Content uploaded by Yang Zhiqing
Author content
All content in this area was uploaded by Yang Zhiqing on Jun 19, 2024
Content may be subject to copyright.
Content uploaded by Yang Zhiqing
Author content
All content in this area was uploaded by Yang Zhiqing on Jun 19, 2024
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=fcpa20
Journal of Comparative Policy Analysis: Research and
Practice
ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/fcpa20
How Does Environmental Regulation Affect
Corporate Green Innovation: A Comparative Study
between Voluntary and Mandatory Environmental
Regulations
Zhiqing Yang, Peiyao Liu & Lianfa Luo
To cite this article: Zhiqing Yang, Peiyao Liu & Lianfa Luo (2024) How Does Environmental
Regulation Affect Corporate Green Innovation: A Comparative Study between Voluntary and
Mandatory Environmental Regulations, Journal of Comparative Policy Analysis: Research and
Practice, 26:2, 130-158, DOI: 10.1080/13876988.2024.2328602
To link to this article: https://doi.org/10.1080/13876988.2024.2328602
Published online: 04 Apr 2024.
Submit your article to this journal
Article views: 119
View related articles
View Crossmark data
How Does Environmental Regulation Affect
Corporate Green Innovation: A
Comparative Study between Voluntary and
Mandatory Environmental Regulations
ZHIQING YANG *, PEIYAO LIU**, & LIANFA LUO
†
*College of Public Administration, Huazhong University of Science and Technology, Wuhan, Hubei, China,
**School of Economics and Management, Wuhan University, Wuhan, Hubei, China,
†
Institute of Quality
Development Strategy, Wuhan University, Wuhan, Hubei, China
(Received 26 September 2023; accepted 3 March 2024)
ABSTRACT Environmental governance in China is a hybrid of mandatory and voluntary regula-
tions in both implementation and enforcement. This paper aims to explore how voluntary environ-
mental regulations (VERs) affect corporate green innovation (CGI) and the role of mandatory
environmental regulations (MERs). The study uses panel data of 1,232 Chinese-listed manufactur-
ing rms from 2004 to 2016 and the main ndings are: rst, VERs have a signicant positive effect
on CGI. Second, no signicant effect has been found of MERs on rms’ CGI. Third, in areas where
mandatory environmental regulations are weak, voluntary environmental regulation can better
serve as a supplement to promote rms’ green innovation. The results are robust to a series of
sensitivity checks. This study provides evidence that voluntary environmental regulation is more
effective in encouraging enterprises to engage in green innovation activities, and it is an important
supplement to mandatory environmental regulation.
Keywords: environmental governance; voluntary regulations; mandatory regulations; green inno-
vation; China
Introduction
Promoting corporate green innovation is an invaluable path for environmental govern-
ance, as it can achieve environmental goals at low or no economic costs. Basically,
there are two kinds of environmental governance in practice: mandatory environmental
regulations (MERs) and voluntary environmental regulations (VERs). The classic
Zhiqing Yang is Associate Professor, College of Public Administration, Huazhong University of Science and
Technology (yangzhiqing@hust.edu.cn).
Peiyao Liu is PhD candidate, School of Economics and Management, Wuhan University (liupeiyao@whu.edu.
cn).
Lianfa Luo is Associate Professor, Institute of Quality Development Strategy, Wuhan University
(luolianfa@whu.edu.cn).
Correspondence Address: Lianfa Luo Associate Professor, Institute of Quality Development Strategy, Wuhan
University, Bayi Road of Wuchang District, Wuhan, Hubei 430072, China E-mail: luolianfa@whu.edu.cn
Journal of Comparative Policy Analysis, 2024
Vol. 26, No. 2, 130–158, https://doi.org/10.1080/13876988.2024.2328602
© 2024 The Editor, Journal of Comparative Policy Analysis: Research and Practice
Porter Hypothesis indicates that pressure from mandatory regulations will promote
corporate green innovation (Porter and van der Linde 1995). However, much recent
literature nds that voluntary regulation, such as voluntary environmental certication,
also signicantly promotes green innovation (Prakash and Potoski 2007; Li et al. 2020;
Ren et al. 2022). The two types of regulatory tools exist simultaneously, but few
studies compare the effect and interplay in promoting corporate green innovation.
Determining the potential different effects of MERs and VERs on green innovation
is one of the most important issues for environmental governance in developing
countries which face greater pressure in development and have a greater need for
green innovation. As a large transitional developing economy, China has undergone a
transformation from mandatory regulation to voluntary regulation, searching for ef-
cient governance under the goals of carbon peaking and neutrality. This research will
shed light on good environmental governance in developing countries in the process of
green production transformation.
Mandatory environmental regulations are often promulgated and implemented by laws
and government regulations with strong “command-and-control” features. In the 1990s,
China’s environmental governance was characterized by nes for pollution, environmen-
tal monitoring, and environmental inspection. The main purpose of environmental
regulations is to punish illegal environmental behavior and the policy instruments are
mostly mandatory. This kind of regulation has been lauded for reducing pollution yet also
criticized as too costly, inexible, and adversarial. Due to the large number of enterprises
under supervision, the regulator and polluter often fall into exhausted “guerrilla warfare”.
Prior studies (Jaffe et al. 2002; Wang 2016) noted that the cost of mandatory regulation
can even far exceed the direct cost of pollution and cannot stimulate rms’ green
production behavior.
Voluntary environmental regulations, also called environmental self-regulation (Anton
et al. 2004; Stoeckl 2010), voluntary agreements (Segerson and Miceli 1998; Delmas and
Montes-Sancho 2010), and voluntary environmental programs (Prakash and Potoski
2007), refer to a rm’s voluntary participation and engagement in environmental
enhancement, such as environmental information disclosure (EIC), environmental man-
agement system certication (EMSC), green production-related subsidies, and voluntary
environmental agreements. Compared to the end-focused treatment of mandatory regula-
tions, voluntary regulations have a pre-intervention target. For instance, the “green club”
theory built by Prakash and Potoski (2007) proposed that well-designed voluntary
environmental programs (or “green clubs”) can complement regulation by inducing
rms to go beyond prescribed standards. They test the “green club” theory that some
companies will join green clubs and gain additional environmental benets as well as a
positive reputation. This reputation can also translate into less pollution. The positive
effects of VER can also be interpreted through signaling theory. Many studies conrmed
that rms participating in voluntary environmental agreements in order to distinguish
themselves from those with poor environmental performance enhance their reputation in
the stock market (Bae et al. 2018). Due to its low cost and lower intervention, China’s
senior ofcials responsible for the environment have increasingly shifted to advocate
such voluntary environmental regulation. The latest important notice released by the
State Council in April 2022 proposed to develop “a unied national market”,
1
requiring
Environmental Protection Bureaus (EPBs) at all levels to rely more on voluntary
Environmental Regulation and Green Innovation 131
governance such as green product certication and green labeling systems to promote
green production and green consumption.
Although a substantial amount of literature recognizes the critical impact of both
mandatory and voluntary environmental governance on green innovation (Prakash and
Potoski 2007; Aragon-Correa et al. 2016; Ren et al. 2022), there have been few attempts
to evaluate the interplay of MERs and VERs with green innovation in the context of a
developing transitional economy.
In this paper, we used long-term rm-level data to assess the impacts of voluntary and
mandatory environmental regulations on corporate green innovation, and the interacting
effects between them. Specically, this study aimed to answer the following two ques-
tions. (1) How do mandatory and voluntary environmental regulations affect green
innovation activities in China? As a large developing economy, China is pursuing a
dual carbon reduction target strategy and green industrial transformation, and voluntary
regulation in China is still a new concept in environmental governance; therefore, the
impact on green innovation should be evaluated and investigated. (2) How do MERs and
VERs interplay to achieve good environmental governance in a developing country like
China? Although mandatory regulation has long been criticized for its high cost and low
efciency, we hold that the best governance may not abandon MERs completely but
coordinate with VERs. We try to nd out which areas or regions are more suited to
MERs, which to VERs, and which need a combination of the two.
Specically, this study lls the research gap in the following three respects:
First, we comparatively analyzed voluntary and mandatory environmental regula-
tions on corporate green innovation in theory and practice. Previous studies have
examined the effect of VERs and MERs respectively on CGI; however, a few focused
on the comparison of voluntary and mandatory regulations in terms of promoting green
innovation. Some studies pointed out that mandatory regulations are not always
effective because of the high cost of compliance, political connections, and enforce-
ment lapses (e.g. Stigler 1975; Marcus 1980). Whereas extant literature has also been
skeptical on the green effect of voluntary regulations because of adverse selection,
moral hazard, and free-riding (e.g. King and Lenox 2000; Potoski and Prakash 2013;
Berliner and Prakash 2015; Testa et al. 2018), this claim has also been supported by
empirical evidence. For example, Aragon-Correa et al. (2016) found that rms with a
better record of voluntary environmental information disclosure actually show worse
environmental progress than others. However, we extend those studies by comparing
these two kinds of regulations in the context of a large developing country seeking to
deliver the Paris climate objectives. Specically, we discuss the features of mandatory
and voluntary environmental regulations in China in theory and practice, and try to
answer the question of why mandatory environmental regulations do not work on
average, and why VERs are more effective in practice.
Second, we focused on the interplay of MERs and VERs. Although VER is still
a new practice in developing countries, literature has advocated implementing it
due to the low cost and high compliance in practice. As a highly centralized
political system, China has long adopted mandatory environmental regulations
mainly based on “command-and-control” government regulation (Wang et al.
2015). Though China is switching to more market-oriented MERs such as environ-
mental taxes or nancial subsidies, studies have also addressed the fact that the
132 Z. Yang et al.
cost of mandatory regulations can far exceed the direct cost of pollution and cannot
stimulate rms’ green production behavior (Jaffe et al. 2002; Wang 2016). In this
paper, we extend the literature by investigating further the interplay of MERs and
VERs. In a large transitional economy like China, we do not believe that manda-
tory regulatory measures are completely ineffective and voluntary regulations are
effective in all areas. With the coexistence of MERs and VERs, we try to explore
the effectiveness of MERs. We test the heterogenous effect of VERs on green
innovation under different MER strength levels, so as to nd out how VERs
substitute or complement the role of MERs in promoting rms’ green innovation.
The results will provide timely policy implications for the reform of MERs in
developing countries.
Third, most of the previous literature focused on developed economies, whereas
few studies have focused on voluntary environmental governance in developing
countries. Many studies conrmed that the institutional and economic development
context can shape how rms operate and perform (Peng et al. 2008; Williams and
Martinez 2012), and affect the implementation and effectiveness of voluntary
environmental regulations (Iatridis and Kesidou 2018). The evidence from devel-
oped economies may not be applicable to developing countries, where the institu-
tional and socioeconomic context is dissimilar and the roles of voluntary and
mandatory regulations in those countries are often very different. The green impact
of voluntary and mandatory environmental governance in China is very important
for the ongoing “Dual Carbon Goals” as well as the world’s carbon reduction
goals.
Besides, we enhanced the technical method in policy evaluation to obtain more precise
and robust results. Specically, we rened the identication strategy in the following two
ways: rst, we specied different degrees of mandatory environmental regulations to
understand the synergetic effect on rms’ green innovation. We rened the measurement
of mandatory regulation by dividing mandatory regulations into punitive environmental
administrative penalties and relatively exible supervision tools such as the Pollution
Information Transparency Index (PITI) to measure different levels of government-man-
dated regulations. The two types of mandatory regulation tools represent hard and soft
governmental constraints respectively, and it can help to look into their interactions in
depth. Second, we use the difference-in-difference (DID) method to gain more robust
conclusions. Li et al. (2020), Jiang et al. (2021), and Zhu et al. (2021) have researched a
topic similar to this study. The main empirical method used in their papers are xed
effect models. The effects cannot be identied well using such a model in a case where
there are endogenous problems. This paper for the rst time uses the DID model to test
the effects of VERs and MERs on green innovation. Besides, we conduct series robust-
ness checks, such as PSM-DID (Propensity-Score-Matching and Difference-in-
Difference) or Imputation DID, to make the research more robust.
The remainder of the paper proceeds as follows. The second section presents a review
of MERs and VERs in China. The third section proposes the hypothesis of this study.
The fourth section describes the empirical study of how MERs and VERs affect
corporate green innovation. The fth section discusses further regressions with compara-
tive analysis between MERs and VERs on green innovation in different industries. The
last section concludes and presents the policy implications of the study.
Environmental Regulation and Green Innovation 133
Background on China’s Mandatory and Voluntary Environmental Regulations
China’s Mandatory Environmental Regulations
As a large transitional developing economy, China has been increasingly aware of the
importance of environmental protection and has built sophisticated mandatory environ-
mental governance systems including a series of environmental laws and regulations
related to water, air, noise, soil, solid waste, etc.
2
China has long been adopting mandatory environmental regulations mainly based on
government regulation (Wang et al. 2015), which is different from the market-oriented
environmental regulations such as environmental taxes or nancial subsidies implemen-
ted by developed countries. Specically, China has promulgated and implemented a
series of relevant laws and regulations to strengthen environmental supervision and
bring the attention of enterprises to environmental protection, such as the
Administrative Measures for Listing and Supervising the Handling of Environmental
Violations (2009), the Environmental Protection Law (2015), the “River Chiefs” system
(2016), the Environmental Protection Tax Law (2018), and the Cleaner Production
Promotion Law of the People’s Republic of China (2019). Local ofcials are the policy-
makers and executors of these mandatory environmental regulations. Therefore, in order
to stimulate local ofcials to implement relevant environmental regulations, since the
11th Five-Year Plan, the central government has established emissions reduction targets
of major pollutants as binding indicators and linked them with the political achievements
of local government ofcials.
In terms of specic administrative means, environmental administrative punishment, a
policy tool to punish illegal environmental behaviors, is an important and widely used
strategy in China’s environmental supervision, using the deterrence of punishment .
However, this punitive method has limited effect and is criticized as too costly, inexible,
and adversarial. Due to the large number of enterprises under supervision, the regulator
and polluter often falls into exhausted “guerrilla warfare”, so that some enterprises may
evade supervision thus increasing the law enforcement cost of supervision, and some
enterprises’ substantive environmental management behavior may change to symbolic
environmental management behavior, which is not conducive to the rms’ long-term
green production behavior (Short and Toffel 2010; He et al. 2018; Iatridis and Kesidou
2018).
Being aware of the limitations of this kind of treatment and rigid supervision, the
Chinese government tried to make changes on environment governance. For example,
government environmental information disclosure has become another relatively soft
environmental regulation tool to improve the efciency of environmental governance.
The former State Environmental Protection Administration issued the Environmental
Information Disclosure Measures (Trial) in 2007, requiring the government and enter-
prises to disclose environmental quality information and environmental supervision
information of management departments to the public, trying to alleviate the information
asymmetry between the central government, local governments, enterprises, and the
public through environmental information disclosure, and improve the efciency of
environmental governance. Since then, the importance of environmental information
disclosure has been emphasized more frequently in the documents issued by the State
Council.
3
Although the disclosure of government environmental information will not act
134 Z. Yang et al.
as a mandatory constraint on the polluting behaviors of enterprises, it can act as an
external soft constraint on enterprises through the disclosure of pollution information and
environmental governance information. In conditions of given environmental regulation
intensity, a higher degree of government disclosure of environmental information often
resulted in higher consciousness of environmental behavior (Kumar and Shetty 2018).
Furthermore, enterprises may seek to change the original production mode through green
technology innovation to reduce the possibility of environmental default. The Institute of
Public and Environmental Affairs (IPE) and Natural Resources Defense Council (NRDC)
jointly developed the “Pollution Information Transparency Index”. Since 2008, it has
systematically evaluated the quality of environmental information disclosure of 120
national key cities for consecutive years and released the evaluation results to the public.
This index quantitatively assesses the degree of environmental information transparency
of these cities and is also conducive to horizontal and vertical comparison.
The Development of Voluntary Environmental Regulations
Due to the complexity of environmental problems and asymmetric information, there are
often policy distortions in the implementation of mandatory environmental regulations,
such as rent seeking and pollution transfer within enterprise groups, which cannot
achieve the expected policy effects and economic dividends (Chen et al. 2021). Some
studies found that the cost of mandatory regulation even far exceeded the direct cost of
pollution and cannot stimulate rms’ green production behavior (Jaffe et al. 2002; Wang
2016). As a supplementary tool for mandatory environmental regulation, the number of
voluntary environmental regulation projects has increased signicantly worldwide in the
past few decades (Testa et al. 2018; York et al. 2018). Blackman (2008) pointed out that
voluntary environmental regulation was constantly evolving under the negative stimulus
of mandatory regulations and policies. Unlike mandatory environmental regulation,
voluntary environmental regulation is based on the voluntary participation of enterprises,
advocated by industry associations, enterprises themselves, or third-party certication
bodies without specic mandatory requirements (Aragon-Correa et al. 2020). And
voluntary environmental regulation focuses on setting goals, strategies, and development
guidelines to improve enterprises’ environmental performance, which provides compa-
nies with more exibility (Arora and Cason 1995). Its core idea is to create incentives for
enterprises to spontaneously provide environmental public goods (Pan et al. 2020).
Therefore, the voluntary environmental regulation has received increasing attention,
and is widely used by companies around the world (Bu et al. 2020).
Compared to mandatory environmental regulations, voluntary environmental regula-
tions are less widely used in China. At present, the most popular voluntary environmental
regulation in China is the ISO14001 environmental management system certication (Bu
et al. 2020) and the “Responsible Care” system of the chemical industry (Pan et al.
2020). Besides, rms’ environmental information disclosure is a voluntary and active
behavior (Jiang et al. 2020). In 1996, the International Organization for Standardization
issued and implemented the ISO14000 series of environmental management standards, of
which only ISO14001 can be applied for certication. ISO14001 has formulated a set of
principles for the internal environmental management of enterprises. Enterprises that
have obtained ISO14001 certication need to formulate written environmental
Environmental Regulation and Green Innovation 135
management regulations and internal management systems, formulate quantitative envir-
onmental objectives, conduct periodic self-verication, and appoint senior managers to
supervise the environmental management system and objectives. In addition, ISO14001
environmental management system certication also requires third-party auditors to
conduct periodic checks of enterprises to ensure that they meet project management
requirements (Lim and Prakash 2014). At present, ISO14001 standards have been revised
three times: ISO14001:1996, ISO14001:2004, and ISO14001:2015. Since 1996, the total
number of organizations that have achieved ISO14001 environmental management
system certication in China has been increasing (Pan et al. 2020). According to the
latest ISO survey data, the number of ISO14001 certicates in China reached 295,501 at
the end of 2022, accounting for 55.87% of the world’s total.
4
As for the “Responsible Care” system, the Ministry of Industry and Information
Technology of the PRC issued Guidelines for the Implementation of Responsible Care
(HG/T4184-2011) as an industry standard on June 15, 2011. Since then, “Responsible
Care” has entered a stage of concrete action and comprehensive promotion in China.
However, this system has not been well applied and popularized in China. By the end of
2020, only 636 chemical companies in China had signed the Responsible Care Global
Charter. In addition, most of the enterprises that joined the Charter of Care for
Responsibilities do not fully follow the normative requirements of the Guidelines for
the Implementation of Responsible Care to formulate target plans, implement self-
assessments, and enforce improvement measures (Pan et al. 2020).
It is worth noting that Chinese ofcials are increasingly advocating such voluntary
environmental regulation due to its low cost and higher efciency. The latest important
document released by the State Council in April 2022 proposed to develop “a unied
national market”.
5
The environmental governance section of the document proposed
introducing green product certication and labeling systems to promote green production
and green consumption. Table 1 is a summary of China’s current environmental regula-
tion system.
Literature Review and Hypotheses Development
Direct Effect of Voluntary Environment Regulation on Corporate Green Innovation
The effectiveness and success of voluntary environmental regulation on corporate green
innovation mainly rely on the rms’ motivations to participate in a specic voluntary
scheme. Basically, VERs affect CGI mainly through the following two channels:
The Internal Pursuit of Prot. A common nding is that rms may choose to adopt
voluntary environmental tools or join in voluntary programs only if their private benets
outweigh their costs at present or in the foreseeable future (Stoeckl 2010). Most notably,
Prakash and Potoski (2007) constructed and tested a “green club” theory that rms join
green clubs and gain additional environmental benets as well as a positive reputation.
Many studies have found knowledge sharing among “green clubs”. For example, Zhu et
al. (2021) used data from China’s iron and steel industry and found that the VERs eased
knowledge sharing barriers by improving human capital, while MERs cannot. Besides,
this reputation can also translate into rms’ environmental progress, which resulted in
less pollution. They also pointed out that well-designed voluntary environmental
136 Z. Yang et al.
regulations (or “green clubs”) can even induce rms go beyond prescribed environmental
standards. Besides, much accounting and nance literature emphasizes the role of
managers’ perceptions and preferences in green innovation efforts and participation in
voluntary environmental programs. A strong consensus exists that managers’ negative
attitudes towards environmental regulations are a great hindrance to rms’ environmental
progress and green efforts (Rivera Ungson et al. 1985; Cordano et al. 2004). Many recent
studies highlighted that voluntary environmental regulations with more exible tools and
lower standard levels will encourage more corporate green innovation. For example, Sun
and Abraham (2021) used China’s high-tech companies to nd that exible environ-
mental regulation tools and a combination of rigid and exible tools are more conducive
for rms to carry out innovative activities. Wang et al. (2021) theoretically analyzed and
conrmed that the environmental regulations with stricter standards for a given voluntary
adoption level may discourage a rm from developing a new green technology in the rst
place when facing intense competition.
The External Good Public Image. Firms that want to publicly show their environmental
performance progress and are more visible to consumers and regulators are more likely
to join voluntary environmental programs. Many studies have pointed out that rms with
larger scale, greater social responsibility, and wider market scope are more likely to
engage in voluntary environmental programs (Anton et al. 2004). Prior studies also found
that enterprises of different types (listed enterprises, private enterprises, and state-owned
Table 1. China’s environmental regulation structure
Mandatory environmental regulations Major tools in practice
(i) Ecological and environmental law
enforcement
Pollution discharge fee; total emission control
(TEC); notication of illegal and criminal
environmental cases
(ii) Environmental inspection Real-time monitoring and publicity of air and
water quality index
(iii) Environmental administrative punishment Fines, orders to rectify or suspend production,
etc.
(iv) Governmental environment information
disclosure
Such as “Measures for environmental
information disclosure (Trial)”, issued by
MEE in 2008
Voluntary environmental regulations
(i) Environmental agreements or programs
(administered by regulators or third parties)
that individual rms are invited to join
Environmental management certication (e.g.
ISO14001); ecological labeling
(ii) Public disclosure initiatives that collect and
disseminate information on participants’
environmental performance
Environmental information disclosure
(iii) Other unilateral commitments made by
rms
Source: Ministry of Ecology and Environment of China (MEE), https://www.mee.gov.cn, http://
english.mee.gov.cn/Resources/laws/; https://www.mee.gov.cn/gkml/zj/jl/200910/t20091022_
171845.htm.
Environmental Regulation and Green Innovation 137
enterprises) all have a common external motivation, which is regulatory pressure (King
and Lenox 2000). Videras and Alberini (2000) attributed all these factors to rms’
pursuit of good publicity, holding that publicity is a key aspect emphasized by rms in
terms of participation in voluntary environmental programs. They pointed out that rms
with larger scale and wider market scope are often more visible to consumers, which can
be interpreted as publicity. Firms participating in voluntary environmental programs or
agreements can distinguish themselves from those with poor environmental perfor-
mances to gain a better reputation (Bae et al. 2018). Although many studies suggested
that the publicity and reputation can nally translate into prot, there are some rms that
attach great importance to their image.
Therefore, we proposed the following hypothesis:
Hypothesis 1 (H1). Voluntary environmental regulation has signicant positive
effects on corporate green innovation behaviors.
Effect of Mandatory Governance and the Interactions on Commitment of Voluntary
Environmental Regulations
Mandatory environmental governance such as environmental surveillance, administrative
punishment, and pollution discharge fees can be effective regulatory tools. Government
holds rms accountable for their environmental impact and rms must abide by the
multiple policies and regulations to gain legitimacy (Berrone et al. 2013), so this kind of
environmental regulation is inevitable for rms’ production and operation (Jiang et al.
2021).
In terms of the association between mandatory environmental regulation and innova-
tion, it is traditionally believed that mandatory environmental regulations are a constraint
on business, impeding innovative activities (Darnall and Carmin 2005; Zhang et al.
2018). In addition, rms have little room for choice and are forced to follow the
regulations mechanically or they will face severe penalties, which may damage their
efciency and reduce their enthusiasm for green innovation (Popp et al. 2011; Zizzo and
Fleming 2011). However, Porter and van der Linde (1995) proposed that mandatory
regulations do not necessarily hinder innovation; on the contrary, properly designed
regulations could foster corporate innovation activities and ultimately improve competi-
tive advantage. Based on the Porter Hypothesis, much research reexamines the relation-
ship between mandatory environmental regulations and innovation from multiple
perspectives, but the conclusions were not consistent. Taking into account the actual
context of China, this study will discuss the relationship between mandatory environ-
mental regulations and corporate green innovation in China. Specically, MERs may
affect CGI in the following three respects.
Firstly, the Chinese political system may affect the efciency of MERs, which may
hinder green innovation. China’s political system is largely centralized within a multi-
divisional hierarchical structure. The central government mainly sets the national eco-
nomic environmental policy and the local governments have some governance discretion
on how to adopt the policy. However, the local governments focus more on economic
development rather than environmental protection due to nancial pressures (Pang et al.
138 Z. Yang et al.
2019). Furthermore, as promotion in the Chinese political system is mainly based on
local economic performance in the long term, local ofcials are intended to promote
economic growth for their careers and ignore environmental performance to some extent
(Chang and Wu 2014; Cull et al. 2017). Therefore, the effectiveness and actual imple-
mentation of MERs may be hindered in practice by conicting interests, which may
impede green innovation.
Secondly, MERs will have a crowding out effect on enterprises’ green innovation
investment activities. Schumpeter’s innovation theory holds that the availability of funds
plays an important role in technology innovation (Drejer 2004). Compared to conven-
tional technological innovation, green technological innovation necessitates substantial
initial capital investment, entails a lengthy prot cycle, involves unpredictable risks, and
requires additional capital investment (Johnstone et al. 2010). The pressure from MERs
means that enterprises struggle to cope with the high nes and administrative penalties
caused by polluting emissions, which squeeze the cost of research and development
innovation. Especially the penalty mechanism of MERs can displace R&D expenditures
and discourage companies from investing in green innovative technologies, thereby
hindering green innovation (Petitjean 2019; Zhang et al. 2020; Wang et al. 2023).
Thirdly, MERs induce strategic behavior in enterprises, which may hinder green
innovation. MERs have not always been fully effective because of the high cost of
compliance and low exibility, which may serve as incentives for enterprises to evade
regulation and limit environmental adaption to end-treatment or emissions control, rather
than working towards source control such as green innovation (Qin and Sun 2020). Some
research pointed out that with the strengthening of MERs, some enterprises’ substantive
environmental management behavior may change to symbolic environmental manage-
ment behavior (Short and Toffel 2010; He et al. 2018; Iatridis and Kesidou 2018). For
example, enterprises choose to “greenwash” in order to deal with the pressure of
legitimacy in the short term (Berliner and Prakash 2015). However, the symbolic
environmental management behavior goes against the enterprises’ long-term green inno-
vation behavior.
In terms of the above review, we proposed the following hypothesis:
Hypothesis 2a (H2a). Mandatory environmental regulations do not have a positive
impact on CGI.
By contrast, more evidence has supported the feasibility of voluntary governance in
developing countries, VERs are considered a useful supplement to MERs (Khanna
and Damon 1999; Alberini and Segerson 2002). VERs allow rms greater latitude
across programs in the market for environmental protection (Blackman et al. 2010;
Picard 2015). MERs impose additional direct costs of controlling pollution and
indirect costs caused by the rising prices of certain factors of regulated rms, while
VERs can encourage enterprises to voluntarily reduce environmentally negative
impacts beyond what MERs require (Carmin et al. 2003; Rennings et al. 2006).
VERs can partially or fully offset the cost of complying with MERs for regulated
enterprises, while mitigating the regulatory burden for governments and addressing
sustainable development (Rennings et al. 2006). Therefore, VERs are a supplement
Environmental Regulation and Green Innovation 139
of MERs, which can help rms to improve their green innovation ability. We
proposed the following hypothesis:
Hypothesis 2b (H2b). The voluntary environmental regulation serves as an impor-
tant supplement of mandatory environmental regulations.
The research framework of this study is shown in Figure 1.
Research Design
Sample and Data Collection
Manufacturing rms are the most important participants in environmental programs and the
main polluters in China. Therefore, to investigate the relationship between environmental
regulations and rms’ green innovation, we construct our sample based on Chinese listed
rms in the manufacturing industry from 2004 to 2016. The reason why we used data from
2004 to 2016 is that we matched the China Customs Database (CCD) to get the information
about rms’ exports. A lot of literature has conrmed that exporting has a signicantly
positive effect on innovation (Grossman and Helpman 1990; Wu et al. 2022), and exports
may inuence the adoption of ISO14001 certication in the enterprises (Tambunlerchai et al.
2012). Therefore, to avoid the endogeneity problem caused by omitted variable bias, we
control the export situation in the model. CCD only provides relevant export information for
enterprises up to 2016 so that we have to limit our data period to 2016.
Our data are obtained from the following sources:
First, the environment regulations data. For the voluntary environmental regulation,
we use ISO14001 certication data as an example which was manually collected from
the listed rms’ certicate information in the iFind Database (https://www.51ind.com).
Specically, ISO14001 certication is valid for three years, and rms need to be audited
once a year during the three years, or the ISO14001 certication will be considered
invalid. Therefore, based on the validity period of the ISO14001 certication obtained by
each enterprise, we construct a dummy variable to indicate the rm’s certication status.
Figure 1. Research framework
140 Z. Yang et al.
For the mandatory environmental regulation, the data of local environmental admin-
istrative punishment cases are obtained from PKULAW (https://m.pkulaw.com/). The
PITI data are obtained from the Institute of Public and Environmental Affairs and
Natural Resources Defense Council. They jointly released the “Pollution Information
Transparency Index” which evaluated the quality of government environmental informa-
tion disclosure in 117 key cities since 2008.
Second, the green innovation data are obtained from the China Research Data Service
Platform (CNRDS), which provides listed rms’ green patent information, including the
authorizations of green invention patents, green utility patents, and total patents.
Third, the nancial data of listed rms are obtained from the China Stock Market and
Accounting Research Database (CSMAR), and the export data are obtained from the
China Customs Database (CCD).
The data processing procedures are as follows: rst, we matched the above data
according to the rms’ code, the year, and the city where the rm was located. Then
we dropped the rms that received a delisting warning status because of their abnormal
nancial or other operational conditions and the rms with missing core information or
which are obvious outliers (i.e. their continuous nancial variables are above the 99th
percentile or below the 1st percentile). Consequently, the nal sample included 6,860
observations from 1,232 manufacturing listed rms.
Variables and Measurements
VERs. There are various forms of voluntary environmental regulations. Among them,
ISO14001 has the highest degree of popularity and recognition (Bu et al. 2020; Jiang et
al. 2021). Therefore, this study focused on rms’ certication of ISO14001, and we
construct a dummy variable as a measurement of VER. If a rm was certied by
ISO14001 and was within its validity period in a given year, it is coded as 1; other
rms are coded as 0.
MERs. As we mentioned earlier regarding the background of China’s mandatory envir-
onmental regulations, there is already a series of environmental regulation tools. China
has long adopted mandatory environmental regulation methods based on government
regulation (Wang et al. 2015). Therefore, this study selects two representative environ-
mental regulations and constructs a comprehensive indicator to consider the strength of
government mandatory environmental regulations in the city where the rm is located.
One tool is environmental administrative punishment, which is a severely punitive
environmental regulation tool. The other tool is the government’s environmental infor-
mation disclosure, which is a relatively exible and soft environmental regulation tool
aimed at improving regional environmental supervision. In this study, we assumed that
the administrative punishment (Punishment) and information disclosure (Supervision)
played equal roles. According to the Cobb–Douglas production function, the expression
of MER is:
Environmental Regulation and Green Innovation 141
The resource endowments and development modes differ in different regions, so the
intensity of government mandatory environmental regulations also varies among
regions. In this study, environmental administrative punishment was measured by the
natural logarithm of the city’s local environmental administrative punishment cases.
The environmental supervision was measured by the Pollution Information
Transparency Index. PITI reects the transparency of pollution information and the
implementation of environmental policies in cities published since 2008. In the
environmental policy research area, PITI has become an important tool (Ye et al.
2015). Therefore, this indicator comprehensively considers the government’s manda-
tory environmental regulation from the perspective of both rigid and soft regulatory
tools.
CGI. Green innovation is a key factor in coordinating economic growth and environ-
mental protection (Magat 1978). Green innovation also helps enterprises comprehen-
sively reduce environmental costs, gain the trust of unique suppliers and customers, and
thus seize a green competitive advantage (Hart 1995). Therefore, this paper mainly
examines the impact of different environmental regulatory tools on rms’ green innova-
tion. Compared to the number of patent applications, the number of patent authorizations
better reects the quality of innovation. Therefore, this study uses the number of green
patent authorizations to measure the rms’ green innovation performance. Specically,
the variables considered in the article include the total number of green patent author-
izations, the number of green invention patent authorizations, and the number of green
utility model patent authorizations.
Detailed variable denitions are presented in Table 2 and descriptive statistics are
presented in Table 3.
Model Specication
The Impact of VER on CGI. We estimate the impact of the voluntary environmental
regulations on corporate green innovation using staggered difference-in-difference meth-
odology. The basic model is specied as:
Where i and t indicate rm i and year t; CGIit represents the green innovation
performance for rm i in year t including the total number of green patent author-
izations, the number of green invention patent authorizations, and the number of
green utility patent authorizations. VERit represents the voluntary environmental
regulation for rm i in year t. The rms that have a valid ISO14001 certication
in a given year are coded as 1, other rms are coded as 0. The parameter β is the
coefcient of our interest, indicating the effect of VER on rm’s green innovation
performance.
Xit contains a set of control variables including rm size (Size), rm’s net assets
returns ratio (Roe), rm’s asset liability ratio (Lev), rm’s operating revenue growth ratio
(Growth), rm’s number of directors (Board), rm’s number of independent directors
142 Z. Yang et al.
Table 2. Denition of variables and measurement indicators
Variable Denition Measurement
Grepatent Number of total green patent
authorizations
Inverse hyperbolic sine (IHS) transformation of the
number
Greinvent Number of green invention
patent authorizations
Same as above
Greutility Number of green utility
patent authorizations
Same as above
VER Voluntary environmental
regulation
Whether the rm have a valid ISO14001 certication in
the observation year, Yes = 1, No = 0.
MER Mandatory environmental
regulation
The natural logarithm of MER, see the measurement
method for details.
Size Firm size The natural logarithm of total assets
Roe Firm’s net assets returns ratio Net prot to net assets at the end of year
Lev Firm’s asset liability ratio Debt to asset ratio
Growth Firm’s operating revenue
growth ratio
Increase in operating revenue to total operating revenue
of the previous year
Board Firm’s number of directors The natural logarithm of the number of directors
Indep Firm’s number of
independent directors
The natural logarithm of the number of independent
directors
Soe State-owned enterprise Yes = 1; no = 0
Top1 Shareholding ratio of the
largest shareholder
Number of shares held by the largest shareholder to
total shares
Inst Shareholding ratio of the
institutional shareholders
Number of shares held by institutional shareholders to
total shares
Export Firm’s export volume The natural logarithm of the export volume
Table 3. Descriptive statistics for variables
Variable Obs Mean Sd Min Max
Grepatent 6,860 0.4261 0.8668 0.0000 6.2344
Greinvent 6,860 0.1693 0.5206 0.0000 5.2575
Greutility 6,860 0.3299 0.7699 0.0000 5.9610
VER 6,860 0.3628 0.4809 0.0000 1.0000
MER 1,715 4.0528 1.0427 1.8863 6.0461
Size 6,860 21.6191 1.0490 19.2244 27.3031
Roe 6,860 0.0320 0.8741 −46.5167 0.8976
Lev 6,860 0.4020 0.1945 0.0291 0.9925
Growth 6,860 0.1803 0.4129 −0.7234 7.9565
Board 6,860 2.1588 0.1893 1.6094 2.8332
Indep 6,860 0.3672 0.0503 0.1818 0.5714
Soe 6,860 0.3700 0.4828 0.0000 1.0000
Top1 6,860 0.3612 0.1428 0.0826 0.7546
Inst 6,860 0.3188 0.2411 0.0000 0.8919
Export 6,860 15.8254 2.5546 1.7918 21.6350
Environmental Regulation and Green Innovation 143
(Indep), whether the rm is state-owned (Soe), shareholding ratio of the largest share-
holder (Top1), shareholding ratio of the institutional shareholders (Inst), and rm’s export
volume (Export). The measurement and descriptive statistics are shown in Table 2 and 3.
μt is the year xed effect, controlling other effects in a certain year that might
inuence all rms in a similar manner. σi is the rm-level xed effect representing all
the time-invariant difference across rms which might affect green innovation. εit is the
error term.
The Impact of MER on CGI. We estimate the impact of the mandatory environmental
regulations on corporate green innovation using a two-way xed effects (TWFE) model.
The basic model has the following specication:
The key variable of interest is MERit which represents the intensity of mandatory
environmental regulation in the city where the rm is located. The coefcient δ provides
the effect of our interest. Other variables are the same as in Equation (2).
Empirical Results
The Direct Effect of VER on CGI. Table 4 reports the main results of the DID estimation
using CGI as the dependent variable. Columns (1) to (6) all control for both rm and year
xed effects. Columns (1) and (2) report the “unconditional” and “conditional” effects of
VER on the number of total green patent authorizations. Column (2) adds the vector of
covariates Xit to control for rm characteristics, which shows that participating in VER
(obtaining ISO14001 certication) is associated with a 13 percentage point increase in
the total number of green patent authorizations. In columns (3) and (4), the dependent
variable is the number of green invention patent authorizations. After controlling for the
rm characteristics, participating in VER is associated with a 6.8 percentage point
increase in the number of green invention patent authorizations. In columns (5) and
(6), the dependent variable is the number of green utility patent authorizations, and the
estimation results in column (6) show that participating in VER is associated with a 9.7
percentage point increase in the number of green utility patent authorizations. The results
in Table 4 all present a signicantly positive relationship between VER and CGI.
Parallel Trend Test. The common trend assumption is the key assumption underpinning
the use of a DID regression to estimate the treatment effect. This assumption requires
that changes in the treated group and in the controlled group are the same in the absence
of certain policy shocks. Before the attainment of ISO14001, the treated and control
groups should experience consistent or parallel changes, indicating that the results are
driven by obtaining the ISO14001 certication and not by an underlying trend between
the two groups. Following Beck et al. (2010), the parallel trend test equation is as
follows:
144 Z. Yang et al.
Where VERi;toþk is a series of dummies jointly denoting a window of some periods
around obtaining ISO14001 certication. Specically, t
0
represents the rst year in which
rm i is certied, and k represents the k
th
year after it is certied, k = −4, −3, −2, −1, 0, 1,
2, 3, 4 + . Other variables are of the same denition as in Equation (2). We take t = −1 as
the base group to conduct the parallel trend tests. βk is the parameter of our interest, and
the results are illustrated in Figure 2. We can nd that before obtaining ISO14001
certication, there was no systematic difference between certied and uncertied rms,
but after obtaining ISO14001 certication, the coefcients began to rise signicantly,
indicating that the difference in green innovation outputs between certied and uncerti-
ed rms come from the ISO14001 certication.
Table 4. Effect of VER on CGI
(1) (2) (3) (4) (5) (6)
Grepatent Grepatent Greinvent Greinvent Greutility Greutility
VER 0.137*** 0.130*** 0.072*** 0.068** 0.104*** 0.097***
(4.07) (3.90) (2.69) (2.55) (3.35) (3.15)
Size 0.131*** 0.067** 0.123***
(3.26) (2.47) (3.45)
Roe 0.000 0.006* −0.004
(0.04) (1.87) (−0.49)
Lev 0.064 0.104 0.007
(0.60) (1.59) (0.07)
Growth −0.058*** −0.033*** −0.052***
(−3.27) (−2.58) (−3.12)
Board 0.041 −0.040 0.095
(0.35) (−0.57) (0.84)
Indep 0.064 0.087 0.055
(0.19) (0.41) (0.19)
Soe −0.038 −0.079 0.022
(−0.57) (−1.33) (0.40)
Top1 −0.269 −0.089 −0.167
(−1.21) (−0.62) (−0.80)
Inst −0.020 −0.061 −0.006
(−0.36) (−1.56) (−0.11)
Export 0.013* 0.008* 0.016*
(1.68) (1.74) (1.67)
Constants 0.378*** −2.670*** 0.144*** −1.348** 0.294*** −2.790***
(31.16) (−2.77) (14.84) (−2.20) (26.25) (−3.06)
Firm-xed effects Yes Yes Yes Yes Yes Yes
Year-xed effects Yes Yes Yes Yes Yes Yes
Observations 6677 6677 6677 6677 6677 6677
R-squared 0.6886 0.6917 0.5828 0.5862 0.6511 0.6545
Notes: ***p < 0.01, **p < 0.05, *p < 0.1. t-statistics in parentheses. Standard errors are clustered at
the rm level.
Environmental Regulation and Green Innovation 145
The Direct Effect of MER on CGI. Table 5 reports the effect of MERs on CGI. Columns
(1) to (3) control for both rm and year xed effects, as well as the covariates Xit . The
results show that the intensity of the government’s mandatory environmental regulation
is insignicant for rms’ green innovation performance, and even reduces the number of
Figure 2. Parallel trend test
Table 5. Effect of MER on CGI
(1) (2) (3)
Grepatent Greinvent Greutility
MER −0.063 −0.088*** −0.027
(−1.63) (−3.13) (−0.68)
Constants 1.720 0.886 0.387
(1.17) (1.01) (0.22)
Controls Yes Yes Yes
Firm-xed effects Yes Yes Yes
Year-xed effects Yes Yes Yes
Observations 1,477 1,477 1,477
R-squared 0.8419 0.7701 0.7924
Notes: ***p < 0.01, **p < 0.05, *p < 0.1. t-statistics in parentheses. Standard
errors are clustered at the rm level.
146 Z. Yang et al.
green invention patent authorizations for enterprises. The ndings are almost similar to
the research of Camison-Zornoza and Boronat-Navarro (2010), Li et al. (2020), and
Wang et al. (2023), which all nd that MERs do not have a positive effect on corporate
technological innovation. A possible reason for this is that mandatory environmental
regulations may increase costs for enterprises and the passive regulation approach is
unlikely to stimulate the internal innovation awareness of enterprises, thus MER does not
have a substantive positive effect on rms’ green innovation behaviors.
The Relationship between VER and MER. To further explore the relationship between
VERs and MERs, this study denes the sample with the highest 25 per cent mandatory
environmental regulation intensity as the high MER group, and the sample with the
lowest 25 per cent mandatory environmental regulation intensity as the low MER group
and conduct the regression on these two sub-samples. The results in Table 6 indicate that
when MER is relatively low, VER has a stronger promoting effect on rms’ green
innovation. However, in the group with the highest MER, the positive effect of VER
on CGI is no longer signicant. This indicates that in areas where government mandatory
environmental regulations are weak, voluntary environmental regulation can serve as a
supplement to promote rms’ green innovation.
Robustness Checks
Randomly Generated VER Status. We also need to rule out the possibility that the nding
is a random result without any substantive meaning. For example, enterprises’ awareness
of social responsibility and environmental protection has generally improved over time,
and enterprises pay more attention to green transformation, which eventually leads to an
Table 6. The effects of VER under different MER intensity
Low MER group High MER group
Grepatent Greinvent Greutility Grepatent Greinvent Greutility
(1) (2) (3) (4) (5) (6)
VER 0.175** 0.118** 0.120** 0.097 0.009 0.086
(2.49) (2.15) (2.07) (1.35) (0.20) (1.31)
Constants −2.204 −1.506 −2.975 2.527 1.190 1.584
(−1.01) (−0.99) (−1.43) (1.45) (1.19) (0.94)
Controls Yes Yes Yes Yes Yes Yes
Firm-xed effects Yes Yes Yes Yes Yes Yes
Year-xed effects Yes Yes Yes Yes Yes Yes
Observations 1,251 1,251 1,251 1,110 1,110 1,110
R-squared 0.6555 0.5450 0.6480 0.6859 0.5419 0.6432
Notes: ***p < 0.01, **p < 0.05, *p < 0.1. t-statistics in parentheses. Standard errors are clustered at
the rm level.
Environmental Regulation and Green Innovation 147
increase in green innovation patents (Chen et al. 2023). In that case, the improvement of
corporate green innovation has nothing to do with adopting VER (ISO14001 certica-
tion). In order to rule out this possibility, referring to the practice of Chetty et al. (2009),
we conduct placebo tests by randomly arranging ISO14001 certied rms and years and
repeat this process 500 times. The results are shown in Figure 3. The estimated
coefcients obtained by random simulation are all distributed around the zero point,
and most of the estimated values have p-values greater than 0.1 (not signicant at the
10 per cent level). The effects of VER on CGI are not driven by random and accidental
factors.
PSM-DID. In addition, there is a possible concern that the choice of enterprises to adopt
VER is not a random behavior, nor is it determined by exogenous factors, but it is a choice
made by enterprises according to their own business conditions. Enterprises with higher
levels of green innovation may be more inclined to carry out VER. Therefore, there may be
a self-selection bias. So we overcome the possible interference of this problem by
performing Propensity Score Matching (PSM).
The specic PSM approach we applied is the 1:1 nearest neighbor matching.
6
Firstly, we
select the matching variables including Size, Roe, Lev, Growth, Board, Indep, Soe, Top1, Inst,
and Export, and compute the propensity score according to these matching variables.
Secondly, we sort the treated group and controlled group based on the propensity score.
Thirdly, we select one rm from the treated group in sequence, then select one rm with the
closest propensity score in the controlled group as the matching object. Furthermore, we
conduct the balanced test and the results are presented in Figure 4, which implies that the
Figure 3. Placebo test
148 Z. Yang et al.
standardized deviation of the matched samples is less than 10 per cent, so that the difference
between the matched treated group and the controlled group decreased signicantly compared
with before matching. The balanced test result shows that in the matched sample, almost all t-
test results accept the null hypothesis that there is no systematic difference between the
matched treated group and the controlled group, thus satisfying the hypothesis of balance.
We re-test Equation (2) with the matched samples and the results are shown in Table 7. It
can be found that VER still has a signicant positive impact on the green innovation output
of enterprises, which is consistent with the benchmark regression results, proving that our
results are robust.
Control Other Factors That Potentially Affect Corporate Green Innovation. Another
possible concern is that other types of certication obtained by enterprises may also have a
potential impact on corporate green innovation. For example, many manufacturing enterprises
will choose to carry out ISO9001 certication. ISO9001 certication requires enterprises to
apply international quality standards, which can improve efciency and reduce costs, thus
helping to promote innovative activities (Abrunhosa and Sa 2008; Pekovic and Galia 2009). In
addition, the Energy Management System Certication has been implemented since 2013. This
certication is also conducive to helping enterprises improve energy efciency and establish the
concept of energy conservation and emissions reduction, which may promote the green
transformation and green innovation activities of enterprises. Therefore, we add the control
variables of whether the enterprise has obtained the ISO9001 certication and whether the
enterprise has obtained the Energy Management System Certication. The results are shown in
Figure 4. Balanced test
Environmental Regulation and Green Innovation 149
Table 8. It can be found that after controlling for the two variables, the DID estimator is still
signicant at the signicance level of 1 per cent.
Control City Time Trends. Furthermore, due to the implementation of many environmental
policies in various cities, these policies may also have an impact on the green innovation of
enterprises. This study cannot identify all environmental protection policies at the city level
and test them one by one. Therefore, by further adding the interaction term of city dummy
variables and time trend terms to the regression equation, we can control the impact of all
urban-level factors on the estimation results in the linear dimension. The results of Table 9
indicate that the effect of VER slightly decreases compared to the benchmark regressions,
but is still signicant at the 1 per cent or 5 per cent level.
Table 8. Control for other factors that potentially affect rms’ green innovation
(1) (2) (3)
Grepatent Greinvent Greutility
VER 0.125*** 0.081*** 0.090***
(3.38) (2.76) (2.67)
Constants −2.707*** −1.367** −2.823***
(−2.81) (−2.24) (−3.09)
ISO9001&Ener_manage Yes Yes Yes
Controls Yes Yes Yes
Firm-xed effects Yes Yes Yes
Year-xed effects Yes Yes Yes
Observations 6,677 6,677 6,677
R-squared 0.6922 0.5870 0.6550
Notes: ***p < 0.01, **p < 0.05, *p < 0.1. t-statistics in parentheses. Standard errors
are clustered at the rm level.
Table 7. PSM-DID analysis
(1) (2) (3)
Grepatent Greinvent Greutility
VER 0.167*** 0.066* 0.149***
(2.84) (1.90) (2.69)
Constants −0.692 −0.056 −0.757
(−0.85) (−0.12) (−0.99)
Controls Yes Yes Yes
Firm-xed effects Yes Yes Yes
Year-xed effects Yes Yes Yes
Observations 2,466 2,466 2,466
R-squared 0.7099 0.6672 0.6791
Notes: ***p < 0.01, **p < 0.05, *p < 0.1. t-statistics in parentheses. Standard
errors are clustered at the rm level.
150 Z. Yang et al.
Imputation DID. Recent literature suggests that the TWFE estimates may still be biased if
there are more than two time periods before and after the treatment (De Chaisemartin and
D’Haultfoeuille 2020; Borusyak et al. 2024; Callaway and Sant’Anna 2021; Goodman-
Bacon 2021; Sun and Abraham 2021). If there are more than two periods and groups are
treated at different points in time, as in the context of the present study, the robustness of
TWFE to treatment heterogeneity may not hold. In other words, treatment effects may be
heterogeneous across different rms or exhibit dynamics or changes across different time
periods. For example, in a TWFE DID regression with two time periods, rms that had
never received the ISO14001 certication serve as the comparison group for rms that
received the ISO14001 certication at some time. However, with multiple time periods and
variation of treatment (ISO14001) timing, the identication of the ATT in a conventional
TWFE DID model relies on comparisons between newly certied rms and (1) never
certied rms; (2) not-yet certied rms; and (3) already certied rms. The rst and
second comparisons take the path of outcomes experienced by the newly certied rms
and adjust it to the path of outcomes experienced by rms that have not (yet) obtained
ISO14001 certication. The third comparison adjusts the path of outcomes for newly
certied rms by the path of outcomes for already certied rms; however, this is not
the path of untreated potential outcomes since it includes treatment effect dynamics,
making it difcult to give a clear causal interpretation of β, which is essentially a weighted
average of ATT, in Equation (1). Ignoring the treatment effect dynamics in TWFE DID
with multiple periods and variation of treatment timing can lead to a negatively weighted
average of ATT, even if the effects of obtaining the ISO14001 certication is positive for
all rms in all time periods or an insignicant weighted average of ATT if the effects of
ISO14001 are statistically insignicant in some cohorts of certied rms, but signicant in
others. While our Equations (1) and (2) using TWFE can capture some of the relative
differences in treatment effects between treated and untreated rms during or after the
validity period of the ISO14001 certication, it cannot capture differences in treatment
timing or the full dynamics of treatment effects. Thus, we employ a staggered DID
estimator with multiple time periods and variations in treatment timing.
Table 10 presents the single average ATT across all treated observations over all post-
award periods (years during and after the validity of the ISO14001 certicate) using the DID
Table 9. Control for the city time trends
(1) (2) (3)
Grepatent Greinvent Greutility
VER 0.109*** 0.053** 0.087***
(3.50) (2.05) (2.94)
Constants −1.185 −0.362 −1.668*
(−1.32) (−0.62) (−1.93)
Controls Yes Yes Yes
Firm-xed effects Yes Yes Yes
Year-xed effects Yes Yes Yes
City time trends Yes Yes Yes
Observations 6,675 6,675 6,675
R-squared 0.7269 0.6290 0.6904
Notes: ***p < 0.01, **p < 0.05, *p < 0.1. t-statistics in parentheses. Standard
errors are clustered at the rm level.
Environmental Regulation and Green Innovation 151
design with staggered adoption of treatment and the imputation approach of Borusyak et al.
(2024). It shows that obtaining an ISO14001 certicate has a positive effect on all the green
patents outcomes. The coefcients from staggered DID using an imputation estimator
suggest that TWFE DID in Table 4 underestimates the effects of the VER (ISO14001
certication), which suggests that it is crucial to capture the full spectrum of impacts over
the years and across different treatment times for individual rms.
Further Regressions
Furthermore, the issue of concern in this study is whether environmental regulatory tools
have heterogeneous effects among industries with different levels of pollution. Referring
to Liu and Liu (2015), this study selects 13 sub-sectors of the manufacturing industry as
heavily polluting industries
7
based on the Guidelines for Industry Classication of Listed
Companies (2012) and the Catalogue of Environmental Protection Verication Industry
Classication Management of Listed Companies (2008). Table 11 examines the
Table 10. Imputation DID
(1) (2) (3)
Grepatent Greinvent Greutility
VER 0.163*** 0.094*** 0.129***
(4.27) (3.16) (3.76)
Controls Yes Yes Yes
Firm-xed effects Yes Yes Yes
Year-xed effects Yes Yes Yes
Observations 5,933 5,933 5,933
Notes: ***p < 0.01, **p < 0.05, *p < 0.1. t-statistics in parentheses. Standard
errors are clustered at the rm level.
Table 11. The effects of VER for different industries
Heavily polluting industries Non-heavily polluting industries
Grepatent Greinvent Greutility Grepatent Greinvent Greutility
(1) (2) (3) (4) (5) (6)
VER 0.061 0.040 0.033 0.184*** 0.093** 0.146***
(1.47) (1.37) (0.95) (3.96) (2.39) (3.26)
Constants −1.382 −0.748 −1.367 −3.458** −1.566* −3.646***
(−1.10) (−0.90) (−1.21) (−2.50) (−1.85) (−2.71)
Controls Yes Yes Yes Yes Yes Yes
Firm-xed effects Yes Yes Yes Yes Yes Yes
Year-xed effects Yes Yes Yes Yes Yes Yes
Obs 2,730 2,730 2,730 3,936 3,936 3,936
R-squared 0.6182 0.5491 0.5974 0.7121 0.6065 0.6632
Notes: ***p < 0.01, **p < 0.05, *p < 0.1. t-statistics in parentheses. Standard errors are clustered at
the rm level.
152 Z. Yang et al.
heterogeneity impacts of VER in heavily polluting and non-heavily polluting industries.
The results indicate that the positively signicant effect of VER on green innovation is
only found in non-heavily polluting industries. Table 12 examines the heterogeneous
effects of MER in heavily polluting and non-heavily polluting industries, and the results
indicate that the role of MER is not heterogeneous across industries and cannot promote
green innovation in enterprises.
Conclusions
This paper uses a DID model to test the effect of VERs and MERs on corporate green
innovation in China. The main conclusions are as follows: rst, VERs are generally
more effective in promoting corporate green innovation in China compared with
MERs. The empirical results show that VERs have a signicant positive effect on
corporate green innovation, while MERs have no or negative effect on green innova-
tion. Second, in areas where government mandatory environmental regulations are
weaker, VERs are found to be more effective in promoting corporate green innovation.
Third, VERs’ effect on green innovation is only signicant in non-heavily polluting
industries.
The managerial and practical implications of this paper are as follows: First, it will
encourage rms to adopt more VERs to improve their ability in green innovation.
China is now experiencing an economic slowdown, and having to reduce carbon
emissions at the same time. Green innovation is of key importance for China. The
government should place more importance on encouraging adoption of VERs.
However, VER is not affordable for many rms. On the one hand, government can
subsidize rms, especially medium-sized and small rms, in their adoption of VERs –
adoption of ISO14001. On the other hand, government can provide more opportunities
for rms to get and understand the information about environmental VERs, such as
Table 12. The effects of MER for different industries
Heavily polluting industries Non-heavily polluting industries
Grepatent Greinvent Greutility Grepatent Greinvent Greutility
(1) (2) (3) (4) (5) (6)
MER −0.076 −0.036 −0.082* −0.053 −0.109*** 0.004
(−1.59) (−1.03) (−1.67) (−1.03) (−2.86) (0.07)
Constants 2.766 0.216 2.780 1.964 0.841 0.355
(1.15) (0.11) (1.31) (0.96) (0.81) (0.14)
Controls Yes Yes Yes Yes Yes Yes
Firm-xed effects Yes Yes Yes Yes Yes Yes
Year-xed effects Yes Yes Yes Yes Yes Yes
Observations 483 483 483 990 990 990
R-squared 0.8478 0.8210 0.8166 0.8369 0.7570 0.7815
Notes: ***p < 0.01, **p < 0.05, *p < 0.1. t-statistics in parentheses. Standard errors are clustered at
the rm level.
Environmental Regulation and Green Innovation 153
providing free consultation services for environmental certication. Second, it will
promote the reform of MERs. Although we nd the effect of MERs on green innova-
tion is insignicant, we will not propose that MER is not important in environmental
governance. Further, we suggest that government should acknowledge the potential
problems with the MER method. As there are many tools for MERs (hard and soft
tools), government can utilize the advantages of each tool, and enrich the toolbox of
environmental governance. Besides, local government can play an active role in VERs
when considering MERs. For example, government should reduce on-site inspections
for rms with voluntary environmental certications. Third, rms can actively adopt
environmental certication and other VERs to meet the challenge of stricter environ-
mental protection. As China has proposed the dual carbon goal, rms will face stricter
regulation on the environment. The transition to green innovation is important for them
to reduce the costs of regulation and increase their competitiveness.
Notes
1. See “Opinions on Accelerating the Construction of a Unied National Market”, issued by the CPC Central
Committee and State Council on April 10, 2022. For more details of the notice, see http://www.gov.cn/
zhengce/2022-04/10/content_5684385.htm.
2. A comprehensive list of environmental laws, environment-related laws and regulations in China can be
found at http://english.mee.gov.cn/Resources/laws/.
3. See Key Points of Government Information Disclosure in 2015, issued by the State Council on April
21, 2015. More details at http://www.gov.cn/zhengce/content/2015-04/21/content_9644.htm. See
Regulations of the People’s Republic of China on the Disclosure of Government Information, issued
by the State Council on April 3, 2019. More details at http://www.gov.cn/zhengce/2020-12/27/
content_5573650.htm.
4. Data from the ISO survey released in Mar 18th, 2024. More details at https://www.iso.org/the-iso-survey.
html.
5. See “Opinions on Accelerating the Construction of a Unied National Market”, issued by the CPC Central
Committee and State Council on April 10, 2022. More details at http://www.gov.cn/zhengce/2022-04/10/
content_5684385.htm.
6. The Kernel matching method is also used to construct the PSM sample and then make the PSM-DID
analysis. We do not present the results. Interested readers can contact the author to obtain them.
7. Heavily polluting industries include C15, C17, C18, C19, C22, C25, C26, C28, C29, C30, C31, C32,
and C33.
Acknowledgements
The paper has greatly beneted from the helpful comment of two anonymous reviewers. We thank Tom
Christensen, Department of Political Science, University of Oslo for the insightful suggestions on the workshop
in College of Public Administration, Huazhong University of Science and Technology. We acknowledge the
nancial support from the National Natural Science Foundation of China (grant no.7217030601,71933004).
Disclosure Statement
No potential conict of interest was reported by the authors.
154 Z. Yang et al.
Funding
This work was supported by the National Natural Science Foundation of China [71933004].
ORCID
Zhiqing Yang http://orcid.org/0009-0001-4971-347X
References
Abrunhosa, A. and Sa, P. M. E., 2008, Are TQM principles supporting innovation in the portuguese footwear
industry? Technovation, 28(4), pp. 208–221. doi:10.1016/j.technovation.2007.08.001
Alberini, A. and Segerson, K., 2002, Assessing voluntary programs to improve environmental quality.
Environmental and Resource Economics, 22(1–2), pp. 157–184. doi:10.1023/A:1015519116167
Anton, W., Deltas, G., and Khanna, M., 2004, Incentives for environmental self-regulation and implications for
environmental performance. Journal of Environmental Economics and Management, 48(1), pp. 632–654.
doi:10.1016/j.jeem.2003.06.003
Aragon-Correa, J. A., Marcus, A., and Hurtado-Torres, N., 2016, The natural environmental strategies of
international rms: Old controversies and new evidence on performance and disclosure. Academy of
Management Perspectives, 30(1), pp. 24–39. doi:10.5465/amp.2014.0043
Aragon-Correa, J. A., Marcus, A. A., and Vogel, D., 2020, The effects of mandatory and voluntary regulatory
pressures on rms’ environmental strategies: A review and recommendations for future research. Academy of
Management Annals, 14(1), pp. 339–365. doi:10.5465/annals.2018.0014
Arora, S. and Cason, T., 1995, An experiment in voluntary environmental regulation: Participation in EPA’s 33/
50 program. Journal of Environmental Economics and Management, 28(3), pp. 271–286. doi:10.1006/jeem.
1995.1018
Bae, S. M., Masud, M. A. K., and Kim, J. D., 2018, A cross-country investigation of corporate governance and
corporate sustainability disclosure: A signaling theory perspective. Sustainability, 10(8), pp. 2611. doi:10.
3390/su10082611
Beck, T., Levine, R., and Levkov, A., 2010, Big bad banks? The winners and losers from bank deregulation in
the United States. The Journal of Finance, 65(5), pp. 1637–1667. doi:10.1111/j.1540-6261.2010.01589.x
Berliner, D. and Prakash, A., 2015, “Bluewashing” the rm? Voluntary regulations, program design, and
member compliance with the United Nations Global compact. Policy Studies Journal, 43(1), pp. 115–138.
doi:10.1111/psj.12085
Berrone, P., Fosfuri, A., Gelabert, L., and Gomez-Mejia, L. R., 2013, Necessity as the mother of “green”
inventions: Institutional pressures and environmental innovations. Strategic Management Journal, 34(8), pp.
891–909. doi:10.1002/smj.2041
Blackman, A., 2008, Can voluntary environmental regulation work in developing countries? Lessons from case
studies. Policy Studies Journal, 36(1), pp. 119–141. doi:10.1111/j.1541-0072.2007.00256.x
Blackman, A., Lahiri, B., Pizer, W., Rivera Planter, M., and Munoz Pina, C., 2010, Voluntary environmental
regulation in developing countries: Mexico’s clean industry program. Journal of Environmental Economics
and Managemnet, 60(3), pp. 182–192. doi:10.1016/j.jeem.2010.05.006
Borusyak, K., Jaravel, X., and Spiess, J., 2024, Revisiting event study designs: Robust and efcient estimation.
The Review of Economics Studies, rdae007. doi:10.1093/restud/rdae007
Bu, M., Qiao, Z., and Liu, B., 2020, Voluntary environmental regulation and rm innovation in China.
Economic Modelling, 89, pp. 10–18. doi:10.1016/j.econmod.2019.12.020
Callaway, B. and Sant’Anna, P. H. C., 2021, Difference-in-differences with multiple time periods. Journal of
Econometrics, 225(2), pp. 200–230. doi:10.1016/j.jeconom.2020.12.001
Camison-Zornoza, C. and Boronat-Navarro, M., 2010, Does regulation perform better than self-regulation? An
analysis of spanish environmental policies. Environment and Planning C: Government and Policy, 28(4), pp.
733–758. doi:10.1068/c0944r
Carmin, J., Darnall, N., and Mil-Homens, J., 2003, Stakeholder involvement in the design of US voluntary
environmental programs: does sponsorship matter? Policy Studies Journal, 31(4), pp. 527–543. doi:10.1111/
1541-0072.00041
Environmental Regulation and Green Innovation 155
Chang, S.-J. and Wu, B., 2014, Institutional barriers and industry dynamics. Strategic Management Journal, 35
(8), pp. 1103–1123. doi:10.1002/smj.2152
Chen, P. Y., Dagestani, A. A., and Kim, S., 2023, Corporate social responsibility and green exploratory
innovation - the moderating role of three environmental regulations. Technology Analysis & Strategic
Management, pp. 1–13. doi:10.1080/09537325.2023.2196585
Chen, Q., Chen, Z., and Liu, Z., 2021, Regulating conglomerates in China: Evidence from an energy
conservation Program. NBER Working Paper, No. w29066. ssrn.com/abstract=3893347
Chetty, R., Looney, A., and Kroft, K., 2009, Salience and taxation: Theory and evidence. American Economic
Review, 99(4), pp. 1145–1177. doi:10.1257/aer.99.4.1145
Cordano, M., Frieze, I., and Ellis, K., 2004, Entangled afliations and attitudes: An analysis of the inuences
on environmental policy stakeholders’ behavioral intentions. Journal of Business Ethics, 49(1), pp. 27–40.
doi:10.1023/B:BUSI.0000013850.14353.b6
Cull, R., Xu, L. C., Yang, X., Zhou, L.-A., and Zhu, T., 2017, Market facilitation by local government and rm
efciency: Evidence from China. Journal of Corporate Finance, 42, pp. 460–480. doi:10.1016/j.jcorpn.
2015.06.002
Darnall, N. and Carmin, J., 2005, Greener and cleaner? The signaling accuracy of US voluntary environmental
programs. Policy Sciences, 38(2–3), pp. 71–90. doi:10.1007/s11077-005-6591-9
De Chaisemartin, C. and D’Haultfoeuille, X., 2020, Two-way xed effects estimators with heterogeneous
treatment effects. American Economic Review, 110(9), pp. 2964–2996. doi:10.1257/aer.20181169
Delmas, M. A. and Montes-Sancho, M. J., 2010, Voluntary agreements to improve environmental quality:
Symbolic and substantive cooperation. Strategic Management Journal, 31(6), pp. 575–601. doi:10.1002/
smj.826
Drejer, I., 2004, Identifying innovation in surveys of services: A Schumpeterian perspective. Research Policy,
33(3), pp. 551–562. doi:10.1016/j.respol.2003.07.004
Goodman-Bacon, A., 2021, Difference-in-differences with variation in treatment timing. Journal of
Econometrics, 225(2), pp. 254–277. doi:10.1016/j.jeconom.2021.03.014
Grossman, G. M. and Helpman, E., 1990, Trade, innovation and growth. American Economic Review, 80(2),
pp. 86–91.
Hart, S., 1995, A natural-resource-based view of the rm. Academy of Management Review, 20(4), pp. 986–
1014. doi:10.2307/258963
He, W., Yang, W., and Choi, S., 2018, The interplay between private and public regulations: Evidence from ISO
14001 adoption among Chinese rms. Journal of Business Ethics, 152(2), pp. 477–497. doi:10.1007/s10551-
016-3280-x
Iatridis, K. and Kesidou, E., 2018, What drives substantive versus symbolic implementation of ISO 14001 in a
time of economic crisis? Insights from Greek manufacturing companies. Journal of Business Ethics, 148(4),
pp. 859–877. doi:10.1007/s10551-016-3019-8
Jaffe, A., Newell, R., and Stavins, R., 2002, Environmental policy and technological change. Environmental
and Resource Economics, 22(1– 2), pp. 41–69. doi:10.1023/A:1015519401088
Jiang, Z., Wang, Z., and Lan, X., 2021, How environmental regulations affect corporate innovation? The
coupling mechanism of mandatory rules and voluntary management. Technology in Society, 65, pp. 101575.
doi:10.1016/j.techsoc.2021.101575
Jiang, Z., Wang, Z., and Zeng, Y., 2020, Can voluntary environmental regulation promote corporate techno-
logical innovation? Business Strategy and the Environment, 29(2), pp. 390–406. doi:10.1002/bse.2372
Johnstone, N., Hascic, I., and Popp, D., 2010, Renewable energy policies and technological innovation:
Evidence based on patent counts. Environmental and Resource Economics, 45(1), pp. 133–155. doi:10.
1007/s10640-009-9309-1
Khanna, M. and Damon, L., 1999, EPA’s voluntary 33/50 Program: Impact on toxic releases and economic
performance of rms. Journal of Environmental Economics and Management, 37(1), pp. 1–25. doi:10.1006/
jeem.1998.1057
King, A. and Lenox, M., 2000, Industry self-regulation without sanctions: The chemical industry’s responsible
care program. Academy of Management Journal, 43(4), pp. 698–716. doi:10.5465/1556362
Kumar, S. and Shetty, S., 2018, Corporate participation in voluntary environmental programs in India:
Determinants and deterrence. Ecological Economics, 147, pp. 1–10. doi:10.1016/j.ecolecon.2017.12.029
Li, D. Y., Tang, F., and Zhang, L., 2020, Differential effects of voluntary environmental programs and
mandatory regulations on corporate green innovation. Natural Hazards, 103(3), pp. 3437–3456. doi:10.
1007/s11069-020-04137-y
156 Z. Yang et al.
Lim, S. and Prakash, A., 2014, Voluntary regulations and innovation: The case of ISO 14001. Public
Administration Review, 74(2), pp. 233–244. doi:10.1111/puar.12189
Liu, Y. G. and Liu, M. N., 2015, Have smog affected earning management of heavy-polluting enterprises?——
based on the political-cost hypothesis. Accounting Research, 3, pp. 26–33+94. (In Chinese)
Magat, W. A., 1978, Pollution control and technological advance: A dynamic model of the rm. Journal of
Environmental Economics and Management, 5(1), pp. 1–25. doi:10.1016/0095-0696(78)90002-5
Marcus, A. A., 1980, Promise and Performance: Choosing and Implementing an Environmental Policy
(Westport, CT: Greenwood Press).
Pan, F., Xu, J., and Xue, L., 2020, Voluntary environmental program: Research progress and future prospect.
China Population Resources and Environment, 30(1), pp. 74–82. doi:10.1002/pam.20617
Pang, R., Zheng, D., Shi, M., and Zhang, X., 2019, Pollute rst, control later? exploring the economic threshold
of effective environmental regulation in China’s context. Journal of Environmental Management, 248, pp.
109275. doi:10.1016/j.jenvman.2019.109275
Pekovic, S. and Galia, F., 2009, From quality to innovation: Evidence from two french employer surveys.
Technovation, 29(12), pp. 829–842. doi:10.1016/j.technovation.2009.08.002
Peng, M. W., Wang, D. Y. L., and Jiang, Y., 2008, An institution-based view of international business strategy:
A focus on emerging economies. Journal of Internaional Business Studies, 39(5), pp. 920–936. doi:10.1057/
palgrave.jibs.8400377
Petitjean, M., 2019, Eco-friendly policies and nancial performance: Was the nancial crisis a game changer
for large US companies? Energy Economics, 80, pp. 502–511. doi:10.1016/j.eneco.2019.01.028
Picard, D., 2015, Making ecotourism sustainable: Refocusing on economic viability. lessons learnt from the
“regional strategic action plan for coastal ecotourism development in the South Western Indian Ocean.
Journal of Sustainable Tourism, 23(6), pp. 819–837. doi:10.1080/09669582.2015.1019512
Popp, D., Hafner, T., and Johnstone, N., 2011, Environmental policy vs. public pressure: Innovation and
diffusion of alternative bleaching technologies in the pulp industry. Research Policy, 40(9), pp. 1253–1268.
doi:10.1016/j.respol.2011.05.018
Porter, M. E. and van der Linde, C., 1995, Toward a new conception of the environment-competitiveness
relationship. Journal of Economic Perspectives, 9(4), pp. 97–118. doi:10.1257/jep.9.4.97
Potoski, M. and Prakash, A., 2013, Do voluntary programs reduce pollution? examining ISO 14001ʹs effec-
tiveness across Countries. Policy Studies Journal, 41(2), pp. 273–294. doi:10.1111/psj.12017
Prakash, A. and Potoski, M., 2007, Collective action through voluntary environmental programs: A club theory
perspective. Policy Studies Journal, 35(4), pp. 773–792. doi:10.1111/j.1541-0072.2007.00247.x
Qin, Y. and Sun, H., 2020, Relationship between voluntary participation in environmental regulation and R&D:
Based on government supervision and media attention. Science and Technology Management Research, 40
(4), pp. 254–262. (In Chinese)
Ren, S., He, D., Yan, J., Zeng, H., and Tan, J., 2022, Environmental labeling certication and corporate
environmental innovation: The moderating role of corporate ownership and local government intervention.
Journal of Business Research, 140, pp. 556–571. doi:10.1016/j.jbusres.2021.11.023
Rennings, K., Ziegler, A., Ankele, K., and Hoffmann, E., 2006, The inuence of different characteristics of the
EU environmental management and auditing scheme on technical environmental innovations and economic
performance. Ecological Economics, 57(1), pp. 45–59. doi:10.1016/j.ecolecon.2005.03.013
Rivera Ungson, G., James, C., and Spicer, B. H., 1985, The effects of regulatory agencies on organizations in
wood products and high technology/electronics industries. Academy of Management Journal, 28(2), pp. 426–
445. doi:10.5465/256209
Segerson, K. and Miceli, T., 1998, Voluntary environmental agreements: Good or bad news for environmental
protection? Journal of Environmental Economics and Management, 36(2), pp. 109–130. doi:10.1006/jeem.
1998.1040
Short, J. L. and Toffel, M. W., 2010, Making self-regulation more than Merely symbolic: The Critical role of
the legal environment. Administrative Science Quarterly, 55(3), pp. 361–396. doi:10.2189/asqu.2010.55.
3.361
Stigler, G. J., 1975, The Citizen and the State: Essays on Regulation. Chicago: University of Chicago Press.
Stoeckl, N., 2010, The private costs and benets of environmental self-regulation: Which rms have most to
gain? Business Strategy and the Environment, 13(3), pp. 135–155. doi:10.1002/bse.405
Sun, L. and Abraham, S., 2021, Estimating dynamic treatment effects in event studies with heterogeneous
treatment effects. Journal of Econometrics, 225(2), pp. 175–199. doi:10.1016/j.jeconom.2020.09.006
Environmental Regulation and Green Innovation 157
Tambunlerchai, K., Kontoleon, A., and Khanna, M., 2012, Assessing participation in voluntary environmental
programmes in the developing world: The role of FDI and export orientation on ISO14001 adoption in
Thailand. Applied Economics, 45(15), pp. 2039–2048. doi:10.1080/00036846.2011.648320
Testa, F., Iraldo, F., and Daddi, T., 2018, The effectiveness of EMAS as a Management tool: A key role for the
internalization of environmental practices. Organization & Environment, 31(1), pp. 48–69. doi:10.1177/
1086026616687609
Videras, J. and Alberini, A., 2000, The appeal of voluntary environmental programs: Which rms participate
and why? Contemporary Economic Policy, 18(4), pp. 449–461. doi:10.1093/cep/18.4.449
Wang, A., Si, L., and Hu, S., 2023, Can the penalty mechanism of mandatory environmental regulations
promote green innovation? evidence from China’s enterprise data. Energy Economics, 125, pp. 106856.
doi:10.1016/j.eneco.2023.106856
Wang, C., Yang, Y., and Zhang, J., 2015, China’s sectoral strategies in energy conservation and carbon
mitigation, Climate Policy. Clinical Lymphoma, Myeloma & Leukemia, 15 Suppl, pp. S60–S80. doi:10.
1080/14693062.2015.1050346
Wang, X., Cho, S.-H., and Scheller-Wolf, A., 2021, Green technology development and adoption: Competition,
regulation, and uncertainty-A global game approach. Management Science, 67(1), pp. 201–219. doi:10.1287/
mnsc.2019.3538
Wang, Y., 2016, On the necessity of the application of voluntary environment agreements in China: From the
perspective of government regulation. Ecological Economy, 32(9), pp. 145–151.
Williams, C. and Martinez, C. A., 2012, Government effectiveness, the global nancial crisis, and multinational
enterprise internationalization. Journal of International Marketing, 20(3), pp. 65–78. doi:10.1509/jim.12.
0078
Wu, L. C., Cong, S., Lin, L., and Li, J. X., 2022, Theoretical and empirical research on the impact of exports on
green technology innovation in enterprises. Southern Economy, 8, pp. 52–72. (In Chinese)
Ye, C. G., Wang, Z., Wu, J. F., and Li, H., 2015, External governance, environmental information disclosure
and the cost of equity nancing. Nankai Business Review, 18(5), pp. 85–96. (In Chinese)
York, J. G., Vedula, S., and Lenox, M., 2018, It’s not easy building green: The impact of public policy, private
actors, and regional logics on voluntary standards adoption. Academy of Management Journal, 61(4), pp.
1492–1523. doi:10.5465/amj.2015.0769
Zhang, H., Zhu, Z., and Fan, Y., 2018, The impact of environmental regulation on the coordinated development
of environment and economy in China. Natural Hazards, 91(2), pp. 473–489. doi:10.1007/s11069-017-
3137-3
Zhang, J., Zhang, W., Song, Q., Li, X., Ye, X., Liu, Y., and Xue, Y., 2020, Can energy saving policies drive rm
innovation behaviors? - evidence from China. Technology Forecasting and Social Change, 154, pp. 119953.
doi:10.1016/j.techfore.2020.119953
Zhu, X., Zuo, X., and Li, H., 2021, The dual effects of heterogeneous environmental regulation on the
technological innovation of Chinese steel enterprises-based on a high-dimensional xed effects model.
Ecological Economics, 188, pp. 107113. doi:10.1016/j.ecolecon.2021.107113
Zizzo, D. J. and Fleming, P., 2011, Can experimental measures of sensitivity to social pressure predict public
good contribution? Economics Letters, 111(3), pp. 239–242. doi:10.1016/j.econlet.2011.02.021
158 Z. Yang et al.
