Carbon Management

ISSN: 1758-3004 (Print) 1758-3012 (Online) Journal homepage: www.tandfonline.com/journals/tcmt20

A long-term and heterogeneous study on the

impact of carbon emission trading policy on
financial performance

Ran Wang, Hanwen Zhang, Yin Feng, Jing Wang & Tianyue Yang

To cite this article: Ran Wang, Hanwen Zhang, Yin Feng, Jing Wang & Tianyue Yang
(2025) A long-term and heterogeneous study on the impact of carbon emission
trading policy on financial performance, Carbon Management, 16:1, 2486627, DOI:
10.1080/17583004.2025.2486627

To link to this article: https://doi.org/10.1080/17583004.2025.2486627

© 2025 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group

Published online: 10 Apr 2025.

Submit your article to this journal

Article views: 680

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tcmt20
CARBON MANAGEMENT
2025, VOL. 16, NO. 1, 2486627
https://doi.org/10.1080/17583004.2025.2486627

A long-term and heterogeneous study on the impact of carbon emission

trading policy on financial performance
Ran Wanga,b, Hanwen Zhanga, Yin Fengc, Jing Wanga and Tianyue Yanga
a
School of Economic & Management, China University of Geosciences, Wuhan, China; bResearch Center of Resource and
Environmental Economics, China University of Geosciences, Wuhan, China; cSchool of Economics and Trade, Hubei University of
Economics, Wuhan, China

ABSTRACT ARTICLE HISTORY

China’s “14th Five-Year Plan“ sets ambitious goals to peak carbon emissions by 2030 and Received 20 November 2024
achieve carbon neutrality by 2060. Since 2013, China has been piloting carbon emission Accepted 23 March 2025
trading policies to internalize environmental externalities through market mechanisms. This
KEYWORDS
paper employs a quasi-natural experiment, using data from Chinese companies spanning
Carbon emission trading
2010 to 2021, to analyze the impact of the pilot carbon emission trading policy on green policy; green technology
technology innovation and financial performance, utilizing a difference-in-differences innovation; financial
approach. The findings indicate that: (1) the carbon emission trading policy positively affects performance; long-term and
firms’ financial performance, with long-term benefits; (2) while the policy initially suppresses heterogeneous; difference-
financial performance by driving green technology innovation, this negative effect dimin­ in-differences model
ishes over time; (3) the long-term effects are particularly significant, with heterogeneity ana­
lysis revealing more pronounced impacts on state-owned enterprises and firms located in
high-carbon price regions. These insights are crucial for understanding the policy’s role in
promoting sustainable corporate development and guiding the ongoing refinement of
China’s national carbon emissions trading market.

HIGHLIGHTS

� The carbon emission trading policy has a more pronounced positive impact on firms’
financial performance over the long term.
� Heterogeneity analysis reveals that the policy’s impact is more significant for state-
owned enterprises and firms located in high-carbon price regions.
� Green technology innovation serves as a catalyst for long-term financial gains by initially
requiring investments that may temporarily suppress financial performance, but ultim­
ately leading to enhanced efficiency and sustainability that boost firms’ financial out­
comes over time.

Introduction civilization. The CETP employs market mechanisms

to promote carbon emission reduction, facilitating
With global warming becoming a pressing issue,
the transformation and upgrading of the economic
countries worldwide are actively exploring green
structure. It is one of the essential measures to
development models to achieve sustainable
achieve the goals of carbon peak and carbon neu­
growth. According to the World Bank (2023), China
trality. The evolution and development of the
has been the world’s largest CO2 emitter since
CETP can be broadly categorized into the follow­
2006. In 2022, China emitted 11.4 billion tons of ing four stages (Figure 1).
CO2, accounting for approximately 30.68% of glo­ The report of the 20th National Congress of the
bal emissions. To achieve its "dual carbon" goals, Communist Party of China explicitly proposed to
China has implemented numerous policies and "accelerate the establishment of a new develop­
guidelines, with the establishment of a carbon ment pattern and promote high-quality develop­
emissions trading policy (CETP) being a significant ment." High-quality economic development entails
step in fulfilling international emission reduction not only ensuring economic growth but also
commitments and enhancing ecological enhancing both its quality and efficiency. As key

CONTACT yin feng [email protected] School of Economic & Management, Hubei University of Economics, Wuhan, China
This article has been corrected with minor changes. These changes do not impact the academic content of the article.
� 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow
the posting of the Accepted Manuscript in a repository by the author(s) or with their consent.
2 R. WANG ET AL.

Figure 1. Stages of evolutionary development of CETP.

drivers of national economic development, techno­ These research findings can aid the government in
logical innovation, and talent attraction, enter­ formulating relevant policy guidelines and provide
prises play a crucial role in advancing high-quality firms with a reference for management decisions.
economic development. Financial performance The main contributions of this paper are as follows:
(FP), as an essential indicator of high-quality cor­ First, this study adopts a dynamic research perspec­
porate economic development [1,2], not only tive and conducts an in-depth exploration of multidi­
reflects a company’s short-term profitability but mensional relationships. While most existing studies
also serves as a critical measure of its long-term examine the impact of CETP from a static perspective,
growth potential and market competitiveness. By this paper takes a dynamic approach to analyze the
optimizing FP, enterprises can better promote green long-term effects of the relationship among CETP,
innovation and industrial upgrading, thereby contri­ GTI, and FP. Specifically, we not only focus on the
buting to high-quality economic development. immediate effects of policy implementation but also
How to realize high-quality economic develop­ delve into how such policies incentivize corporate GTI
ment by implementing the CETP is a key issue that and their long-term impact mechanisms on FP.
needs to be addressed. Taking a long-term per­ Secondly, whereas previous studies have primarily
spective into account, green technological innov­ explored the heterogeneous effects of CETP on pilot
ation (GTI) is a fundamental way to ensure enterprises from the perspectives of ownership struc­
ture and industry differences, this paper classifies pilot
economic benefits while combating pollution.
regions into high-carbon-price and low-carbon-price
However, GTI is characterized by high investment,
areas based on the carbon trading prices in each
high risk, and long cycles, which often discourage
province and city. This classification further verifies
some firms [3]. Most studies on the CETP have
the differential impact of CETP on the FP of enter­
focused on the macro level, exploring its impact
prises in high- and low-carbon-price areas. Carbon
on regional economies [4]. Some studies have
pricing is a current research hotspot, as it effectively
examined the policy’s effect on investment effi­
reflects carbon market supply and demand conditions
ciency, energy utilization, and competitiveness
as well as policy implementation. It plays a crucial
[5–7], or investigated the respective impacts of the
role in corporate cost control and competitive
CETP on GTI and FP, but have failed to analyze its
dynamics, while also serving as an important refer­
impact mechanism in depth [8]. ence for achieving carbon reduction targets.
This study constructs a quasi-natural experiment The rest of the paper is structured as follows.
with the carbon emissions trading pilot policy and Section 2 provides a review of the relevant litera­
employs a difference-in-differences (DID) model for ture. Section 3 describes the research method­
empirical research. Focusing on the micro level, it ology, including the empirical analysis used to
aims to explore the actual impact of the CETP on construct the research model, as well as data sour­
firm FP driven by GTI and analyzes the impact of ces and variable definitions. Sections 4 and 5 dis­
the CETP on multi-period FP from a dynamic per­ cuss the empirical results of the paper in detail.
spective to examine the long-term nature of the The final section summarizes the paper and pro­
policy’s effects. Additionally, heterogeneity tests vides policy recommendations.
are conducted across three aspects—property
rights, carbon price differences, and regional differ­
ences—to thoroughly investigate the relationship
between environmental regulation, GTI and FP.
 CARBON MANAGEMENT 3

Literature review an indispensable goal for their sustainable opera­

tions [21], and actively guide the flow of resources
Balancing the relationship between economic
from high-pollution, high-energy-consumption sec­
development and environmental pollution is a crit­
tors to low-energy-consumption, environmentally
ical issue, and in this dilemma, the concept of
friendly sectors [22]. Regarding the relationship
emissions trading emerges as a significant innov­
between CETP and GTI, existing research has
ation. Emissions trading was first proposed by
proved that the establishment of China’s carbon
American economist Dales [9]. Later, Coase [10]
emission trading market can effectively promote
suggested using market trading and clear property
the development of low-carbon technology and
rights to internalize negative externalities, adopt­
encourage enterprises to carry out green innov­
ing market-oriented environmental regulation poli­
ation activities [23,24]. The carbon trading system
cies to coordinate economic development and
uses price leverage to encourage enterprises to
environmental protection. This approach has
become an important policy tool for governments reduce emissions from the source and save capital
to promote carbon reduction among firms [11]. In [25], which can eliminate the “double externalities”
July 2021, China’s nationwide carbon market to a large extent. However, some studies come to
began trading, marking the start of the carbon a different conclusion, that the CETP has no signifi­
market’s official operations. cant effect on GTI, or even inhibits the green
However, the development of the carbon emis­ innovation activities of firms [26,27], and that dif­
sion trading system is still immature, mainly focus­ ferent types of environmental regulations have a
ing on the emission reduction effects in pilot cities non-linear relationship with green economic devel­
[12]. Regarding the impact of CETP on the FP of opment [28]. In addition, other scholars’ studies
enterprises, scholars at home and abroad have not show that China’s CETP has unique regional effects
yet reached a consensus. Some scholars believe on green innovation, and not all pilot trading sys­
that CETP can effectively improve the carbon emis­ tems are effective in stimulating local green innov­
sion efficiency in pilot areas [13]. By optimizing the ation [29].
production process, it promotes the optimization At present, there is a lack of research on the
and upgrading of the industrial structure [14], "Strong Porter’s hypothesis" of whether CETP can
helps enterprises gain an advantage in resource improve firm FP by GTI, and only some papers have
allocation, and thus improves the FP of enterprises. theoretically illustrated the relationship among the
In addition, some scholars believe that the inten­ three. Theoretically, CETP can incentive firms to carry
sity of environmental regulation is also positively out technological innovation, reduce long-term
correlated with the long-term profitability - main­ abatement costs, have good energy-saving and
taining ability of enterprises. However, other schol­ emission reduction effects [6,30], and have synergis­
ars hold that the CETP may reduce the FP, tic emission reduction effects on other pollutants
especially in the short term. Implementing the [31], and enable firms to own more relative quotas,
CETP requires bearing additional compliance costs increase their operating income, and thus improve
[15], which has a negative impact on the FP of their FP [32]. Some scholars’ empirical studies have
enterprises. Or enterprises may achieve carbon also concluded that the mediating effect of GTI
emission reduction by reducing production [16], between CETP and firm FP is not significant. Shen
and it may even lead to a financial crisis [17]. and Huang [33], Zhang et al. [34], found that the
Furthermore, differences in enterprise ownership emission reduction effect of carbon market is
structure, enterprise scale, and regional character­ achieved by enterprises at the expense of economic
istics can also affect the impact of CETP on enter­ goals, rather than the effect produced by long-term
prise FP [18]. emission reduction technology. Empirically believe
Led by the CETP, green measures such as car­ that CETP can effectively improve firm FP, but the
bon emission limitation, carbon emission fees, and implementation of carbon emissions trading cannot
carbon emission trading scheme have been promote Chinese firms to increase the level of
actively implemented at the firm level [19], and investment in R&D [35], nor can it improve the value
carbon emission databases have been established of the firm [36].
[8] to enhance energy conservation and the Existing research has extensively examined the
release of a green finance policy to enhance relationship between CETP, GTI and FP. However,
energy conservation and emission reduction [20]. two key areas remain underexplored: (1) a deeper
Many firms regard carbon emission reduction as investigation into the long-term impact
4 R. WANG ET AL.

mechanisms of CETP across different time horizons, trading, different firms can sell or buy carbon quo­
and (2) an analysis of the heterogeneous effects of tas freely in the carbon trading market [36], and
CETP across firms with varying characteristics, which firms with sufficient quotas will benefit from the
could aid in refining policy design. Addressing these cash flow obtained from a large number of free
gaps through the lenses of long-term impact and quotas [37], which promotes the optimal allocation
firm heterogeneity would offer a more comprehen­ of resources and improves the FP of firms.
sive evaluation of CETP’s overall effectiveness and Secondly, to maximize profits, firms will make a
provide valuable insights to support the sustainable trade-off between emission reduction costs and
development of a low-carbon economy. the purchase of carbon quotas, so the CETP can
help to reduce the cost of firm emission reduction,
improve FP, and increase the overall value of the
Methodology
firm. Therefore, the CETP can help reduce the cost
Research framework of firm emission reduction, improve FP, and
To realize the high-quality development of increase the overall value of firms [20,38]. From
Chinese firms, the following three questions are the traditional point of view, environmental regula­
tion may increase the cost of firm environmental
answered (Figure 2): (1) Does CETP promote the
governance, impose constraints on firm produc­
improvement of FP, and what is the role of firm
tion, and have a negative impact on FP [33].
GTI in the process? (2) What kind of differentiated
However, research from a dynamic perspective
impacts does CETP show in diverse regions, carbon
shows that CETP can bring additional economic
prices, and ownership of firms? (3) Does the
benefits through investment activities such as cost
impact of CETP on FP have a long-term effect? By
reduction, risk reduction, and trading gains, thus
answering the above questions, we can guide
enhancing the FP of firms.
firms to pay attention to the investment in GTI,
Based on the above analysis, hypothesis 1 is
and the lagging or long-term effect of CETP, and
proposed.
guide the government to think about the CETP
system according to local conditions. H1: CETP can enhance firms’ FP.

Research hypothesis
CETP and GTI
CETP and FP CETP is a market-based environmental regulatory
Some scholars have argued that there are syner­ tool centered on carbon trading price, which can
gies between CETP and FP. Firstly, based on quota stimulate firms to improve technology and R&D

Figure 2. Research ideas on the impact path of CETP on FP from the perspective of GTI.
 CARBON MANAGEMENT 5

innovation through the price mechanism. Based H3: CETP reduces FP by promoting firms’ GTI.
on the innovation theory, to realize sustainable
The specific theoretical relationship model is
and healthy development, firms will carry out GTI
shown in Figure 3:
to improve production efficiency and reduce car­
bon emissions, obtain more relative quotas, and
reduce long-term emission reduction costs [39]. In Variable settings
addition, firms with a higher level of GTI can more
Sample selection and data sources
actively disclose carbon emission information. On
This paper conducts an empirical study using panel
the one hand, it can directly and effectively allevi­
ate firm financing constraints, and obtain tax data of listed companies on the Shanghai and
reduction and green subsidies [40], reduce regula­ Shenzhen A-share in 8 pilot industries across 30 prov­
tory pressure, and receive potential benefits. On inces in China from 2010 to 2021. Listed companies
the other hand, low-carbon subsidies, as a kind of in the 8 pilot industries, namely petrochemical, chem­
"good signal", effectively reduce the information ical, building materials, steel, non-ferrous metals,
asymmetry between external investors and firms, paper - making, power, and aviation, in seven pilot
bring a large number of innovation resources for provinces and cities (Beijing, Shanghai, Tianjin,
firms, and urge firms to innovate behavior [26], Chongqing, Hubei, Guangdong, and Shenzhen) are
reflecting stronger social responsibility and better selected as the experimental group, while firms in
reputation. the same industries in non-pilot provinces and cities
Based on the above analysis, hypothesis 2 is serve as the control group. In addition, the following
proposed: treatments are applied to the sample: (1) Firms
marked with ST, �ST, etc. are excluded; (2) Firms with
H2: CETP can promote firms’ GTI.
severe data missing are excluded; (3) Firms listed
after 2010 are excluded. Finally, data of 453 enter­
Mediating effects of GTI prises and 5,436 valid observations are obtained.
The strong Porter’s hypothesis suggests that strict The financial data of listed companies, such as
and reasonable environmental regulations can enterprise size (Size), enterprise age (Age), total
force firms to carry out GTI and improve produc­ asset turnover (TUR), asset - liability ratio (DAR),
tion efficiency through technological and product and ownership concentration (OC), are sourced
innovation, thereby compensating for the environ­ from the China Stock Market & Accounting
mental costs invested in the previous period, and Research Database (CSMAR). The patent data used
although more resources invested in carbon posi­ to measure GTI come from the China National
tivity will reduce current FP, it will improve future Intellectual Property Administration, and other
FP. However, there are different conclusions about macroeconomic data are obtained from the China
the verification of this hypothesis in academics. Statistical Yearbook and the China Environment
Some scholars support the above view, finding the Yearbook.
impact of GTI inputs is lagging [41]. Some scholars
do not support the Strong Porter hypothesis, find­
ing that environmental regulations incentivize Independent variable
technological innovation, but do not lead to The explanatory variable is the multiplication term
improved business performance due to high (Treat � Time) of the time dummy variable (Time)
inputs [42]. To further explore the role of CETP on and the CETP variable (Treat). Since the opening of
firm FP by influencing GTI, a mediation effect test the carbon market in all seven pilot provinces and
is conducted. The previous analysis that CETP will cities was completed in 2014, the time dummy
prompt firms to carry out GTI. And GTI is a long- variable Time is taken as 1 in 2014 and later years,
term dynamic evolution of the process, high inputs and 0 in the rest of the years. If the firm is a listed
and longer return periods will make the firms’ firm in the eight pilot industries in the seven pilot
short-term decline. Therefore, it is assumed that provinces and cities and belongs to the experi­
the decline of FP is caused by the implementation mental group, the policy dummy variable Treat is
of CETP when companies carry out GTI activities. taken as 1 (Table 1). If the firm belongs to the
Based on the above analysis, hypothesis 3 is listed firms in the same industry in non-pilot prov­
proposed: inces and belongs to the control group, it takes 0.
The specific definitions are shown in Table 1.
6 R. WANG ET AL.

Figure 3. Theoretical relationship model.

Table 1. Independent variable.

Indicator name Implication Indicator values Meaning of value
Treat Whether the firm is in the pilot region of the carbon emission trading market 0 No
1 Yes
Time time dummy variable(Whether the firm is in a position to implement the policy in 0 No
the current year and the year after its implementation) 1 Yes
Treat � Time Whether the firm is in a carbon emission trading pilot region and at the same time, 0 No
whether it is in the year of implementation of carbon emission trading 1 Yes

Dependent variable into account the three aspects of firm characteris­

The dependent variable is FP. The measurement tics, financial status and firm governance, and it is
method of firms’ FP indicators can be roughly div­ proposed to select SIZE to respond to the eco­
ided into two categories. One is market-adjusted nomic strength of the firm; AGE to respond to the
indicators, such as Tobin’s Q value and EVA, etc. stability and credibility of the firm, and TUR to
The other is accounting-based indicators, such as respond to the efficiency of the utilization of
the EBIT and the return on total assets. However, assets; DAR reacts to the financial risk, solvency,
China’s market is weak, not yet a strong and effi­ profitability of the firm, and OC reacts to the con­
cient market, the use of market-adjusted indicators trol and governance structure of the firm, as con­
for research is more likely to work well than trol variables. The specific definitions are shown in
accounting-based indicators, and the relevance is Table 2.
weaker. So accounting-based indicators are used
to measure FP. Most scholars prefer to measure Empirical model
the FP of firms by Return on asset (ROA) and the
return on equity (ROE) in the profitability indica­ CETP provides a quasi-natural experimental sce­
nario that can overcome endogeneity. By compar­
tors [43]. Finally, the ROA is selected to measure
ing the difference between the effects of the
the FP of firms, and the ROE is adopted as a
experimental group and the control group before
replacement indicator for robustness testing.
and after the implementation of policies, it elimi­
nates factors that do not change over time and
Intermediary variable other unobservable factors and strips the disposi­
The mediating variable is GTI. Existing studies gen­ tional effects of the policies from confounders.
erally measure it through the number of green Therefore, DID is used to conduct a fixed-effects
patent applications, the number of eco-label certif­ test.
ications that have been obtained, and the unit
energy consumption of products. Considering the
Parallel trend test model construction
gradual increase in the availability and accuracy of
The DID method is a research approach based on
patent data, the number of firm green patent
fixed effects, mainly used for the inter-temporal
applications is drawn as a measure of GTI [44].
evaluation of policy implementation effects.
Specifically, the ratio of green patent applications
Compared with traditional policy evaluation meth­
to per ten million operating revenues is used to
ods, the DID method is more scientific in model
exclude the scale effect of business operations.
setting. The important premise is that before the
implementation of the policy, the treatment group
Control variables and the control group have the same trend, that
According to the previous literature [45,46], the is, in the absence of CETP intervention, the
selection of control variables in this paper takes dependent variable has the same trend in the
 CARBON MANAGEMENT 7

Table 2. Variable definition.

Variable Definition variable name variable symbol Variable Definition
Dependent variable Financial performance ROA Total Assets Net Ratio ¼ Net Profit
/ Average Total Assets
Independent variable time dummy variable Treat � Time If the firm is in both the carbon
emission trading pilot year and
the pilot province or city, take
1; for the rest, take 0.
Intermediary variable Green technology innovation GTI Green patent applications/10
million business revenue
Control variable firm Size Size Ln (Total assets at year-end)
firm Age Age Ln (Years of company
establishment þ 1)
Total asset turnover TUR Net operating income/average
total assets
Asset-liability ratio DAR Total liabilities/total assets
Equity concentration OC The shareholding ratio of the
firm’s largest shareholder

experimental group and the control group [47], located in a pilot province or city and 0 when it is
which can largely avoid the interference of endo­ located in a non-pilot province or city. The remaining
geneity problems.Therefore, the DID method has variables are the same as in model Equation (3.1).
become the mainstream approach for evaluating
policy implementation effects [48,49]. In this paper, The mediating effect of GTI model construction
we adopt the DID method and use the CETP as a This paper draws on the test analysis of Wen and
quasi - natural experiment to study the impact of Ye [50] on the mediating effect to carry out the
the CETP on the FP of Chinese enterprises before mediating effect test. The mediation effect models
and after its establishment. Based on this, we have are constructed as follows:
established the following DID model:
GTIit ¼ a0 þ a1 time � treatit þ kX þ yeari
ROAit ¼ q0 þ q1 treati þ q2 trendt þ q3 treat � trendit
þ companyt þ eit (3.3)
þ kX þ yeart þ companyi þ eit
ROAit ¼ h0 þ h1 time � treatit þ h2 GTIit þ kX þ yeari
(3.1)
þ companyt þ eit
“Trend” is the time trend variable, “Treat” is the (3.4)
implementation of CETP. When the firm is located
in the pilot provinces and cities, the value is 1. On GTIit is the mediating variable, and the rest of
the contrary, its value is 0. If q3 is not significant, it the variables are the same as in model (3.1).
means that there is no significant difference
between the experimental and control groups Results
before the pilot, satisfying the parallel trend
Parallel trend test
assumption and the prerequisites of DID. X repre­
sents a set of control variables, i.e. firm size, firm The trend chart of firms’ FP in the experimental
age, total asset turnover, asset-liability ratio, and and control groups for the period 2010-2021
equity concentration. Yeart and companyi denote shows initially that there is a common trend
the time-fixed effect and individual fixed effect, between the experimental and control groups
respectively. eit is the random error term. until the pilot year 2014 (Figure 4). There is no sig­
nificant difference between the FP of listed compa­
nies in pilot industries in pilot provinces and listed
CETP and FP model construction
companies in pilot industries in non-pilot provin­
To test the impact of CETP on FP, this paper con­
ces before the implementation of the CETP, and
structs a DID fixed-effects model as a basis regres­
there is a common trend. The prerequisites of DID
sion model as follows:
are satisfied.
ROAit ¼ b0 þ b1 time�treatit þ kX þ yeart
þ companyi þ eit (3.2) Analysis of regression results
ROAit denotes FP of firm i in the pilot industry in Table 3 reports the regression results of the impact
year t, “time” is a time dummy variable, taking 1 in of CETP on firms’ FP, with columns (1) and (2) indi­
2014 and after, and 0 before 2014. Treat is the cating the impact of CETP on FP after controlling
implementation of CETP, taking 1 when the firm is for individual and time-fixed effects. Column (1)
8 R. WANG ET AL.

Figure 4. Parallel trend test.

Table 3. Regression results of the impact of CETP on FP. Table 4. Regression results of the mediating effect of
(1) (2) firm GTI.
variant ROA ROA (1) (2) (3)
Treat � Time 0.174��� 0.091�� variant ROA GTI ROA
(3.81) (2.01) Treat � Time 0.091�� 0.097��� 0.098��
Size 0.409��� (2.01) (4.98) (2.17)
(14.28) GTI −0.075��
Age 0.060 (−2.36)
(1.26) Size 0.409��� −0.101��� 0.401���
TUR −0.076�� (14.28) (−8.26) (13.93)
(−2.34) Age 0.060 0.007 0.061
DAR 0.051� (1.26) (0.32) (1.27)
(1.85) TUR 0.076�� −0.057��� −0.080��
OC 0.004��� (2.34) (−4.14) (−2.47)
(3.28) DAR −0.051� −0.037��� 0.048�
Constant 23.021��� 18.779��� (−1.85) (−3.17) (1.74)
(1,223.68) (67.37) OC 0.004��� −0.002��� 0.004���
Individual fixed effect Yes Yes (3.28) (−5.10) (3.11)
Time fixed effect Yes Yes Constant 18.779��� 1.237��� 18.872���
sample size 5436 5436 (67.37) (10.37) (67.07)
R-squared 0.645 0.648 Individual fixed effect Yes Yes Yes
Note: �、��、���denote the 10%, 5%, and 1% significance levels, Time fixed effect Yes Yes Yes
respectively, and the values in parentheses are t-values. sample size 5436 5,436 5,436
R-squared 0.648 0.464 0.648
Note: �、��、���denote the 10%, 5%, and 1% significance levels,
shows the regression results without adding con­ respectively, and the values in parentheses are t-values.
trol variables, yielding a regression coefficient of
0.174 for CETP(Treat � Time) and FP(ROA), which is costs but also enable enterprises to obtain add­
significantly positive at the 1% significant level. itional income by selling surplus quotas [45],
Column (2) is the regression result of adding con­ thereby improving corporate FP. In addition, the
trol variables, and it is concluded that the regres­ CETP can also promote energy conservation and
sion coefficient of the two is 0.091, and it is emission reduction, enhance stakeholders’ confi­
significantly positive at the 5% significant level, dence in the firm, and help the firm obtain more
and this result is the same as the research of Lin favorable resources to improve its market competi­
Zhihong (2022), and it is concluded that the CETP tiveness, thus increasing FP [52].
can significantly improve the FP of firms, support­
ing hypothesis 1.
Analysis of mediation effects
This result shows that the CETP not only
reduces carbon emissions but also improves eco­ Table 4 reports the regression results of the media­
nomic performance and realizes the synergistic ting effect of firms’ GTI. Column (2) tests the
development of the economy. By exploring the impact of CETP on firms’ GTI, and the regression
reasons, it is found that the implementation of the coefficient of the CETP (Treat � Time) is 0.097,
CETP helps to internalize carbon emission costs, which is significant at the 1% significant level, sug­
prompting enterprises to optimize internal gesting that the CETP of the pilot region can sig­
resource allocation, reduce energy consumption, nificantly promote firms’ GTI. Column (3) is used to
and carbon emissions [51]. This mechanism can test the impact of CETP (Treat � Time) on firm FP
not only effectively reduce corporate production (ROA) through GTI. As can be seen from the table,
 CARBON MANAGEMENT 9

Table 5. Results of sobel’s test for mediated effects. Table 6. Bootstrap test results for mediated effects.
Results of Sobel’s test GTI 95% significant interval
Sobel Z value −2.134�� Bootstrap test results Effect BootSE minimum maximum
Indirect effect −0.007�� Direct effect 0.098 0.047 0.013 0.198
(−2.134) Indirect effect −0.007 0.003 −0.014 −0.001
Direct effect 0.098�� Total effect 0.091 0.047 0.007 0.190
(2.165)
Total effect 0.091�� Note: �、��、���denote the 10%, 5%, and 1% significance levels,
(2.010) respectively, and the values in parentheses are t-values.
Note: �、��、���denote the 10%, 5%, and 1% significance levels,
respectively, and the values in parentheses are t-values. financial performance (FP) while GTI has a damp­
ening effect. Finally, using ROE as an alternative FP
the regression coefficient of GTI is −0.075, which is indicator supports previous findings, reinforcing
significantly negative at the 5% significant level, the significance of GTI in the CETP-FP relationship.
indicating that GTI has a mediating effect, and firm Overall, these tests affirm the reliability of the
GTI has an inhibitory effect on FP. The regression study’s conclusions.
coefficient of CETP (Treat � Time) is 0.098, which is
significantly positive at the 5% significant level,
Robustness test
indicating that CETP has a direct effect on FP.
According to the test analysis of the mediation Placebo test
effect model, it can be seen that there is a "mask­ To ensure the robustness and reliability of the
ing effect" of GTI on the impact of CETP on FP. research conclusions, this paper conducted a pla­
The research results show that GTI plays a medi­ cebo test. Given the limitation of the data sample
ating role between the CETP and corporate FP. size, this paper changed the pilot start time of the
The reasons are as follows: The implementation of CETP to test whether a randomly set policy time
the CETP sets carbon emission limits and trading could produce a significant treatment effect.
mechanisms, which further forces enterprises to Specifically, this paper advanced the pilot time of
carry out GTI [26,34] to meet the requirements of the CETP by two years, assuming that the pilot
compliant operation. Secondly, GTI helps to work started in 2012, and retested it according to
develop clean production technologies and pro­ the benchmark regression model mentioned ear­
duce more green products or services to meet lier to verify the robustness of the policy effect. In
market demands. This can not only expand the the placebo test, in order to maintain the consist­
market share of enterprises but also enhance their ency of the treatment group and the control
brand image [53]. Therefore, the CETP promotes group with the original study, the setting of the
corporate GTI. However, the short term investment treatment group and the control group in this
in corporate GTI is large, and the transformation of paper was consistent with the original study. If the
green innovation achievements is slow, which can­ original research conclusion was driven by the dif­
not be promptly reflected in FP. So in the short ference in unobservable characteristics between
term, GTI has a significant inhibitory effect on FP the treatment group and the control group, then
[54,55]. this virtual policy setting should be able to draw
To further verify whether this mediating effect conclusions similar to the original research.
exists, this paper conducts the Sobel test on the Columns (1) and (2) in Table 7 show the regression
model and adds the Bootstrap test [56]. The indir­ results of FP (ROA) without control variables and
ect effect of the two tests is −0.007, the direct with control variables, respectively. The results
effect is 0.098, and the total effect is 0.091, all of show that, regardless of whether control variables
which are significant at the 5% level, and the dir­ are included, the regression coefficient of the CETP
ect effect is greater than the total effect, indicating is not significant. This result indicates that the pos­
that the masking effect is established (Tables 5 sibility of omitted variables or inherent differences
and 6). between the treatment group and the control
To be clear, the robustness of the results is con­ group before the policy implementation is low,
firmed through various tests. The placebo test, which further supports the robustness of the
advancing the CETP implementation date, shows research conclusions of this paper.
no significant coefficients, suggesting minimal
interference from omitted variables. The PSM-DID PSM-DID model
model, which accounts for differences among Due to the heterogeneity of sample companies in
firms, indicates that CETP positively impacts firm macro environment, corporate resources, and
10 R. WANG ET AL.

Table 7. Results of placebo test. Table 8. Regression results of PSM-DID model.

(1) (2) (1) (2) (3)
Variables ROA ROA Variables ROA GTI ROA
Treat � Time 0.048 0.050 Treat � Time 0.091�� 0.095��� 0.098��
(0.75) (0.79) (2.10) (4.86) (2.16)
Control Variables No Yes GTI −0.077��
Constant 24.101��� 26.414��� (−2.42)
(103.09) (38.66) Control Variables Yes Yes Yes
Individual Fixed Effects Yes Yes Constant 18.785��� 1.238��� 18.880���
Time Fixed Effects Yes Yes (67.26) (10.35) (66.97)
Sample Size 5,436 5,436 Individual Fixed Effects Yes Yes Yes
R-squared 0.645 0.648 Time Fixed Effects Yes Yes Yes
Note: �、��、���denote the 10%, 5%, and 1% significance levels, Sample Size 4968 4968 4968
respectively, and the values in parentheses are t-values. R-squared 0.649 0.465 0.649
Note: �、��、���denote the 10%, 5%, and 1% significance levels,
respectively, and the values in parentheses are t-values.
other aspects, this paper used the PSM-DID model
for robustness tests [57]. The treatment group of
Table 9. Results of alternative indicator test for financial
this paper is the listed companies in the pilot performance.
industries of the pilot provinces and cities, and the (1) (2) (3)
control group is the listed companies in the pilot Variables ROE GTI ROE
industries of the non-pilot provinces and cities, Treat � Time 0.153�� 0.053� 0.156��
(2.60) (1.68) (2.53)
with 116 companies in the treatment group and GTI −0.057��
(−1.85)
337 companies in the control group. In the match­ Control Variables Yes Yes Yes
ing process, this paper selected region, industry, Constant 1.185��� 0.769� 1.186���
(4.33) (1.69) (4.33)
total asset growth rate, and financial leverage as Individual Fixed Effects Yes Yes Yes
matching variables, and used the Logit model to Time Fixed Effects Yes Yes Yes
样本数 5436 5,436 5,436
calculate the propensity score, with the implemen­ R-squared 0.358 0.327 0.358
tation of the CETP as the explained variable. Note: �、��、���denote the 10%, 5%, and 1% significance levels,
respectively, and the values in parentheses are t-values.
Subsequently, this paper conducted one-to-one
matching for the treatment group and the control As can be seen from the regression results in
group according to the propensity score [58]. After Table 9, the impact coefficient of the CETP on cor­
the matching was completed, this paper used the porate FP (ROE) in column (1) is 0.153, which is sig­
matched samples to construct the DID model for nificantly positive at the 5% level. The impact
regression analysis. After PSM matching, 39 com­ coefficient of the CETP on corporate GTI in column
panies were excluded from the common support (2) is 0.053, which is significantly positive at the
range, with 1 in the treatment group and 38 in the 10% level. The coefficient of GTI in column (3) is −
control group, and finally 414 companies were suc­ 0.057, which is significant at the 5% confidence
cessfully matched. The balance test results after level, and the direct effect coefficient of the CETP
matching show that the data has achieved a good (Treat � Time) on FP (ROE) is 0.156, which is signifi­
balance on the matching variables. The benchmark cant at the 5% confidence level.The positive and
regression results based on the PSM-DID model negative signs and significance of the regression
are shown in Table 8. The results show that the coefficients have not changed significantly, which
CETP has a significant promoting effect on corpor­ is basically consistent with the above - mentioned
ate FP, and at the same time, this policy has a cer­ empirical conclusions. This shows that the media­
tain inhibiting effect on FP through promoting ting role of GTI between the CETP and FP is signifi­
corporate GTI. This result is basically consistent cant, further verifying the robustness of the
with the previous DID regression results, further research conclusions of this paper.
verifying the robustness of the research conclu­
sions of this paper.
Further discussion
Long-term effects analysis
Alternative indicator test
To test the robustness of the results, this paper fol­ A test of the long-term effect of CETP on FP
lowed and used the Return on Equity (ROE) as an FP with a lag of one to three periods has been
alternative variable for corporate FP to test the analyzed. The results show that the impact of CETP
promoting effect of the CETP on FP and the medi­ on FP with a lag of one to three periods is signifi­
ating effect of GTI. cantly positive (Table 10), indicating that CETP has
 CARBON MANAGEMENT 11

Table 10. Regression results of the long-term effects of CETP.

One period behind Two periods behind Three periods behind
variant ROA ROA ROA
Treat � Time 0.103�� 0.121��� 0.145���
(2.26) (2.63) (3.09)
Size 0.402��� 0.394��� 0.387���
(13.24) (12.29) (11.31)
Age 0.108�� 0.180��� 0.258���
(2.03) (3.00) (3.85)
TUR −0.071�� −0.076�� −0.064�
(−2.10) (−2.10) (−1.68)
DAR 0.055 0.049 0.041
(1.50) (1.31) (1.04)
OC 0.004��� 0.004��� 0.004���
(3.07) (2.95) (2.80)
Constant 18.688��� 18.532��� 18.344���
(61.75) (56.31) (50.91)
Individual fixed effect Yes Yes Yes
Time fixed effect Yes Yes Yes
sample size 4,983 4,530 4,077
R-squared 0.657 0.671 0.687
Note: �、��、���denote the 10%, 5%, and 1% significance levels, respectively, and the values in parentheses are t-
values.

a long-term promotion effect on firms’ FP. The sig­ The test shows that the regression coefficients
nificance level of the impact of the policy on FP of CETP are 0.110, 0.127, and 0.149 respectively,
increases, and the regression coefficient gradually indicating that the direct effect of CETP on FP is
increases, indicating that with the gradual gradually increasing. The regression coefficients of
improvement of the carbon emission trading mar­ GTI are −0.070, −0.063, and −0.056 respectively,
ket, the policy has a more and more obvious effect indicating that the inhibitory effect of GTI on FP is
on the promotion of FP. gradually weakened. In the short, GTI increases
firm development costs and reduces firm FP.
A long-term test of the mediating effect of GTI. However, in the long run, firms actively carrying
Based on the previous analysis, it is concluded that out GTI can not only reduce the marginal abate­
the impact of CETP on FP is long-term. Therefore, ment cost and reduce the cost expenditure of
we would like to further study whether the conclu­ firms purchasing carbon emission quotas in the
sion that CETP reduces FP by promoting GTI is also long term [28] but also improve the efficiency of
long-term in nature. The methods of other scholars resource utilization, increase production efficiency,
are referred to do the regression of the effect enrich product diversification [60].
The research of Zhou et al. [32] shows that the
based on the mediation effect regression model to
CETP has a long-term effect on carbon emission
explore the long-term nature of the mediation
reduction. This paper studies the long-term impact
effect of GTI [59].
of the CETP on corporate GTI and FP. Through con­
To explore the long-term effect, the mediation
tinuous market-based mechanisms, enterprises can
effect model including the lagged effect is con­
be encouraged to incorporate green and low-car­
structed as follows:
bon development into their strategic plans and
ROAit ¼ c0 þ c1 time � treatin þ kX þ yeari gradually transform from a high-carbon to a low-
þ companyt þ eit (5.1) carbon operation model. This can reduce environ­
mental compliance costs and improve resource
GTIit ¼ g0 þ g1 time � treatin þ kX þ yeari utilization efficiency in the long run. However, in
þ companyt þeit (5.2) the initial stage of GTI, enterprises need to invest a
large amount of R&D funds and resources, which
ROAit ¼ l0 þ l1 time � treatin þl2 GTIit þkX þ yeari will have an inhibitory effect on FP [61]. But as the
þ companyt þeit technology gradually matures, the R & D costs are
(5.3) amortized, production efficiency is improved, and
the economic benefits of green technology grad­
The regression of lag behind the first and third ually emerge, and the inhibitory effect weakens.
periods is constructed, and the subscript n can be Secondly, the market acceptance and demand for
taken as t-1, t-2, and t-3; when n ¼ t-1, it means green technology gradually increase. The products
lag behind the first period. The regression results and services developed by enterprises through GTI
are shown in Table 11. can better meet market demands and enhance
12 R. WANG ET AL.

Table 11. Mediated effects regression results including lagged effects.

One period behind Two periods behind Three periods behind
(5.1) (5.2) (5.3) (5.1) (5.2) (5.3) (5.1) (5.2) (5.3)
variant ROA GTI ROA ROA GTI ROA ROA GTI ROA
Treat � Time 0.103�� 0.091��� 0.110�� 0.121��� 0.085��� 0.127��� 0.145��� 0.077��� 0.149���
(2.26) (4.50) (2.40) (2.63) (4.13) (2.74) (3.09) (3.61) (3.18)
GTI −0.070�� −0.063� −0.056�
(−2.17) (−1.91) (−1.65)
Control variable Yes Yes Yes Yes Yes Yes Yes Yes Yes
Constant 18.688��� 1.346��� 18.782��� 18.532��� 1.446��� 18.624��� 18.344��� 1.628��� 18.436���
(61.75) (10.09) (61.46) (56.31) (9.81) (56.01) (50.91) (9.87) (50.58)
Individual fixed effect Yes Yes Yes Yes Yes Yes Yes Yes Yes
Time fixed effect Yes Yes Yes Yes Yes Yes Yes Yes Yes
sample size 4,983 4,983 4,983 4,530 4,530 4,530 4,077 4,077 4,077
R-squared 0.657 0.484 0.657 0.671 0.512 0.671 0.687 0.532 0.687
Note: �、��、���denote the 10%, 5%, and 1% significance levels, respectively, and the values in parentheses are t-values.

Table 12. Impact of CETP on FP in pilot provinces and cities.

(1) (2) (3) (4) (5)
Beijing Guangdong Shanghai Shenzhen Hubei
variant ROA ROA ROA ROA ROA
Treat � Time 0.323��� 0.146� 0.225�� 0.110 −0.132�
(2.61) (1.71) (2.53) (1.23) (−1.78)
Size 0.215��� 0.230��� 0.280��� 0.330��� 0.259���
(6.04) (6.74) (8.50) (9.82) (7.47)
Age 0.146��� 0.123�� 0.106�� 0.041 0.171���
(2.68) (2.37) (1.98) (0.78) (3.32)
TUR −0.001 −0.034 −0.041 −0.022 0.005
(−0.04) (−1.03) (−1.14) (−0.60) (0.14)
DAR 0.324��� 0.245��� 0.043 0.246��� 0.227���
(3.79) (2.94) (1.14) (2.89) (2.70)
OC 0.006��� 0.006��� 0.005��� 0.006��� 0.005���
(4.42) (4.74) (3.74) (5.01) (4.55)
Constant 20.154��� 20.112��� 19.822��� 19.364��� 19.681���
(60.59) (62.82) (62.86) (60.93) (60.61)
Individual fixed effect Yes Yes Yes Yes Yes
Time fixed effect Yes Yes Yes Yes Yes
sample size 4,183 4,357 4,340 4,343 4,275
R-squared 0.623 0.620 0.630 0.623 0.618
Note: �、��、���denote the 10%, 5%, and 1% significance levels, respectively, and the values in parentheses are t-
values.

market competitiveness, further weakening the ini­ The regression results show that the intermediary
tial inhibitory effect. Finally, the CETP helps enter­ effect of GTI of pilot firms in the three regions is
prises to adapt in advance to potentially stricter significant, but the strength of the intermediary
environmental regulations in the future, enhancing effect varies. Among them, CETP has the strongest
the stability and sustainability of corporate promotional effect on the FP of firms in the pilot
operations. industries in Beijing, and the intermediary effect of
GTI accounts for the largest absolute value of the
total effect (Table 13).
Heterogeneity analysis
Analysis of regional heterogeneity Carbon price heterogeneity analysis
CETP improves the FP of the pilot firms in Beijing, Under the institutional arrangement of the carbon
Guangdong, and Shanghai, but it cannot signifi­ market, firms will decide whether to adopt GTI for
cantly improve the FP of the pilot firms in autonomous emission reduction or to purchase
Shenzhen and Hubei (Table 12), which reflects the carbon quotas for emission reduction through the
heterogeneity of the implementation effect of the carbon market by comparing the estimation of
policy in different regions. investment in GTI and their own marginal emis­
The study of the intermediary effect of GTI of sion reduction cost, to maximize the benefits.
firms in different provinces and cities, according to Therefore, the relationship between CETP, GTI, and
the previous empirical evidence concluded that FP is related to the level of carbon emission trad­
CETP can improve FP of firms in the pilot indus­ ing price. In this paper, we divide the price [34]
tries of Beijing, Guangdong Province, and and take the lower quartile of the daily closing
Shanghai, so it is only necessary to verify the inter­ price of carbon trading in seven pilot provinces
mediary effect of GTI of firms in the three regions. and cities in 2013-2021 as the standard. The
 CARBON MANAGEMENT 13

Table 13. Mediating effects of firms’ GTI in different provinces and cities.
Beijing Guangdong Shanghai
variant ROA GTI ROA ROA GTI ROA ROA GTI ROA
Treat � Time 0.323��� 0.269��� 0.341��� 0.146� 0.150��� 0.154� 0.221�� 0.057� 0.225��
(2.61) (5.25) (2.74) (1.71) (3.94) (1.81) (2.49) (2.10) (2.53)
GTI −0.065� −0.055� −0.065�
(−1.72) (−1.68) (−1.69)
Control variable Yes Yes Yes Yes Yes Yes Yes Yes Yes
Constant 20.154��� 1.122��� 20.227��� 20.112��� 1.233��� 20.179��� 19.822��� 1.158��� 19.898���
(60.59) (8.17) (60.34) (62.82) (8.61) (62.51) (62.86) (9.36) (62.48)
Individual fixed effect Yes Yes Yes Yes Yes Yes Yes Yes Yes
Time fixed effect Yes Yes Yes Yes Yes Yes Yes Yes Yes
sample size 4,183 4,183 4,183 4,357 4,357 4,357 4,340 4,340 4,340
R-squared 0.623 0.407 0.623 0.620 0.477 0.620 0.630 0.340 0.630
Note: �、��、���denote the 10%, 5%, and 1% significance levels, respectively, and the values in parentheses are t-values.

Table 14. Daily closing prices of carbon emission rights quarter points in provinces and cities (unit:
Yuan/ton).
High carbon price area Low carbon price area
standard price Beijing Shanghai Guangdong Hubei Shenzhen Tianjin Chongqing
17.465 50.200 31.978 18.317 16.748 13.510 12.860 3.490

provinces and cities with a daily closing price of high-emission enterprises, which may lead to a
the quarter-points that are higher than the stand­ temporary decline in FP. On the contrary, in
ard price during this period are categorized as the regions with low carbon prices, the carbon emis­
high-carbon price area and the provinces and cit­ sion costs of enterprises are relatively low, which
ies with prices lower than the standard price are may reduce the financial burden in the short term
categorized as the low-carbon price area. The daily but may also weaken the motivation of enterprises
closing lower quartile prices and the area for each to carry out GTI, affecting the long-term improve­
pilot province and city in 2013-2021 are shown in ment of competitiveness and FP. Therefore, the
Table 14. heterogeneity of carbon prices not only affects
Benchmark regression and intermediary effect corporate GTI but also further affects changes in
regression are carried out for high-carbon price corporate FP.
area firms and low-carbon price area firms respect­
ively. The results show that CETP can significantly Analysis of property rights heterogeneity
improve the FP of high carbon price area firms; at Considering the special institutional background of
the same time, CETP has a promoting effect on the China, there are many differences in the business
GTI of firms in both high carbon price area and models of state-owned firms and private firms
low carbon price areas. It has a stronger promoting under different ownership types. In this paper,
effect on GTI of high carbon price area firms. The according to the nature of property rights, the
GTI of firms in high-carbon price areas has an sample is divided into state-own-firms and non-
intermediary effect between CETP and FP, while state-own-firms, and the empirical results show
the GTI of firms in low-carbon price areas does not that compared with non-state-owned firms, the
have an intermediary effect (Table 15). CETP has a more significant role in the promotion
The volatility and regional differences of carbon of the FP of state-owned firms and GTI, and the
prices directly affect the carbon emission costs and mediating effect of the GTI of state-owned firms is
benefits of enterprises, and thus have different more significant (Table 16).
impacts on FP. Many enterprises purchase carbon
quotas to achieve carbon emission reduction tar­
Conclusion and policy implications
gets. In regions with high carbon prices, enter­
prises will face higher carbon emission costs, Against the background of China’s Dual Carbon
which forces them to increase investment in GTI, Goals and high-quality economic development,
optimize production processes, and reduce carbon the Difference-in-difference method is used to
emission intensity. This can achieve cost savings explore the impact of carbon emissions trading
and efficiency improvements in the long run and policies on firm financial performance by using A-
have a positive impact on FP. However, in the share listed companies in eight pilot industries in
short term, high carbon prices may increase the seven pilot regions from 2010-2021. The results
operating pressure of enterprises, especially for show that the CETP significantly enhances firm FP
14 R. WANG ET AL.

Table 15. Tests for differences in carbon price heterogeneity.

High carbon price area Low carbon price area
(1) (2) (3) (4) (5) (6)
Model(3.2) Model(3.3) Model(3.4) Model(3.2) Model(3.3) Model(3.4)
variant ROA GTI ROA ROA GTI ROA
Treat � Time 0.234��� 0.150��� 0.247��� 0.043 0.068��� 0.047
(3.32) (5.05) (3.50) (0.80) (3.02) (0.87)
GTI −0.086�� −0.056�
(−2.45) (−1.66)
Size 0.331��� −0.103��� 0.322��� 0.364��� −0.105��� 0.358���
(10.22) (−7.53) (9.89) (12.17) (−8.34) (11.90)
Age 0.158��� 0.007 0.159��� 0.023 0.009 0.023
(3.03) (0.32) (3.04) (0.46) (0.43) (0.47)
TUR 0.000 −0.046��� −0.004 −0.083�� −0.065��� −0.087���
(0.01) (−3.08) (−0.10) (−2.52) (−4.61) (−2.62)
DAR 0.076�� −0.053��� 0.071�� 0.049 −0.059��� 0.045
(2.17) (−3.58) (2.04) (1.28) (−3.67) (1.19)
OC 0.005��� −0.002��� 0.005��� 0.005��� −0.002��� 0.004���
(4.31) (−4.22) (4.15) (3.86) (−3.10) (3.78)
Constant 19.113��� 1.242��� 19.220��� 19.288��� 1.253��� 19.359���
(61.08) (9.39) (60.87) (66.86) (10.31) (66.40)
Individual fixed effect Yes Yes Yes Yes Yes Yes
Time fixed effect Yes Yes Yes Yes Yes Yes
sample size 4,525 4,525 4,525 4,954 4,954 4,954
R-squared 0.639 0.418 0.639 0.634 0.458 0.634
Note: �、��、���denote the 10%, 5%, and 1% significance levels, respectively, and the values in parentheses are t-values.

Table 16. Tests for differences in property rights heterogeneity.

State-owned firms non-state-owned firms
(1) (2) (3) (4) (5) (6)
Model(3.2) Model(3.3) Model(3.4) Model(3.2) Model(3.3) Model(3.4)
variant ROA GTI ROA ROA GTI ROA
Treat � Time 0.314��� 0.041��� 0.320��� 0.0004 0.112�� 0.003
(4.50) (5.71) (4.57) (0.01) (2.68) (0.06)
GTI −0.124�� −0.034
(−2.17) (−0.92)
Size 0.386��� −0.072��� 0.377��� 0.228��� −0.124��� 0.224���
(9.13) (−5.01) (8.89) (5.16) (−5.48) (5.03)
Age −0.636��� 0.041 −0.631��� 0.281��� 0.008 0.281���
(−7.00) (1.33) (−6.94) (5.06) (0.27) (5.06)
TUR −0.092�� −0.055��� −0.099�� −0.036 −0.064��� −0.038
(−2.03) (−3.58) (−2.18) (−0.79) (−2.73) (−0.84)
DAR 0.218� −0.130��� 0.202� 0.010 −0.033�� 0.008
(1.95) (−3.41) (1.80) (0.35) (−2.37) (0.31)
OC −0.001 −0.001 −0.002 0.003� −0.004��� 0.002
(−0.82) (−1.03) (−0.86) (1.65) (−4.97) (1.56)
Constant 21.278��� 0.815��� 21.379��� 19.793��� 1.509��� 19.844���
(46.92) (5.26) (46.93) (47.21) (7.03) (46.92)
Individual fixed effect Yes Yes Yes Yes Yes Yes
Time fixed effect Yes Yes Yes Yes Yes Yes
sample size 2628 2628 2628 2,808 2,808 2,808
R-squared 0.634 0.464 0.634 0.640 0.461 0.640
Note: �、��、���denote the 10%, 5%, and 1% significance levels, respectively, and the values in parentheses are t-values.

and this promotion has a long-term effect, which impact of carbon emissions trading, a govern­
implies that the company’s development prospects ment-guided market economic policy, on the FP of
are better. Secondly, CETP reduces current FP by firms in recent years, and at the same time enrich
promoting firm GTI, but the masking effect of GTI the theoretical research on the micro level of GTI.
gradually diminishes in the long run. In addition, It guides the sustainable development of firms and
the impact of CETP on FP in different provinces provides a reference for relevant departments to
and cities is differentiated, and the promotion of improve the trading system of the carbon market.
FP is more significant for firms in high-carbon price Based on the research in this paper, the follow­
areas and SOEs. After the robustness tests such as ing suggestions are made: First, the government
the placebo test, PSM-DID test, and alternative should encourage firms to carry out GTI. Local car­
indicator test, the conclusions still support the bon emissions trading markets can provide differ­
hypothesis proposed in this paper. ent carbon emission quotas according to the
The findings of this paper expand and enrich green innovation performance of firms. At the
the research on the economic consequences of same time, it should gradually weaken the division
the CETP, discuss more comprehensively the of the regional carbon market, accelerate the
 CARBON MANAGEMENT 15

construction of a national integrated carbon pric­ Econ Manage. 2023;45(06):161–176. doi:10.19616/j.

ing mechanism, and establish a national unified cnki.bmj.2023.06.009.
03. Zeng CL, Liu L, Li JT, et al. Environmental assessment
carbon emissions monitoring system and account­
and firm green innovation: a quasi-natural experi­
ing standards. Second, the effect of CETP is ment based on pilot natural resource asset audit for
affected by firm ownership, the government outgoing leadership in official academic context.
should pay more attention to NSOEs, provide a fair Account Res. 2022;(03):107–122.
financing platform for NSOEs, narrow the gap 04. Cui JB, Dai J, Wang ZX, et al. Does environmental
between NSOEs and SOEs in terms of resources regulation induce green innovation? A panel study of
Chinese listed firms. Technol Forecasting Soc Change.
and information channels, and encourage NSOEs
2022;176:121492. doi:10.1016/j.techfore.2022.121492.
to actively carry out innovative activities. Thirdly, 05. Hong QQ, Cui LH, Hong PH. The impact of carbon
to make the national carbon market operate stably emission trading on energy efficiency: evidence from
in the long run, it is important to have laws and quasi-experiment in China’s carbon emission trading
regulations to follow, and the regulatory author­ pilot. Energy Econ. 2022;110:106025. doi:10.1016/j.
ities should introduce relevant laws and regula­ eneco.2022.106025.
06. Hu Y, Ren S, Wang Y, et al. Can carbon emission trad­
tions as soon as possible.
ing scheme achieve energy conservation and emis­
However, since the implementation of the CETP sion reduction? Evidence from the industrial sector in
is still ongoing, future research could further China. Energy Econ. 2020;85:104590. doi:10.1016/j.
extend the observation period and consider a eneco.2019.104590.
broader range of indicators, such as the ESG per­ 07. Tang M, Walsh G, Lerner D, et al. Green innovation,
formance of enterprises, to comprehensively evalu­ managerial concern, and firm performance: an empir­
ical study. Bus Strat Env. 2018;27(1):39–51. doi:10.
ate the multi-dimensional effects of the policy.
1002/bse.1981.
Secondly, this paper mainly analyzes the impact of 08. Chen Y, Xu Z, Zhang Z, et al. Does the carbon emis­
CETP on the FP of enterprises. As the global efforts sion trading scheme boost firm environmental and
to address climate change intensify, the role of car­ financial performance in China? J Cleaner Prod. 2022;
bon sinks in the carbon emission trading system is 368:133151. doi:10.1016/j.jclepro.2022.133151.
becoming increasingly prominent. Future research 09. Dales JH. Land, water, and ownership. Can J Econ.
1968;1(4):791–804.
can combine the dynamic changes in the carbon
10. Coase RH. The problem of social cost: the citations.
sink market to analyze how enterprises can opti­ Chi.-Kent L. Re-v. 1995;71:809.
mize their carbon emission trading strategies 11. Cai J, Luo DN, Xiao XY. Analysis of the Emission
through carbon sink projects and thereby enhance Reduction Effects of Pilot Carbon Emission Rights
their FP. Trading Policies: from the Perspectives of Carbon
Emission Volume and Carbon Emission Efficiency.
Syst Engin Theory Pract. 2024;44(09):2838–2858.
Funding 12. Ma G, Qin J, Zhang Y. Does the carbon emissions trad­
ing system reduce carbon emissions by promoting two-
This work was supported by the he National Social Science
way fdi in developing countries? Evidence from Chinese
Foundation of China “Research on Co-ootive Strategy and
listed companies and cities. Energy Econ. 2023;120:
Realization Mechanism of Chinese Cobalt Supply Chain
106581. doi:10.1016/j.eneco.2023.106581.
Members Based on ESG (Grant No. 24CGL100); Philosophy
13. Chen Y, Mu H. Natural resources, carbon trading poli­
and Social Science Project of Hubei Provincial Department
cies and total factor carbon efficiency: a new direc­
of Education “Research on Realization Path and
tion for China’s economy. Resour. Pol. 2023;86:
Countermeasures of Ecological Product Value in Qingjiang
104183. doi:10.1016/j.resourpol.2023.104183.
River Basin under Dual Carbon Target (Grant No. 23Q179).
14. Tan J, Zhang JH. Has the carbon trading mechanism
forced industrial structure upgrading? – An analysis
Disclosure statement based on synthetic control method. Econ Manage
Stud. 2018;39(12):104–119.
No potential conflict of interest was reported by the 15. Xu J, Li HY, Han XX. The impact effect of market-
author(s). based environmental regulation on enterprise value:
evidence from the pilot policy of carbon emission
rights trading. China Popul Resources Environ. 2024;
References
34(07):88–100.
01. Handfield, Robert, Soumen Ghosh, and Stanley 16. Zhang H, Duan M. China’s pilot emissions trading
Fawcett. “Quality-Driven Change and Its Effects on schemes and competitiveness: an empirical analysis of
Financial Performance.” Qual Manag J. 1998;5(3):13– the provincial industrial sub-sectors. J Environ Manage.
30. doi:10.1080/10686967.1997.11918813. 2020;258:109997. doi:10.1016/j.jenvman.2019.109997.
02. Wang B, Kang Q. Digital transformation and the per­ 17. Zhao X, Shang Y, Ma X, et al. Does carbon trading
formance of enterprise sustainable development. lead to green technology innovation: recent evidence
16 R. WANG ET AL.

from Chinese companies in resource-based indus­ trading——based on the “Porter Hypothesis” & PSM-
tries. 2022. DID Test[J]. collected essays on finance and econom­
18. Lv M, Bai M. Evaluation of China’s carbon emission ics, 2020;36(3):68.
trading policy from firm innovation. Financ. Res. Lett. 33. Shen HT, Huang N. Can carbon emission trading
2021;39:101565. doi:10.1016/j.frl.2020.101565. mechanism enhance firm Value. Finan Trade Econ.
19. Stagliano AJ. Carbon trading reporting: the case of 2019;40(01):144–161.
Spanish companies. Int Adv Econ Res. 2017;23(2): 34. Zhang Q ,Zheng Y ,Kong D. Regional environmental
231–243. doi:10.1007/s11294-017-9634-z. governance pressure, executive experience, and firm
20. Zhang Z, Wang J, Feng C, et al. Do pilot zones for environmental investment - A quasi-natural experi­
green finance reform and innovation promote energy ment based on Ambient Air Quality Standards (2012).
savings? Evidence from China. Energy Econ. 2023; Econ Res. 2019;54(06):183–198.
124:106763. doi:10.1016/j.eneco.2023.106763. 35. Carbone JC, Rivers N. The impacts of unilateral climate
21. Sun H, Pofoura AK, Mensah IA, et al. The role of envi­ policy on competitiveness: evidence from computable
ronmental entrepreneurship for sustainable develop­ general equilibrium models. Rev Environ Econ Policy.
ment: evidence from 35 countries in Sub-Saharan 2017;11(1):24–42. doi:10.1093/reep/rew025.
Africa. Sci Total Environ. 2020;741:140132. doi:10. 36. Zhang N, Zhang W. Can sustainable operations achieve
1016/j.scitotenv.2020.140132. economic benefits and energy savings for manufactur­
22. Ren X, Shao Q, Zhong R. Nexus between green ing industries in China? Ann Oper Res. 2020;290(1-2):
finance, non-fossil energy use, and carbon intensity: 145–168. doi:10.1007/s10479-018-2955-3.
empirical evidence from China based on a vector 37. Feng Q, Wu Z, Zhang W. Carbon emissions market
error correction model. J Clean Prod. 2020;277: adjustment in the electricity supply sector: A govern­
122844. doi:10.1016/j.jclepro.2020.122844. ment perspective [J]. J Clean Prod. 2020;275:123–132.
23. Aftab J, Abid N, Cucari N, et al. Green human 38. Figueiredo NC. Firm-specific impacts of CO2 prices
resource management and environmental perform­ on the stock market value of the Spanish power
ance: the role of green innovation and environmental industry. Energy Policy. 2016;94:492–501. doi:10.
strategy in a developing country. Bus Strat Env. 2023; 1016/j.enpol.2016.01.005.
32(4):1782–1798. doi:10.1002/bse.3219. 39. Cohen MA, Adeline T. The impact of environmental
24. Jiang Y, Guo Y, Bashir MF, et al. Do renewable regulation on firm and country competitiveness: a
energy, environmental regulations and green innov­ meta-analysis of the Porter hypothesis. J Assoc Environ
ation matter for China’s zero carbon transition: evi­ Res Econ. 2018;5(2):371–399. doi:10.1086/695613.
dence from green total factor productivity. J Environ 40. Liu H, Zhou HY, Hou CR. Local government debt and
Manage. 2024;352:120030. doi:10.1016/j.jenvman. firm Innovation. Account Res. 2020;(09):163–177.
2024.120030. 41. Jie MH, Wang J, Liu DM. Environmental regulation,
25. He YN. The Impact and Mechanism of Carbon technological innovation, and firm performance.
Trading Market on Enterprises’ Innovation Strategies. Nankai Business Rev. 2014;17(06):106–113.
China Popul Resour Environ. 2022;32(7):37–48. 42. Yu W, Chen Q, Chen H. Environmental regulation,
26. Tian JQ, Han XY. Carbon emission trading and firm technological innovation, and operational perform­
financing constraints: evidence from China’s carbon ance: an empirical analysis based on 37 industrial
trading pilots. Friend Account. 2023;(06):104–111. sectors. Res Manage. 2017;38(02):18–25.
27. Wang X, Liu C, Wen Z, et al. Identifying and analyzing 43. Xie X, Wang M, Jiage R. Green technological innov­
the regional heterogeneity in green innovation effect ation and firm performance under government fiscal
from China’s pilot carbon emissions trading scheme incentives: an empirical study based on content ana­
through a quasi-natural experiment. Comput. Ind. lysis method. Manage Rev. 2020;32(05):109–124.
Eng. 2022;174:108757. doi:10.1016/j.cie.2022.108757. 44. Qi SZ, Lin S ,Cui JB. Can environmental equity trading
28. Liu J-Y, Woodward RT, Zhang Y-J. Has carbon emis­ markets trigger green innovation? Evidence from
sions trading reduced PM2. 5 in China?. Environ Sci green patent data of listed companies in China. Econ
Technol. 2021;55(10):6631–6643. doi:10.1021/acs.est. Res. 2018;53(12):129–143.
1c00248. 45. Zhang GX, Fan MM, Ma RK, et al. Synergistic emission
29. Zhao X, Sun BW. The influence of Chinese environ­ reduction effects of carbon trading policies. China
mental regulation on corporation innovation and Popul Resources Environ. 2022;32(03):1–10.
competitiveness. J. Clean. Prod. 2016;112:1528–1536. 46. Zhao ZZ, Cheng Z, L€ u DS. Does national low-carbon
doi:10.1016/j.jclepro.2015.05.029. strategy improve total factor productivity of firms?
30. Dong F, Dai Y, Zhang S, et al. Can a carbon emission evidence from low-carbon city pilots: a quasi-natural
trading scheme generate the Porter effect?Evidence experiment. Industr Econ Res. 2021;06):101–115. doi:
from pilotareas in Chinay. Sci Total Environ. 2019;653: 10.13269/j.cnki.ier.2021.06.008.
565–577. doi:10.1016/j.scitotenv.2018.10.395. 47. Jacobson. Earnings losses of displaced workers. Am
31. Ding L, Wang Y, Wang K. Synergistic effects and influ­ Econ Rev. 1993;83(4):685–709.
encing mechanism of pollution reduction and carbon 48. Xu H, Itoh H. Density economies and transport geo­
emission reduction through carbon emission rights gra-phy: evidence from the container shipping indus­
trading. Adv Clim Change Res. 2023;19(06):786–798. try. J Urban e-Conom. 2018;105:121–132.
32. Chang Z, Haijing CAI, Meijuan LIU. Research on enter­ 49. Wang X, Cao F, Ye K. Mandatory corporate social respon­
prise financial effects of the carbon emission sibility (CSR) r-eporting and financial reporting quality:
 CARBON MANAGEMENT 17

evidence from a quasi-natural ex-periment. J Bus Ethics. Chinese manufacturing firms. Research Management.
2018;152(1):253–274. doi:10.1007/s10551-016-3296-2. 2016;37(06):103–110.
50. Wen ZL, Ye BJ. Mediation analysis: methods and model 56. Li WW, Li EL. Business, model innovation, R&D invest­
development. Adv Psychol. 2014;22(05):731–745. ment and growth performance of entrepreneurial
51. Li GM, Zhang WJ. Research on industrial carbon emis­ firms. Finan Econ Monthly. 2021;(04):34–43.
sions and emission reduction mechanism under chi­ 57. Heckman JJ, Ichimura H, Todd P. Matching as an
na’s carbon trading. China Popul Resour Environ. econometric evaluation estimator. Rev Econ Studies.
2017;27(10):141–148. 1998;65(2):261–294. doi:10.1111/1467-937X.00044.
52. Yang R, Peng AQ. Environmental regulation, techno­ 58. Leng CY, Liu JP. Policy effects, emission reduction
logical innovation, and the performance of heavily pol­ paths and implications for carbon peak of low-carbon
pilot programs: a PSM-DID test based on 275 cities in
luting firms. J East China Normal Univ. (Philosophy and
China. J Hubei Univer (Philosophy and Social
Social Science Edition). 2021;53(01):129–141. þ73–74.
Sciences Edition). 2022;49(06):159–168.
53. Luo Y, Li X, Qi X, et al. The impact of emission trading
59. Jin JC. The relationship between outward foreign dir­
schemes on firm competitiveness: evidence of the medi­
ect investment from china’s manufacturing sector
ating effects of firm behaviors from the guangdong ETS.
and host country economic growth: an empirical ana­
J Environ Manage. 2021;290:112633. doi:10.1016/j.jenv­
lysis based on countries along the belt and road ini­
man.2021.112633. tiative. Res Finan Econ Issues. 2021;(04):108–115. doi:
54. Pan CL, Tian H. Research on the impact of forward- 10.19654/j.cnki.cjwtyj.2021.04.011.
looking environmental strategies on firm green 60. Lin ZH, Zhao SY. The impact of carbon emission trading
innovation performance - the chain mediating role of policies on firm financial performance: evidence from a
green intellectual capital and absorptive capacity. multiple periods double difference model examination.
Finan Econ Series. 2016;(07):85–93. doi:10.13762/j. Certif Publ Account China. 2022;(10):62–69.
cnki.cjlc.2016.07.007. 61. Feng C, Shi B, Kang R. Does Environmental policy
55. Yi S, Xue QZ. Green supply chain management and reduce enterprise innovation?—evidence: from China.
green innovation an empirical study based on Sustainability. 2017;9(6):872. doi:10.3390/su9060872.