall, B., Rosenberg, N. (Eds.), Handbook of the Economics of Innovation.
performance. Eur. Account. Rev. 14 (1), 41–61. Elsevier Press., pp. 873–937.
He, L.Y., Zhang, L.H., Zhong, Z.Q., Wang, D.Q., Wang, F., 2019. Green credit, renewable Porter, M.E., Van-der-Linde, C., 1995. Toward a new conception of the environment-
energy investment and green economy development: empirical analysis based on competitiveness relationship. J. Econ. Perspect. 9 (4), 97–118.
150 listed companies of China. J. Clean. Prod. 208, 363–372. Saunila, M., Ukko, J., Rantala, T., 2018. Sustainability as a driver of green innovation
Howell, S.T., 2017. Financing innovation: evidence from R&D grants. Am. Econ. Rev. investment and exploitation. J. Clean. Prod. 179, 631–641.
107 (4), 1136–1164. Schaltenbrand, B., Foerstl, K., Azadegan, A., Lindeman, K., 2018. See what we want to
Huang, L., Lei, Z., 2021. How environmental regulation affect corporate green see? The effects of managerial experience on corporate green investments. J. Bus.
investment: evidence from China. J. Clean. Prod. 279, 123560. Ethics 150, 1129–1150.
Jaffe, A.B., Peterson, S.R., Portney, P.R., Stavins, R.N., 1995. Environmental regulation Shleifer, A., Vishny, R., 1994. Politicians and firms. Q. J. Econ. 109 (4), 995–1025.
and the competitiveness of U.S. manufacturing: what does the evidence tell us? Stucki, T., 2018. Which firms benefit from investments in green energy technologies?
J. Econ. Lit. 33 (1), 132–163. —the effect of energy costs. Res. Pol. 48 (3), 546–555.
Ji, Q., Zhang, D., 2019. How much does financial development contribute to renewable Surroca, J., Tribó, J.A., Waddock, S., 2010. Corporate responsibility and financial
energy growth and upgrading of energy structure in China? Energy Pol. 128, performance: the role of intangible resources. Strat. Manag. J. 31 (5), 463–490.
114–124. Tang, M.F., Walsh, G., Lerner, D., Fitza, M.A., Li, Q.H., 2018. Green innovation,
Jiang, Z., Wang, Z., Li, Z., 2018. The effect of mandatory environmental regulation on managerial concern and firm performance: an empirical study. Bus. Strat. Environ.
innovation performance: evidence from China. J. Clean. Prod. 203, 482–491. 27 (1), 39–51.
Kaustia, M., Rantala, V., 2015. Social learning and corporate peer effects. J. Financ. Trumpp, G., Guenther, T., 2017. Too little or too much? Exploring U-shaped relationship
Econ. 117 (3), 653–669. between corporate environmental performance and corporate financial
Kim, E.H., 2013. Deregulation and differentiation: incumbent investment in green performance. Bus. Strat. Environ. 26, 49–68.
technologies. Strat. Manag. J. 34 (10), 1162–1185. Wang, Q., Dou, J.S., Jia, S.H., 2015. A meta-analytic review of corporate social
King, A.A., Lenox, M.J., 2001. Does it really pay to be green? An empirical study of firm responsibility and corporate financial performance: the moderating effect of
environmental and financial performance. J. Ind. Ecol. 5 (1), 105–116. contextual factors. Bus. Soc. 4, 1–39.
Klassen, R.D., McLaughlin, C.R., 1996. The impact of environmental management on Wang, K., Zhang, H.M., Tsai, S.B., Wu, L.D., Xue, K.K., Fan, H.J., Zhou, J., Chen, Q.,
firm performance. Manag. Sci. 42 (8), 199–1214. 2018. Does a board chairman’s political connection affect green investment? — from
Lee, C.W., 2010. The effect of environmental regulation on green technology innovation a sustainable perspective. Sustainability 10 (3), 1–14.
through supply chain integration. Int. J. Sustain. Econ. 2, 92–112. Weche, J.P., 2018. Does green corporate investment really crowd out other business
Leiter, A.M., Parolini, A., Winner, H., 2011. Environmental regulation and investment: investment? Ind. Corp. Change 28 (5), 1279–1295.
evidence from European industry data. Ecol. Econ. 70 (4), 759–770. Yang, X.L., He, L.Y., Xia, Y.F., Chen, Y.F., 2019. Effect of government subsidies on
Liao, X.C., Shi, X.P., 2018. Public appeal, environmental regulation and green renewable energy investments: the threshold effect. Energy Pol. 132, 156–166.
investment: evidence from China. Energy Pol. 119, 554–562. You, D., Zhang, Y., Yuan, B., 2019. Environmental regulation and firm eco-innovation:
Long, X., Chen, Y., Du, J., et al., 2017. The effect of environmental innovation behavior evidence of moderating effects of fiscal decentralization and political competition
on economic and environmental performance of 182 Chinese firms. J. Clean. Prod. from listed Chinese industrial companies. J. Clean. Prod. 207, 1072–1083.
166, 1274–1282. Zhang, D.Y., Rong, Z., Ji, Q., 2019. Green innovation and firm performance: evidence
Long, X.L., Chen, B., Park, B., 2018a. Effect of 2008’s Beijing Olympic Games on from listed companies in China. Resour. Conserv. Recycl. 144, 48–55.
environmental efficiency of 268 China’s cities. J. Clean. Prod. 172, 1423–1432. Zhou, H.N., 2019. Internal control, environmental policy and corporate environmental
Long, X.L., Wu, C., Zhang, J.J., Zhang, J., 2018b. Environmental efficiency for 192 investment—empirical evidence from listed companies in heavy polluting industries.
thermal power plants in the Yangtze River Delta considering heterogeneity: a Comm. Finan. Acc. (6), 88–91.

11