02 - Fenvs-11-1305539

TYPE Original Research
PUBLISHED 08 January 2024

DOI 10.3389/fenvs.2023.1305539
How do smart city pilots affect the

OPEN ACCESS ESG performance of
EDITED BY
Vinay Kandpal,
University of Petroleum and Energy
manufacturing firms? evidence
Studies, India
REVIEWED BY
from China
Otilia Manta,
Romanian Academy, Romania
Serife Eyupoglu,
Haisheng Tang 1,2, Jeng-Bang Wang 3 and Chung-Ya Ou 2*
Near East University, Cyprus 1
Party School of the Chaozhou Committee of C.P.C, Chaozhou, China, 2School of Public Administration,
*CORRESPONDENCE Nanfang College, Guangzhou, China, 3College of Art Design, Hsuan Chuang University, Hsinchu, Taiwan
Chung-Ya Ou,
[email protected]
RECEIVED 02 October 2023

ACCEPTED 18 December 2023 Introduction: The existing literature extensively covers factors influencing
PUBLISHED 08 January 2024
environmental, social, and governance (ESG) performance. However, there’s a
CITATION
lack of studies exploring the relationship between smart cities and ESG
Tang H, Wang J-B and Ou C-Y (2024),
How do smart city pilots affect the ESG performance. This gap is surprising, particularly considering the significant role
performance of manufacturing firms? that governments in developing countries play in leading smart city initiatives.
evidence from China.
Front. Environ. Sci. 11:1305539. Methods: As the impact of smart city pilot (SCP) policies, initiated by governments,
doi: 10.3389/fenvs.2023.1305539
on the ESG performance of manufacturing firms lacks verification, our study
COPYRIGHT leverages SCP as a quasi-experiment to examine the effects of smart city
© 2024 Tang, Wang and Ou. This is an
development on this performance. Data was collected from 2,229 listed
open-access article distributed under the
terms of the Creative Commons manufacturing firms in China from 2009 to 2020.
Attribution License (CC BY). The use,
distribution or reproduction in other Results: Our findings indicate a substantial increase in ESG performance among
forums is permitted, provided the original manufacturing firms due to SCP implementation. Heterogeneity analysis reveals that
author(s) and the copyright owner(s) are
the positive influence of SCP is primarily advantageous for non-state-owned
credited and that the original publication
in this journal is cited, in accordance with enterprises. Geographically, SCP significantly enhances corporate ESG
accepted academic practice. No use, performance in eastern and central China, with more muted effects observed in
distribution or reproduction is permitted
the western region. Two key mechanisms driving the enhancement of corporate ESG
which does not comply with these terms.
performance due to SCP are the promotion of green innovation investment and the
improvement of internal control quality.
Discussion: This paper provides valuable insights for policymakers and business
leaders in China and other emerging economies. It guides them in fortifying ESG
performance, thereby facilitating sustainable corporate growth.
KEYWORDS
smart city pilot policy, environmental, social, and governance performance, green
innovation, internal control, difference in differences
1 Introduction
China is currently undergoing a pivotal economic transformation as its rapid
industrialization has led to high economic growth (He et al., 2022). This industrial
boom, however, has propelled the nation to the position of the world’s largest energy
consumer and carbon dioxide emitter, with its downsides being evident in escalating
pollution, increased energy consumption, and carbon emissions (Hu and Zheng, 2021).
Specifically, the manufacturing industry, which is notably energy-intensive and pollution-
prone, accounts for approximately 60% of China’s total energy consumption and over half
of its carbon emissions (Xu and Lin, 2016). Recognizing these challenges, the China
Frontiers in Environmental Science 01 frontiersin.org

Tang et al. 10.3389/fenvs.2023.1305539
government has intensified its focus on the environment, social, program on ESG performance. While numerous studies have
and governance (ESG) practices of manufacturing firms (Ge et al., analyzed the factors affecting ESG engagement or performance
2022), because these practices are pivotal at enhancing from diverse perspectives, (Gillan et al., 2021), few, like the study
productivity and mitigating environment pollution in the sector of Sun and Saat (2023), delve into the impact of intelligent
(Gao et al., 2022). For instance, in September 2018 the China manufacturing on ESG at the enterprise level. Recognizing the
Securities Regulatory Commission (CSRC) updated the Code of pivotal role of governments in fostering smart city development
Corporate Governance for listed Companies, laying down in developing nations (Tan and Taeihagh, 2020), our study
foundational guidelines for ESG practice disclosures. This introduces a novel perspective to the ESG literature, investigating
commitment was further underlined in 2021 when CSRC the causal relationship between a government-initiated smart city
refined and standardized ESG information disclosure pilot and ESG outcomes.
requirements for these firms. Such moves underscore a growing Second, our research enriches the smart cities’ discourse by
emphasis within the China market on ESG-related information of examining the impact on ESG performance at the micro-firm level.
listed entities (Y. Yan et al., 2023). It is clear that government While numerous studies have begun exploring the implications of
authorities are increasingly sensitive to corporate ESG smart city policies on sustainability, their focus largely remains on
performance in order to fulfil key green economy policy goals. broader themes like innovation (Caragliu and Del Bo, 2019; Wang
The literature has analyzed the factors influencing ESG and Deng, 2022), environmental (Li et al., 2020; Chu et al., 2021;
performance from various perspectives. Aspects examined include Feng and Hu, 2022; Xu and Yang, 2022), carbon emissions (Cavada
ownership structure (McGuinness et al., 2017; Wang et al., 2023), et al., 2015; Yigitcanlar and Kamruzzaman, 2018; Guo et al., 2022),
board structure (Menicucci and Paolucci, 2022), leadership and green total factor productivity (Jiang et al., 2021; Dong et al.,
characteristics, (Gillan et al., 2021), pledges of controlling 2022; Wang et al., 2022), among others. Notably, most studies target
shareholders (Huang et al., 2022), M&A activity (Barros et al., macro-level affects, overlooking micro-firm implications. For
2022), CEO payment (Ikram et al., 2019), CEO confidence instance, while Chen (2022) assesses the influence of smart city
(McCarthy et al., 2017), CSR committee (Baraibar-Diez and pilot policies on a firm’s total productivity, our paper extends this
Odriozola, 2019), executive incentive (Jang et al., 2022) etc. exploration to firm-level ESG performance. We posit that SCP can
Nevertheless, there is a noticeable gap in examining the significantly enhance manufacturing firms’ ESG outcomes, which
relationship between smart cities and ESG performance (Barykin are essential for addressing challenges brought about by swift
et al., 2023). The scant research available primarily emphasizes the industrialization.
influence of businesses’ spontaneous investments in smart Third, we delve into the heterogeneous effects of SCP. Our
technology, driven by market competition pressures, on ESG findings suggest that non-state-owned enterprises (non-SOEs)
(Sun and Saat, 2023). Many of these studies tend to neglect the derive greater benefits from SCP than SOEs. This aligns with the
pivotal role governments play in advancing smart cities. observations of Sun and Saat (2023). Given that non-SOEs typically
In developed countries the rise of smart cities stems from both confront stiffer financing constraints (Su et al., 2022) and heightened
technological advancements and urban demands for efficiency and competition (Liu et al., 2022), they benefit more from the
sustainability (Angelidou, 2015). Conversely, smart cities in infrastructure set forth by smart city initiatives, reflected through
developing countries encounter distinct challenges like budget enhanced ESG performance. Furthermore, we discern that SCP’s
constraints, poor technology-related infrastructure, and lack of influence is more pronounced in east and central regions than in the
skilled human capital. In these nations, governments play a west region. In contexts where cities, especially in developing
crucial role of addressing infrastructure needs, generating nations, exhibit both resource scarcity and limited proclivity
revenue, setting regulatory frameworks to counter technological towards sustainability, smart city pilots might not notably bolster
risks, nurturing human capital, ensuring digital inclusivity, and corporate ESG outcomes.
upholding environmental sustainability. For instance, Fourth and finally, we pinpoint two pivotal pathways through
governments in countries like China and India have embarked which SCP impacts ESG performance in manufacturing firms:
on smart city initiatives (Tan and Taeihagh, 2020). However, corporate internal control and green innovation. Echoing the
over a decade later, pressing concerns linger: Have these findings of Chen (2022), we observe that manufacturing firms
government-led initiatives positively impacted manufacturing proactively leverage smart city pilot strategies to bolster their
firms’ ESG performance and fostered sustainable development? ESG metrics via green innovation. Additionally, the integration
To explore this topic, China’s phased smart city pilot program of advanced artificial intelligence (AI) and digital tools through
offers an insightful context. Our analysis draws from data collated SCP initiatives fortifies internal governance and risk management in
from the China Stock Market and Accounting Research Database listed manufacturing corporations, leading to improved
(CSMAR), the Huazheng ESG rating index (HESG), and the City ESG outcomes.
Statistics Yearbook, by considering 2,229 listed manufacturing firms The structure of this paper is organized as follows. Section 2 outlines
from 2009 to 2020. Utilizing both the difference in differences (DID) the policy background and delves into the underlying mechanisms.
and the propensity score matching-difference in differences (PSM- Section 3 details our research design, encompassing data sources,
DID) methodologies, we gauge the influence of the smart city variables, and model specification. Section 4 offers the empirical
initiative on firms’ ratings. The preliminary findings lean towards results along with robustness tests for validation. Section 5 examines
a beneficial impact. heterogeneity. Sections identifies key channel affecting the outcomes.
This paper offers three key additions to the exiting literature. Section 7 and Section 8 conclude the study and offer research
First, our work explores the influence of the smart city pilot (SCP) implications, respectively.

Tang et al. 10.3389/fenvs.2023.1305539
2 Policy background and prioritizing the use of clean energy and development of green
mechanism analysis technology (Vázquez et al., 2018; Chien et al., 2022). Research
has indicated that smart city development can bolster green total
2.1 Smart city policy in China factor productivity (Jiang et al., 2021; Wang et al., 2022), reduce
pollution (Li et al., 2020; Chu et al., 2021; Feng and Hu, 2022), and
Since the inception of the concept of smart cities by IBM in 2008, carbon emissions (Cavada et al., 2015; Yigitcanlar and
many developed countries in Europe and the United States have Kamruzzaman, 2018; Guo et al., 2022).
pioneered their respective smart city policies. China, however, As for social responsibility, smart city initiatives can significantly
entered the smart city development arena somewhat later. This influence stakeholders, including shareholders, employees, and
late onset came amidst China’s rapid economic expansion, which consumers. Smart cities utilize digital technology infrastructure to
unfortunately led to a plethora of environmental issues (Xu improve individuals’ lives and increase organizational productivity
et al., 2019). and competitiveness (Lima, 2020). This infrastructure allows
In response to these environmental challenges, China has companies to transition from a product-centric approach to one
initiated a variety of measures, including environment regulation, that prioritizes user experience (Yong, 2020). As a result, firms
carbon trading, carbon taxing (Zhao et al., 2017; Shi, 2018; Chang operating within smart cities can enhance economic efficiency and
and Han, 2020; Song et al., 2020) etc. One crucial component of workers’ rights and interests, including better working conditions
these actions was the construction of smart cities, which is an (López-Arranz, 2019) and improved labor remuneration and labor
integral part of China’s new urbanization drive. The China protection (Qi et al., 2020).
government, since 2010, has continuously launched relevant On the corporate governance front, smart cities improve access
policies to guide and encourage the construction of smart cities, to information and enrich both its depth and breadth. Digital
all aiming towards the principle of “no disease, less disease, and technologies used in smart cities facilitate information exchange
quick treatment” in urban areas. (Gupta and Bose, 2022) and knowledge sharing, (Lin et al., 2002),
In late 2012, China’s Ministry of Housing and Urban-Rural leading to more rational decision-making (Qi et al., 2020). Smart
Development officially announced the smart city pilot initiative, cities provide an efficient platform for firms to communicate
which covered 90 places: 37 prefecture-level cities, 50 districts dynamic changes to stakeholders, helping to minimize
(counties), and 3 towns. The initiative dictated a 3–5-year information asymmetry (Chen et al., 2014). Based on the
establishment period for each pilot city. During this period, the considerations, this paper proposes the first hypothesis.
China development Bank partnered with local governments to
invest in intelligent infrastructure, such as cloud computing, 5G Hypothesis 1. Smart city construction can increase manufacturing
network, high-efficiency online platform, efficient organization, etc., firms’ ESG performance.
to realize the intelligence of city management (Yang, 2018). In
2013 the government expanded the initiative with the second batch
of smart city pilots, which included an additional 103 cities, 2.3 Mechanism of green innovation
including 83 at the prefecture level and 20 at the county and
town level (MOHURD, 2013). Green innovation, defined as technological activities related to
By 2015 the third batch list of smart city pilots was rolled out, green processes or products (Chen et al., 2006), serves as a pivotal
increasing the total to 290. This expansion effectively accelerated driver for green transformation within enterprises (Huang et al.,
China’s new urbanization efforts. For these smart city initiatives, the 2021). This paper posits that the construction of smart cities will
central government has developed a three-level indicator system to bolster ESG performance by fostering green innovation in
access the construction achievement of the pilot cities. These manufacturing enterprise.
indicators include network infrastructure, intelligent management First, smart city construction can exert a positive influence on
and services, and industrial upgrading and new industries manufacturing enterprise green innovation. By offering a reliable
(MOHURD, 2012). This system introduced both incentives and public digital infrastructure, smart cities pave the way for green
pressure on local governments to meet evaluation benchmarks, innovation (Filiou et al., 2023). Within these smart cities, the
resulting in diverse policies and actions compared to non-pilot cities. embedded AI and Internet of Things (IoT) technology
infrastructure facilitates technology-intensive industries and helps
develop high-tech sectors (Qian et al., 2023). The growth of these
2.2 Smart city and enterprise ESG sectors stimulates agglomeration of innovation elements, attracts
performance high-end talent, expedites the transformation of scientific and
technological achievements, encourages economic entities to
Smart city indicatives have far-reaching implications for various invest in research and development, and enhances manufacturing
aspects of manufacturing firms, including environmental, enterprises’ technological innovation abilities. Moreover, smart city
responsibility, and governance (ESG). From an environmental construction seeks harmony between humanity and the natural
standpoint, smart cities apply and integrate a vast array of new environment, implying an emphasis on clean energy usage and
digital technologies. These technologies play a pivotal role in energy green technology development (Vázquez et al., 2018; Chien et al.,
and environmental conservation, directly impacting energy usage 2022). This focus not only fine-tunes green technologies (Hao et al.,
and pollutant emissions. Moreover, smart cities aim to foster 2023), but also speeds up the incorporation of energy-efficient and
harmony between humanity and the natural environment, eco-friendly technologies in the manufacturing sector (Witkowski,

Tang et al. 10.3389/fenvs.2023.1305539
2017). In essence, smart city initiatives create a conducive Internal control is critical for integrating ESG (Harasheh and
environment for green innovation selection (Gibbs and O’Neill, Provasi, 2023). Koo and Ki (2020) argue that firms with weak
2018) and elevate the quantity and quality of urban green innovation internal controls tend to have low ESG ratings, suggesting that
(Qiu, 2022). robust internal control bolsters corporate sustainability. Effective
Second, green innovation significantly enhances manufacturing internal control enhances corporate environmental protection
enterprises’ ESG performance. As green innovation activities investment (Yang et al., 2020), ensures law compliance (Le et al.,
increase, the sophistication of green technology improves, and 2022), and increases green levels (Li and Shen, 2021).
the cost of green production decreases. This dynamic enables Additionally, internal control effectively detects fraud risks
companies to more effectively assume environmental that curtail improper behaviors damaging corporate reputation
responsibility (Peattie and Ratnayaka, 1992) and boost their ESG (Hao et al., 2018) and helps encourage managers to make
performance. Moreover, a commitment to fostering green shareholder-friendly decisions and to proactively fulfill social
innovation indicates a company’s willingness and propensity to responsibilities (Liu, 2018). By improving the supervision and
assume social responsibility and to prioritize environmental incentive system for managers, internal control can mitigate
protection, which in turn enhance its ESG performance. agency conflicts and curtail self-serving behaviors by managers
Concurrently, through green innovation activities, companies can (Abbott et al., 2007). Consequently, effective internal control
project a positive image to external stakeholders, establishing a cultivates a conducive business environment for an enterprise
favorable corporate green image (Xie et al., 2019). This positive and ensures long-term financial forecasts and decisions, thereby
image helps attract more investor attention and capital support, enhancing their environmental, social, and governance
thereby providing resources to implement ESG principles. From performance and aiding in achieving sustainable development
these premises, we propose the next hypothesis. (Yang et al., 2020). Given the above, we propose the following
hypothesis.
Hypothesis 2. Smart city initiatives further enhance
manufacturing enterprises’ ESG performance by promoting green Hypothesis 3. Smart city initiatives can boost manufacturing
innovation. enterprises’ ESG performance by improving the level of internal
control within the pilot area enterprises.
2.4 Mechanism of corporate internet

control levels 3 Research design
The concept of internal control pertains to various control 3.1 Sample design
activities aimed at achieving a company’s management objectives
and gauging the level of internal governance and risk control within This paper considers 2,229 Chinese A-share listed
listed companies (Zhong et al., 2022). First, the construction of manufacturing firms from 2009 to 2020 as the research sample.
smart cities applies emerging information technologies to spur The selection of listed manufacturing firms is primarily due to their
organizational structure and enhance the level of internal comprehensive disclosure of key research content such as “ESG
governance (Acemoglu et al., 2007). Specifically, smart city performance.” These firms provide reliable data through trusted
construction embeds crucial information and personnel into the databases, making them ideal for this investigation. The ESG rating
information management system, facilitating a shift from vertical index of Chinese listed firms dates back to 2009, with ESG data
management towards a flatter, information-centric, networked obtained from the Wind database.
management. This transition strengthens interdepartmental Additional micro-data related to the corporations come from
relationships and improves information transmission efficiency the CSMAR database, which is the most comprehensive and
within the enterprise. Furthermore, by utilizing visual data accurate financial and economic database in China. It includes
workflows, IT governance, and data mining, smart city fundamental information on governance structure, financial
construction significantly improves the information transparency status, and operational conditions of listed companies. Macro-
of business decisions (Solana-González et al., 2021), thereby data related to cities are sourced from the China City
enhancing the quality of supervision and ensuring efficient Statistical Yearbook.
management. To ensure the integrity and accuracy of the research findings,
Second, smart cities leverage technologies like big data and the following data processing steps have been taken. 1)
artificial intelligence to digitize, network, and intelligently manage Companies in the ST and *ST categories, denoting firms with
risk. This approach mitigates one-sided information from scattered financial issues or other abnormalities that could potentially skew
data and enhances data processing efficiency and security. In terms the research conclusion, are excluded from the sample. 2) Any
of risk model construction, multidimensional data analysis is sample with missing relevant data are also excluded. 3) The
performed on business positions and processes, allowing continuous financial variables are winsorized at the 1% and
differentiated supervision and responses in various scenarios 99% quartiles to mitigate the effect of extreme values. After
based on big data analysis results. Ultimately, enterprise risk this thorough screening, the paper ends up with 13,923 firm-
management and internal control evolve from fragmented to year sample observations. Data processing and regression
systematic and from uniform to diverse (Fan, 2022). analyses are carried out using STATA17.0.

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 1 Variable definitions.
Symbol Variable Measurement

HESG ESG performance According to Huazheng ESG rating, the assigned value is 1–9 from low to high
SCP Smart city pilot A city approved as a smart city gets a treatment variable (SCP) of 1, while it gets 0 if it has never been
approved as a smart city
IC Internal control effectiveness The effectiveness of internal control is measured by checking for deficiencies in a company’s
evaluation report. If none are present, the variable is set to 1 and otherwise 0
Gpatent Green innovation Natural logarithm of total green invention patents granted to a firm plus one
Size Enterprise size Natural logarithm of the total assets of the enterprise
Lev Financial leverage Total financial liabilities/total assets
Roa Return on total assets Corporate net profit/total assets
Pcontrol Proportion of control rights owned by the actual Control rights held by actual controller/total control rights available in the company
controller
Top1 Percentage of shares held by the largest shareholder Number of shares held by the largest shareholder as a percentage of the total number of the firm’s
shares
TobinQ Tobin’s Q value Stock market value of total assets
Age Age since listing Age since listing is calculated by subtracting the year of listing from the current year
lnpgdp Per capita GDP Natural logarithm of per capita GDP
rgdp GDP growth rate GDP growth rate
lnso2 Industrial sulfur dioxide emissions Natural logarithm of industrial sulfur dioxide emissions
lnpm10 Annual average concentrations of inhalable particulate Natural logarithm of annual average concentrations of inhalable particulate matter
matter (mg/m3)
lnno Industrial nitrogen oxide emissions Natural logarithm of industrial nitrogen oxide emissions
lnsewage Centralized treatment rate of sewage treatment plants Natural logarithm of the centralized treatment rate of sewage treatment plants
lnwaste Harmless treatment rate of domestic waste Natural logarithm of the harmless treatment rate of domestic waste
3.2 Variable definitions internal control deficiencies from the listed companies’
internal control evolution reports. In this context, a lack of
3.2.1 Dependent variable: ESG score (ESG) deficiencies indicates effective internal controls and gives this
Our dependent variable is corporate ESG performance (ESG). The variable a value of 1. Conversely, the presence of deficiencies set
ESG rating index is a recognized tool for the quantitative assessment of its value to 0.
firms’ ESG performance. We employ the Huazheng ESG rating index Green innovation (Gpatent): Green innovation is quantified as
(HESG) to gauge this performance, as it encompasses A-share listed the natural logarithm of total green invention patents granted to a
manufacturing firms over an extended timeframe. HESG is comprised firm plus one.
of 3 primary indicators, 14 secondary indicators, 26 tertiary indicators,
and over 130 underlying indicators. For robustness testing, we also use 3.2.4 Control variables
the SynTao Green Finance Agency ESG evaluation index (SESG) and Based on the literature, we select the control variables at two
the Bloomberg ESG evaluation index (BESG), but these indices have levels: firm and city. Data on firm characteristics are taken from
limitations. SESG was first announced in 2015, while Bloomberg covers China Stock Market and Accounting Research Database
about 1,000 Chinese listed firms. Compared to HESG, both SESG and (CSMAR) and encompass the following: 1) enterprise size; 2)
BESG exhibit more data gaps. financial leverage; 3) return on total assets; 4) proportion of
control rights owned by the actual controller; 5) percentage of
3.2.2 Independent variables: smart city pilot (SCP) shares held by the largest shareholder; 6) Tobin’s Q value; and 7)
The treatment variable, smart city pilot (SCP), is a binary age since listing. Data on city characteristics are from the China
variable reflecting the city of a firm’s registration. If a city is City Statistical Yearbook (2009–2019) and include: 8) per capita
approved as a smart city, then its SCP takes a value of 1 and GDP; 9) GDP growth rate; 10) industrial sulfur dioxide
otherwise 0. emissions; 11) annual average concentrations of inhalable
particulate matter (mg/m3); 12) industrial nitrogen oxide
3.2.3 Mediating variables emissions; 13) centralized treatment rate of sewage treatment
Internal control effectiveness (IC): We determine the plants; and 14) harmless treatment rate of domestic waste. The
effectiveness of internal control using the disclosure of variables appear in Table 1.

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 2 Descriptive statistics for all variables. 13,923 observations with a mean value of 4.046 and a standard
deviation of 1.038. It ranges from 1 to 8, showing a moderate level of
Variable Obs Mean SD Min Max
variability in ESG performance among the firms in our sample. Our
HESG 13,923 4.046 1.038 1 8 key independent variable, SCP, has the same number of
SCP 13,923 0.330 0.470 0 1
observations, reflecting a mean of 0.330 with a standard
deviation of 0.470 and suggesting that roughly 33% of the
Size 13,923 22.052 1.166 19.816 26.297
manufacturing firms in our sample are located in smart cities.
Lev 13,923 0.462 0.223 0.025 0.914
Roa 13,923 0.033 0.064 −0.329 0.179

4.2 Baseline regression results
Pcontrol 13,923 37.290 15.706 4.150 76.220
Top1 13,923 34.139 14.004 8.735 74.995 Table 3 presents the regression results of SCP on ESG
performance. To account for heteroscedasticity, this paper
TobinQ 13,923 2.002 1.152 0.857 7.627
provides results employing robust standard errors. Firm and time
Age 13,923 9.278 6.811 0.000 26.000 effects are fixed across all columns. The initial regression equation
without control variables shows SCP with a significant coefficient of
lnpgdp 13,923 11.287 0.554 9.822 12.223
0.162 at the 1% level, implying a 0.162% increase in a firm’s ESG
rgdp 13,923 8.304 3.183 0.350 16.500 performance with every 1% rise in SCP policy.
lnso2 13,923 10.016 1.407 6.652 12.432 Subsequent models add firm characteristics (column 2), city
characteristics (column 3), and both firm and city characteristics
lnpm10 13,923 3.755 0.374 2.944 4.771
(column 4). Regardless of these added characteristics, SCP
lnno 13,923 9.899 0.934 7.476 11.851 significantly boosts the ESG of firms in pilot cities.
Column (5), applying city-level clustered robust standard errors,
lnsewage 13,923 4.499 0.099 4.086 4.605
reports an SCP regression coefficient of 0.159 at the 1% significance
lnwaste 13,923 4.496 0.461 1.520 4.605 level. This consideration acknowledges the unique situation of
Beijing, Tianjin, Shanghai, and Chongqing—municipalities
3.3 Model design directly under central government regulation with high political
status, population density, and economic development (Sun et al.,
China’s smart city pilot policy is akin to a quasi-natural 2021). When we exclude these four municipalities in column (6), the
experiment. Pilot cities function as the experimental group, SCP regression coefficient is 0.171, which is still significant at the 1%
while non-pilot cities serve as the control group. The level. These findings support Hypothesis 1.
difference in differences (DID) model is widely used in the
literature for policy effect evaluation. Given that China had
three batches of national smart city pilots in 2012, 2013, and 4.3 Robustness test
2015, we adopt a dynamic DID approach, as Beck et al. (2010)
and Wang et al. (2011) and set up the following econometric 4.3.1 Parallel trend test
regression model (see Eq. 1): The difference in differences (DID) model necessitates that both
the treatment and control groups fulfill the parallel trend
HESGi,t α + βSCPi,t + θXi,t + μi + ωt + εi,t (1) assumption. This means that there should be no significant
difference in ESG performance between these two groups before
Here, HESGi,t represents firm i’s ESG performance in year t.
the event under study. To ensure this, we structured our estimation
The key variable SCPi,t is a binary variable that takes on the
equation as follows (see Eq. 2):
value of 1 if firm I is based in a smart city and 0 otherwise. The
−2
associated coefficient β represents the net effect of SCP policy. 8
HESGit α0 + j−5 αj Di,t+j + j0 αj Di,t+j + βi Xit + μi + ωt + εi,t
Moreover, μi and ωt are vectors of city and year dummy
(2)
variables that account for firm and year fixed effects,
respectively, while εi,t is the error term. We use regression Here, D equals 1 if the city, in which the firm is located, initiated SCP
models with firm and year fixed effects and control variables in a particular year and 0 otherwise. Figure 1 illustrates the results of
to estimate policy effects, thereby accommodating unobserved the parallel trend test for the first 5 years and the last 8 years of a
time-invariant heterogeneity. city’s implementation. The regression coefficients from α−5 to α−2
are all insignificant, signifying no substantial difference in ESG
performance between the treatment and control groups before
4 Empirical results SCP implementation. This suggests that both groups comply with
the parallel trend assumption before the introduction of SCP.
4.1 Descriptive statistics
4.3.2 PSM-DID
Table 2 shows the results of descriptive statistics. HESG, our Firms may factor in economic and other attributes of their place
primary dependent variable indicating ESG performance, has of incorporation, potentially leading to selection bias. To address

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 3 Baseline regression results.
(1) (2) (3) (4) (5) (6)
Variable HESG HESG HESG HESG HESG HESG

SCP 0.162*** 0.165*** 0.154*** 0.159*** 0.159*** 0.171***
(0.034) (0.033) (0.034) (0.034) (0.051) (0.048)
Size 0.152*** 0.151*** 0.151*** 0.147***
(0.023) (0.023) (0.026) (0.032)
Lev −0.130** −0.128** −0.128* −0.161**
(0.062) (0.062) (0.073) (0.082)
Roa 1.336*** 1.332*** 1.332*** 1.323***
(0.169) (0.169) (0.206) (0.197)
Pcontrol −0.000 −0.000 −0.000 −0.000
(0.001) (0.001) (0.001) (0.001)
Top1 0.005*** 0.005*** 0.005** 0.006***
(0.002) (0.002) (0.002) (0.002)
TobinQ −0.035*** −0.034*** −0.034** −0.042***
(0.010) (0.010) (0.014) (0.013)
Age 0.182 0.180 0.180 −0.041
(0.140) (0.142) (0.195) (0.110)
lnpgdp 0.102* 0.083 0.083 0.106
(0.059) (0.058) (0.072) (0.078)
rgdp −0.005 −0.006 −0.006 −0.005
(0.005) (0.005) (0.006) (0.006)
lnso2 −0.043** −0.037** −0.037 −0.021
(0.019) (0.019) (0.024) (0.028)
lnpm10 −0.029 −0.024 −0.024 −0.007
(0.064) (0.064) (0.101) (0.094)
lnno −0.062* −0.043 −0.043 −0.064
(0.033) (0.033) (0.045) (0.046)
lnsewage 0.124 0.180 0.180 0.145
(0.142) (0.140) (0.228) (0.182)
lnwaste 0.016 0.012 0.012 −0.030
(0.038) (0.037) (0.051) (0.049)
Constant 3.992*** −1.151 3.413*** −1.972 −1.972 0.230
(0.013) (1.396) (0.997) (1.715) (2.284) (1.807)
Firm FE Yes Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes Yes
City cluster No No No No Yes No
Adjusted R2 0.457 0.468 0.458 0.469 0.469 0.465
Observations 13,923 13,923 13,923 13,923 13,923 12,068
Notes: (1) *** indicates significance at p < 0.01, ** indicates significance at p < 0.05, and * indicates significance at the p < 0.1. (2) Robust standard errors are in parentheses.

Tang et al. 10.3389/fenvs.2023.1305539
FIGURE 1
Parallel trend test of the impact of SCP on HESG. Note: The black dash line shows the confidence interval for the regression coefficients at the 5%
significance level. The gray solid line represents the 0 scale line on the vertical axis, indicating that the regression coefficient is 0. The gray dotted line is the
0 scale line on the horizontal axis, indicating the year in which the policy is implemented.
FIGURE 2
Kernel density test: (A) Before year-by-year and (B) After year-by-year matching.
this, we utilize propensity score matching (PSM) as a strategy to samples obtained using PSM and ensures the reliability of the
mitigate such bias (Dehejia and Wahba, 2002). Following the paper’s estimation results.
methodologies of Heyman et al. (2007) and Böckerman and In this study we select three types of samples to evaluate the
Ilmakunnas (2009), our study conducts yearly propensity score impact: those with non-zero matching weights (_weight! = .), those
matching. This approach emphasizes yearly matching, primarily satisfying common support (On_Support), and those undergoing
within the current year, preventing mismatches such as pairing the frequency-weighted regression (Weight_Reg). Table 4 presents the
2009 sample with the 2020 sample. new regression results derived from these newly obtained samples.
Figures 2A, B represent the kernel density function plots before The regression coefficients of SCP are significantly positive with no
and after matching, respectively. By comparing these figures, it is considerable changes in their magnitude and significance level,
evident that the treatment and control groups are considerably suggesting credibility of the regression outcomes.
more aligned after one-to-two matching. The reduced distance This study decides to select different samples to test the impact
between the groups affirms this. It demonstrates the validity of the effect. These samples are of three types: those matching weights not

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 4 Impact of SCP on ESG within PSM-DID
(1) (2) (3)
Variable HESG HESG HESG

SCP 0.089** 0.099*** 0.109**
(2.125) (2.870) (2.084)
Size 0.161*** 0.160*** 0.184***
(5.486) (6.809) (4.580)
Lev −0.119 −0.131** −0.052
(−1.473) (−2.093) (−0.453)
Roa 1.138*** 1.254*** 1.009***
(5.377) (7.337) (2.991)
Pcontrol −0.000 −0.000 −0.000
(−0.123) (−0.108) (−0.296)
Top1 0.005** 0.005*** 0.002
(2.513) (3.431) (0.707)
TobinQ −0.053*** −0.036*** −0.056***
(−3.764) (−3.408) (−3.048)
Age 0.282 0.176 0.245
(1.200) (1.246) (1.170)
lnpgdp 0.023 0.062 0.059
(0.301) (1.063) (0.629)
rgdp −0.001 −0.007 −0.003
(−0.167) (−1.275) (−0.345)
lnso2 −0.011 −0.028 −0.008
(−0.430) (−1.451) (−0.254)
lnpm10 0.011 −0.009 0.023
(0.128) (−0.146) (0.226)
lnno −0.026 −0.053 −0.039
(−0.658) (−1.628) (−0.797)
lnsewage 0.007 0.163 0.068
(0.042) (1.149) (0.308)
lnwaste −0.034 −0.006 0.004
(−0.653) (−0.150) (0.068)
Firm FE Yes Yes Yes
Year FE Yes Yes Yes
weight! = Yes No No
On_support No Yes No
Weight_Reg No No Yes
Adjusted R2 0.457 0.465 0.527
Observations 8,711 13,532 8,711
Notes: (1) *** indicates significance at p < 0.01, ** indicates significance at p < 0.05, and * indicates significance at p < 0.1. (2) Robust standard errors are in parentheses.

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 5 Placebo test using the fake treatment time.
(1) (2) (3) (4) (5)
Variable HESG HESG HESG HESG HESG

Advance_1 0.058
(0.054)
Advance_2 0.022
(0.060)
Advance_3 −0.030
(0.077)
Advance_4 0.080
(0.120)
Advance_5 0.088
(0.141)
Size 0.148*** 0.147*** 0.147*** 0.148*** 0.148***
(0.026) (0.026) (0.026) (0.026) (0.026)
Lev −0.103 −0.103 −0.102 −0.104 −0.103
(0.074) (0.074) (0.074) (0.074) (0.074)
Roa 1.469*** 1.469*** 1.468*** 1.468*** 1.468***
(0.212) (0.212) (0.212) (0.212) (0.212)
Pcontrol −0.000 −0.000 −0.000 −0.000 −0.000
(0.001) (0.001) (0.001) (0.001) (0.001)
Top1 0.005** 0.005** 0.005** 0.005** 0.005**
(0.002) (0.002) (0.002) (0.002) (0.002)
TobinQ −0.040*** −0.040*** −0.040*** −0.040*** −0.040***
(0.015) (0.015) (0.015) (0.015) (0.015)
Age 0.144 0.144 0.144 0.144 0.144
(0.193) (0.192) (0.192) (0.192) (0.192)
lnpgdp 0.068 0.071 0.075 0.070 0.069
(0.075) (0.075) (0.075) (0.075) (0.075)
rgdp −0.012* −0.012* −0.012* −0.012* −0.012*
(0.007) (0.007) (0.007) (0.007) (0.007)
lnso2 −0.027 −0.027 −0.026 −0.027 −0.026
(0.026) (0.026) (0.026) (0.026) (0.026)
lnpm10 0.053 0.058 0.060 0.058 0.058
(0.101) (0.101) (0.101) (0.101) (0.101)
lnno −0.052 −0.053 −0.054 −0.055 −0.055
(0.046) (0.046) (0.046) (0.046) (0.046)
lnsewage 0.223 0.221 0.218 0.224 0.225
(0.217) (0.220) (0.221) (0.220) (0.219)
(Continued on following page)

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 5 (Continued) Placebo test using the fake treatment time.
(1) (2) (3) (4) (5)

lnwaste −0.026 −0.027 −0.027 −0.027 −0.027
(0.042) (0.042) (0.043) (0.042) (0.042)
Constant −1.615 −1.630 −1.645 −1.646 −1.660
(2.239) (2.231) (2.238) (2.223) (2.225)
Firm FE Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes
Adjusted R2 0.466 0.466 0.466 0.466 0.466
Observations 13,923 13,923 13,923 13,923 13,923
For the second placebo test, we create a fictitious group. We

choose an imaginary treatment group, containing the same number
of cities as the original treatment group. This procedure was
repeated 1,000 times. Figure 3 illustrates the kernel density plot
of the SCP coefficients for these 1,000 repetitions, with the estimates
after random sampling primarily hovering around 0. This infers that
the remaining omitted coefficients are 0, suggesting that the
regression outcomes are not swayed by unobserved variables, and
that the influence of SCP on ESG performance is notably robust.
4.3.4 Alternative measures of ESG performance

To confirm the robustness of our findings, we undertake a series
of supplementary tests. Initially, we replace year-end data with the
quarterly average of Huazheng’s ESG scores (QHESG). The
outcome, presented in column (1) of Table 6, shows a
FIGURE 3 significantly positive regression coefficient for SCP at the 1% level.
Kernel density plot of the placebo test coefficients (the
Apart from the Huazheng database, various authoritative
virtual treatment).
rating agencies or financial information providers also assess
enterprises’ ESG performance, effectively promoting sustainable
equal to zero, those that meet common support (On_Support), and behaviors (Chen and Xie, 2022). In this context, we substitute the
those that undergo frequency-weighted regression (Weight_Reg). baseline regression model’s Huazheng ESG scoring data with
Using these new samples obtained after matching, the new SynTao Green Finance Agency (SESG) and the ESG scoring
regression results appear in Table 4. The regression coefficients data of BloomBerg (BESG) respectively to perform robustness
of SCP are significantly positive, and there are no significant changes checks. The control variables and fixed effects align with the
in the magnitude and significant level of these regression baseline regression model. The regression outcomes for SynTao
coefficients, indicating the regression results are credible. Green Finance Agency’s ESG rating data and BloomBerg’s ESG
score data are presented in columns (2) and (3) of Table 6. Here
4.3.3 Placebo test again, we observe a positive and statistically significant impact of
We execute two distinct placebo tests to eliminate the effect of SCP on manufacturing firms’ ESG performance.
any unobserved variable on the regression results of SCP and to
enhance the robustness of the estimates (Dong et al., 2022). In the 4.3.5 High dimension fixed effect
first placebo test, we simulate a scenario where SCP was initiated In order to account for the time-invariant heterogeneity across
during the pre-reform period. We postulate that the smart city provinces or cities, as well as to buffer any province-specific
pilot was implemented 1–5 years ahead of the actual schedule. economic and political fluctuations (for example, those triggered
The results are displayed in Table 5. The coefficients are not by changes in provincial leadership or the central government’s
significant. This implies that SCP only influences the ESG regional policies), we incorporate the covariates of province fixed
performance of manufacturing firms in smart cities following effects (FE), Prov ×Year fixed effects (FE), and City fixed effects (FE)
the actual policy implementation. Put simply, the ESG effect as shown in Table 7. The primary results are consistent—that SCP
associated with the smart city pilot holds significance. significantly boosts the ESG performance of firms.

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 6 Alternative measures of ESG.
(1) (2) (3)
Variable QHESG SESG BESG

SCP 0.149*** 0.553*** 0.785**
(0.029) (0.166) (0.389)
Size 0.135*** 0.223 1.252***
(0.020) (0.156) (0.215)
Lev −0.142*** −0.006 −0.290
(0.053) (0.296) (0.588)
Roa 1.113*** 0.453 2.692**
(0.143) (0.469) (1.185)
Pcontrol 0.001 −0.001 −0.018*
(0.001) (0.003) (0.009)
Top1 0.004*** −0.001 0.050***
(0.001) (0.007) (0.016)
TobinQ −0.027*** 0.041 0.325***
(0.009) (0.036) (0.092)
Age 0.154 0.310* 0.840
(0.137) (0.159) (0.811)
lnpgdp 0.085* 0.505** −0.664
(0.049) (0.251) (0.478)
rgdp −0.003 −0.013 0.070
(0.004) (0.021) (0.043)
lnso2 −0.026 0.136** −0.151
(0.016) (0.063) (0.157)
lnpm10 −0.007 0.304 −0.497
(0.055) (0.294) (0.625)
lnno −0.041 0.022 −0.024
(0.029) (0.113) (0.319)
lnsewage 0.232* 0.015 −4.903***
(0.121) (0.685) (1.424)
lnwaste −0.008 −0.050 −0.177
(0.035) (0.075) (0.415)
Constant −1.757 −13.854** 19.287
(1.567) (5.806) (13.784)
Firm FE Yes Yes Yes
Year FE Yes Yes Yes
Adjusted R2 0.548 0.671 0.811
Observations 13,923 1,175 4,202

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 7 Robustness test of high dimension FE.
(1) (2) (3)
Variable HESG HESG HESG
SCP 0.156*** 0.169*** 0.152***
(0.034) (0.041) (0.034)
Size 0.152*** 0.158*** 0.151***
(0.023) (0.023) (0.023)
Lev −0.137** −0.104* −0.141**
(0.062) (0.062) (0.062)
Roa 1.307*** 1.237*** 1.302***
(0.170) (0.173) (0.172)
Pcontrol 0.000 0.000 0.000
(0.001) (0.001) (0.001)
Top1 0.005*** 0.006*** 0.005***
(0.002) (0.002) (0.002)
TobinQ −0.037*** −0.034*** −0.037***
(0.010) (0.011) (0.010)
Age 0.187 0.213 0.189
(0.142) (0.148) (0.142)
lnpgdp 0.054 0.090 0.060
(0.061) (0.088) (0.066)
rgdp −0.004 0.006 −0.005
(0.005) (0.008) (0.005)
lnso2 −0.034* −0.047* −0.031
(0.019) (0.027) (0.019)
lnpm10 −0.017 −0.072 −0.013
(0.064) (0.085) (0.065)
lnno2 −0.040 −0.080* −0.038
(0.034) (0.048) (0.036)
lnwater 0.153 0.096 0.148
(0.141) (0.163) (0.144)
lnwaste 0.013 −0.008 0.015
(0.037) (0.059) (0.038)
Constant −1.677 −1.526 −1.804
(1.733) (1.962) (1.780)
Firm FE Yes Yes Yes
Year FE Yes Yes Yes
Province FE Yes Yes No
Prov×Year FE No Yes No
City FE No No Yes
2
Adjusted R 0.471 0.474 0.472
Observations 13923 13923 13923
Note: (1) *** indicates significance at p < 0.01, ** indicates significance at p < 0.05, and * indicates significance at p < 0.1. (2) Robust standard errors are in parentheses.

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 8 Regression results based on firm heterogeneity.
SOE NOE East Central West
(1) (2) (3) (4) (5)

SCP 0.072 0.226*** 0.152*** 0.141** 0.069
(0.048) (0.048) (0.045) (0.072) (0.082)
Size 0.235*** 0.194*** 0.127*** 0.217*** 0.201***
(0.038) (0.032) (0.028) (0.053) (0.066)
Lev 0.013 −0.166** −0.139* −0.209 0.165
(0.107) (0.075) (0.073) (0.147) (0.184)
Roa 0.107 1.418*** 1.454*** 1.023** 0.981**
(0.275) (0.210) (0.208) (0.400) (0.444)
Pcontrol −0.007** 0.001 −0.000 −0.000 0.001
(0.003) (0.001) (0.001) (0.002) (0.003)
Top1 0.001 0.007*** 0.007*** 0.004 0.000
(0.003) (0.002) (0.002) (0.003) (0.004)
TobinQ 0.006 −0.025* −0.045*** −0.044* 0.038
(0.017) (0.013) (0.013) (0.023) (0.030)
Age 0.497*** −0.056 0.145 0.485** 0.385
(0.160) (0.183) (0.162) (0.220) (0.893)
lnpgdp 0.014 0.047 −0.038 0.094 0.059
(0.087) (0.078) (0.083) (0.141) (0.177)
rgdp −0.007 0.000 0.003 −0.010 0.006
(0.008) (0.007) (0.007) (0.012) (0.012)
lnso2 −0.051* 0.001 −0.046** 0.051 −0.164***
(0.030) (0.025) (0.023) (0.050) (0.054)
lnpm10 0.034 −0.033 0.064 −0.375*** 0.316
(0.101) (0.086) (0.079) (0.137) (0.201)
lnno 0.055 −0.107** −0.039 −0.190** 0.027
(0.050) (0.043) (0.044) (0.088) (0.080)
lnsewage 0.449** 0.077 −0.064 −0.031 0.841***
(0.205) (0.190) (0.220) (0.272) (0.254)
lnwaste −0.074 0.025 −0.112 0.047 −0.064
(0.045) (0.059) (0.094) (0.068) (0.056)
Constant −9.883*** 0.104 1.723 −4.310 −9.207
(2.797) (1.935) (2.077) (3.559) (10.760)
Firm FE Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes
Adjusted R2 0.506 0.476 0.478 0.434 0.471
Observations 4,649 9,274 9,509 2,519 1895

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 9 Mechanism testing.
(1) (2) (3) (4)
IC HESG Gpatent HESG

SCP −0.048** 0.102** 0.053** 0.126***
(0.024) (0.051) (0.027) (0.045)
IC −0.076**
(0.033)
Gpatent 0.067***
(0.021)
Size −0.024* 0.160*** 0.049*** 0.140***
(0.014) (0.033) (0.018) (0.030)
Lev −0.056* −0.185** 0.075** −0.156**
(0.030) (0.083) (0.036) (0.077)
Roa −0.382*** 1.325*** −0.092 1.325***
(0.081) (0.202) (0.076) (0.189)
Pcontrol −0.000 0.000 0.000 −0.000
(0.000) (0.001) (0.001) (0.001)
Top1 0.001 0.007*** −0.001 0.006***
(0.001) (0.002) (0.001) (0.002)
TobinQ −0.017*** −0.044*** 0.001 −0.036***
(0.005) (0.013) (0.006) (0.012)
Age −0.042 0.154 0.116** 0.143
(0.043) (0.174) (0.059) (0.172)
lnpgdp 0.038 0.059 0.011 0.054
(0.035) (0.079) (0.040) (0.071)
rgdp −0.001 −0.005 0.000 −0.008
(0.003) (0.007) (0.003) (0.006)
lnso2 −0.031*** −0.039 0.027* −0.041*
(0.010) (0.025) (0.014) (0.024)
lnpm10 0.043 −0.032 −0.047 0.008
(0.039) (0.097) (0.045) (0.092)
lnno −0.016 −0.064 0.016 −0.054
(0.021) (0.049) (0.026) (0.045)
lnsewage −0.008 0.035 −0.150 0.157
(0.083) (0.203) (0.096) (0.186)
lnwaste 0.000 0.023 −0.035 −0.015
(0.021) (0.047) (0.024) (0.042)
Constant 1.125 −0.789 −1.372 −0.823
(0.753) (2.243) (1.025) (2.161)
Firm FE Yes Yes Yes Yes
(Continued on following page)

Tang et al. 10.3389/fenvs.2023.1305539
TABLE 9 (Continued) Mechanism testing.
(1) (2) (3) (4)
IC HESG Gpatent HESG

Year FE Yes Yes Yes Yes
Adjusted R2 0.499 0.468 0.655 0.463
Observations 12,135 12,135 13,321 13,321
Notes: (1)*** indicates significance at p < 0.01, ** indicates significance at p < 0.05, and * indicates significance at p < 0.1. (2) Robust standard errors are in parentheses.
5 Heterogeneity test analysis a mediation effect model to verify the mechanism of the smart city pilots
on ESG. This model can be segmented into three stages. The first stage
We begin by evaluating the heterogeneity effect between enterprises (see Eq. 3) is the same as Eq. 1. In the second stage, we assess whether the
of different ownerships. State-owned enterprises (SOEs) and non-state- smart city pilots affect internal control or green innovation (see Eq. 4). In
owned enterprises (non-SOEs), operating in distinct business the third stage, we incorporate the internal control or green innovation
environments and using varied decision-making mechanisms in into the dynamic DID model and verify their statistical significance (see
China, present interesting contrasts (Dong et al., 2022). Table 8 Eq. 5):
columns (1) and (2) display the results of our ownership-focused
analysis. Our data reveal that SCP do not affect the ESG performance HESGi,t α1 + β11 SCPi,t + θ1 Xi,t + μi + ωt + εi,t (3)
of SOEs, but they do enhance the ESG performance of non-SOEs. This ICi,t or Gpatenti,t α + 2
β21 SCPi,t + θ Xi,t + μi + ωt + εi,t
2
(4)
conclusion aligns with Sun’s research (Sun and Saat, 2023), suggesting
HESGi,t α + 3
β31 SCPi,t + β32 IC i,t 3
or Gpatent i,t + θ X i,t + μi + ωt
that non-SOEs have a larger potential for ESG performance
improvement than SOEs. Notably, the former, facing more intense + εi,t
financing constraints (Su et al., 2022) and marked competition (Liu (5)
et al., 2022) than the latter, benefit more from the infrastructure Table 9 displays the results of the mechanism test. The regression
developed by smart city pilots in terms of ESG performance coefficient of SCP in model (1) is significantly negative, while the
enhancement. regression coefficient of IC in model (2) is also significantly negative.
We next consider the influence of regional differences on enterprise This indicates that SCP could enhance corporate ESG performance by
ESG practices (Wang et al., 2023). China, following a development model mitigating corporate internal control deficiencies. On the other hand, the
that prioritizes efficiency since its reform and opening up, displays stark regression coefficient of SCP in model (3) is significantly positive, and the
regional disparities (Wang et al., 2023). These disparities translate to regression coefficient of Gpatent in model (4) is likewise significantly
notable differences in resources, industrial characteristics, and positive. This suggests that SCP could enhance corporate ESG
government policies across regions. (Liu et al., 2021; Wen et al., performance by fostering firms’ green patents.
2023). For a more comprehensive understanding of SCP’s influence
on enterprise ESG performance, we classify cities into three types: east
region (East), central region (Central), and west region (West). A separate 7 Conclusion
regression is conducted for each to understand SCP’s influence on
enterprise ESG performance more comprehensively. This paper considers the implementation of the smart city pilot
Table 8 columns (3)–(5) present results for the east, central, and policy as a quasi-natural experiment. Utilizing data from
west regions, respectively. SCP significantly boosts corporate ESG 2,229 manufacturing firms listed in China between 2009 and 2020,
performance in the east and central regions. However, in the west we investigate the potential of SCP to enhance corporate ESG
region, SCP does not notably influence corporate ESG performance. performance. We also delve into the mechanisms by which SCP
The east and central regions, boasting a larger concentration of labor impacts corporate ESG performance. Our study illuminates how SCP
and industry (Li et al., 2023), foster green innovation through increased serves as crucial infrastructure to augment corporate ESG performance in
labor input, information infrastructure, and fiscal expenditure (Z. Yan developing countries. The specific findings of the study are as follows.
et al., 2023). Conversely, the west region, characterized by First, SCP can notably bolster corporate ESG performance, which is a
underdeveloped technology and economic infrastructure, prioritizes result that remains robust after a series of tests. We find that, compared
economic growth (Qiu, 2022). Despite initiating smart city pilots, these with non-pilot cities, smart city construction amplifies corporate ESG
cities within this region show a lack of both resources and willingness to performance by 15.9%. This demonstrates the remarkable success of
promote sustainable development through improving enterprises’ ESG China’s smart city pilot policy. The effect persists at being significantly
performance. positive after the parallel trend test, PSM-DID, the placebo test, alternative
measures of ESG performance, and high dimension fixed effect inclusion.
Second, the SCP’s effects vary across different firms. Our
6 Mechanism testing heterogeneity test based on corporate ownership reveals that the
impact of SCP on improving ESG performance is restricted to
We proceed to examine the potential pathways and mechanism. non-SOEs. SOEs do not exhibit a significant impact of SCP on
Following the model proposed by Baron and Kenny (1986), we construct ESG performance. Compared to them, non-SOEs derive greater

Tang et al. 10.3389/fenvs.2023.1305539
benefit from the infrastructure developed by smart city pilots, as Third, the research also uncovers in the underdeveloped west region
evidenced by enhanced ESG performance. The heterogeneity test with poor infrastructure that ESG performance enhancement is
based on corporate location indicates that SCP significantly improves considerably less. As such, meeting basic infrastructure needs, securing
corporate ESG performance in the east and central region, but not in funding, and crafting regulatory frameworks are crucial during the smart
the west region. In cities lacking both resources and willingness to city construction phase. These steps will ensure cities have the requisite
promote sustainable development, smart city pilots have failed to resources and inclination to drive sustainable development.
improve enterprise ESG performance. Fourth, the results reveal that green innovation is a vital pathway to
Third and finally, SCP improves corporate ESG performance via elevate ESG performance among manufacturing firms. Hence,
two mechanisms: internal control and green innovation. SCP government regulations should facilitate knowledge protection and
diminishes corporate internal control deficiencies and augments the incentivize technological innovation. Doing so can promote renewable
volume of corporate green innovation, both of which subsequently energy usage and foster a conducive institutional for green innovation.
advance corporate ESG performance. Our exploration for the internal Fifth, business managers should not only recognize the merits of
mechanisms through which smart city construction enhances corporate smart city policies, but also actively synchronize their corporate
ESG performance contributes to understanding how to fully unlock the strategies with these policies to foster sustainable growth.
benefits of smart city construction. Additionally, they should concentrate on improving innovation
This study, in a theoretical context, delves into SCP’s influence on investment and internal control to leverage SCP’s potential in
manufacturing firms’ ESG performance and expounds on two enhancing ESG performance optimally.
mechanisms via which SCP strengthens corporate ESG performance.
However, the study is not without limitations. SCP’s influence on
corporate ESG performance extends beyond the two explored Data availability statement
channels, warranting future research to uncover additional channels
like information environment and resource allocation efficiency. The The original contributions presented in the study are included in
study’s focus also is confined to the heterogeneity of corporate ownership the article/Supplementary material, further inquiries can be directed
and location, calling for future research to analyze SCP’s impact on ESG to the corresponding author.
performance from more diverse perspectives. Furthermore, this study
treats corporate ESG performance as a unified entity. Future research may
consider the individual impact of SCP on the E, S, and G components. Author contributions
Lastly, policy creation and its impact are often influenced by external
factors (Praharaj et al., 2018), such as economic policy uncertainty (Sharif HT: Formal Analysis, Methodology, Validation,
et al., 2021) and political uncertainty (Sohail et al., 2022), leading to their Writing–original draft. J-BW: Data curation, Investigation,
consideration in future studies. Software, Writing–review and editing. C-YO: Conceptualization,
Investigation, Supervision, Validation, Writing–review and editing.
8 Implications
Funding
SCP not only enhances firms’ total factor productivity (Chen,
2022), but also bolsters ESG performance. This in turn helps the The author(s) declare that no financial support was received for
China government combat industrialization-induced challenges and the research, authorship, and/or publication of this article.
stride towards sustainable development. Consequently, this study
offers several implications.
First, from a policy perspective, our empirical analysis provides Acknowledgments
critical insights for policymakers in developing economies. To address a
broad spectrum of environmental challenges, China initiated a smart The authors like to thank “Construction and Practice of
city pilot policy in 2012. This policy aimed at revolutionizing the Industry-University-Research-Education Collaborative Talent
traditional urban development model through digital technological Cultivation Model in Public Administration Majors for Digital
reform, thereby creating opportunities for both innovations and Governance.” The Higher Education Association of Guangdong
green development (Cao et al., 2019). This paper aligns with these Province, No. 23GYB96; “Study on the Coupling Relationship
efforts, showcasing how SCP bolsters ESG performance and lays the Between Regional Economic Resilience and Rural Revitalization
groundwork for a transition to sustainability. in the Context of Industrial Transfer in the Eastern, Western, and
Second, our research underscores the pivotal role of SCP in Northern Counties of Guangdong Province.” Department of
boosting corporate ESG performance, particularly in developing education of Guangdong Province, No. 2023WTSCX153.
countries, due to its potential to exert a system-level impact. It also
highlights the importance of policy for advancing digital infrastructure
to facilitate sustainability transition in cities. Policymakers should Conflict of interest
therefore persist in fostering public digital infrastructure
development (such as AI and IoT), along with the corresponding The authors declare that the research was conducted in the
expertise. This support will help city stakeholders overcome barriers absence of any commercial or financial relationships that could be
to AI and IoT implementation and develop necessary capabilities. construed as a potential conflict of interest.

Tang et al. 10.3389/fenvs.2023.1305539
Publisher’s note organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
All claims expressed in this article are solely those of the authors claim that may be made by its manufacturer, is not guaranteed or
and do not necessarily represent those of their affiliated endorsed by the publisher.
References
Abbott, L. J., Parker, S., Peters, G. F., and Rama, D. V. (2007). Corporate governance, Intelligence Management (SSIM). 2022 2nd International Conference on Social Sciences
audit quality, and the Sarbanes-Oxley Act: evidence from internal audit outsourcing. and Intelligence Management (SSIM), Taichung, Taiwan, 24-26 November 2022,
Account. Rev. 82 (4), 803–835. doi:10.2308/accr.2007.82.4.803 85–89. doi:10.1109/SSIM55504.2022.10047950
Acemoglu, D., Aghion, P., Lelarge, C., Van Reenen, J., and Zilibotti, F. (2007). Feng, Y., and Hu, S. (2022). The effect of smart city policy on urban haze pollution in
Technology, information, and the decentralization of the firm. Q. J. Econ. 122 (4), China: empirical evidence from a quasi-natural experiment. Pol. J. Environ. Stud. 31 (3),
1759–1799. doi:10.1162/qjec.2007.122.4.1759 2083–2092. doi:10.15244/pjoes/143775
Angelidou, M. (2015). Smart cities: a conjuncture of four forces. Cities 47, 95–106. Filiou, D., Kesidou, E., and Wu, L. (2023). Are smart cities green? The role of
doi:10.1016/j.cities.2015.05.004 environmental and digital policies for Eco-innovation in China. World Dev. 165,
106212. doi:10.1016/j.worlddev.2023.106212
Baraibar-Diez, E., and Odriozola, M. D. (2019). CSR committees and their effect on
ESG performance in UK, France, Germany, and Spain. Sustainability 11 (18), 5077. Gao, J., Chu, D., Zheng, J., and Ye, T. (2022). Environmental, social and governance
doi:10.3390/su11185077 performance: can it be a stock price stabilizer? J. Clean. Prod. 379, 134705. doi:10.1016/j.
jclepro.2022.134705
Baron, R. M., and Kenny, D. A. (1986). The moderator–mediator variable distinction
in social psychological research: conceptual, strategic, and statistical considerations. Ge, G., Xiao, X., Li, Z., and Dai, Q. (2022). Does ESG performance promote high-
J. personality Soc. Psychol. 51 (6), 1173–1182. doi:10.1037/0022-3514.51.6.1173 quality development of enterprises in China? The mediating role of innovation input.
Sustainability 14 (7), 3843. doi:10.3390/su14073843
Barros, V., Verga Matos, P., Miranda Sarmento, J., and Rino Vieira, P. (2022). M&A
activity as a driver for better ESG performance. Technol. Forecast. Soc. Change 175, Gibbs, D., and O’Neill, K. (2018). Future green economies and regional development:
121338. doi:10.1016/j.techfore.2021.121338 a research agenda. Transitions Regional Econ. Dev., 287–309. doi:10.4324/
9781315143736-13
Barykin, S. E., Strimovskaya, A., Sergeev, S., Borisoglebskaya, L., Dedyukhina, N.,
Sklyarov, I., et al. (2023). Smart city logistics on the basis of digital tools for ESG goals Gillan, S. L., Koch, A., and Starks, L. T. (2021). Firms and social responsibility: a
achievement. Sustainability 15 (6), 5507. doi:10.3390/su15065507 review of ESG and CSR research in corporate finance. J. Corp. Finance 66, 101889.
doi:10.1016/j.jcorpfin.2021.101889
Beck, T., Levine, R., and Levkov, A. (2010). Big bad banks? The winners and losers
from bank deregulation in the United States. J. finance 65 (5), 1637–1667. doi:10.1111/j. Guo, Q., Wang, Y., and Dong, X. (2022). Effects of smart city construction on energy
1540-6261.2010.01589.x saving and CO2 emission reduction: evidence from China. Appl. Energy 313, 118879.
doi:10.1016/j.apenergy.2022.118879
Böckerman, P., and Ilmakunnas, P. (2009). Unemployment and self-assessed health:
evidence from panel data. Health Econ. 18 (2), 161–179. doi:10.1002/hec.1361 Gupta, G., and Bose, I. (2022). Digital transformation in entrepreneurial firms
through information exchange with operating environment. Inf. Manag. 59 (3),
Cao, W., Zhang, Y., and Qian, P. (2019). The effect of innovation-driven strategy on
103243. doi:10.1016/j.im.2019.103243
green economic development in China—an empirical study of smart cities. Int.
J. Environ. Res. Public Health 16 (9), 1520. doi:10.3390/ijerph16091520 Hao, D. Y., Qi, G. Y., and Wang, J. (2018). Corporate social responsibility, internal
controls, and stock price crash risk: the Chinese stock market. Sustainability 10,
Caragliu, A., and Del Bo, C. F. (2019). Smart innovative cities: the impact of Smart
1675–1697. doi:10.3390/su10051675
City policies on urban innovation. Technol. Forecast. Soc. Change 142, 373–383. doi:10.
1016/j.techfore.2018.07.022 Hao, X., Li, Y., Ren, S., Wu, H., and Hao, Y. (2023). The role of digitalization on green
economic growth: does industrial structure optimization and green innovation matter?
Cavada, M., Hunt, D. V., and Rogers, C. D. (2015). “Do smart cities realise their
J. Environ. Manag. 325, 116504. doi:10.1016/j.jenvman.2022.116504
potential for lower carbon dioxide emissions?,” in Proceedings of the institution of civil
engineers-engineering sustainability (United Kingdom: Thomas Telford Ltd), 243–252. Harasheh, M., and Provasi, R. (2023). A need for assurance: do internal control
systems integrate environmental, social, and governance factors? Corp. Soc. Responsib.
Chang, N., and Han, C. (2020). Cost-push impact of taxing carbon in China: a price
Environ. Manag. 30 (1), 384–401. doi:10.1002/csr.2361
transmission perspective. J. Clean. Prod. 248, 119194. doi:10.1016/j.jclepro.2019.119194
He, K., Zhu, N., Jiang, W., and Zhu, C. (2022). Efficiency evaluation of Chinese
Chen, J. C., Cho, C. H., and Patten, D. M. (2014). Initiating disclosure of
provincial industrial system Based on network DEA method. Sustainability 14 (9), 5264.
environmental liability information: an empirical analysis of firm choice. J. Bus.
doi:10.3390/su14095264
ethics 125, 681–692. doi:10.1007/s10551-013-1939-0
Heyman, F., Sjöholm, F., and Tingvall, P. G. (2007). Is there really a foreign ownership
Chen, P. (2022). The impact of smart city pilots on corporate total factor productivity.
wage premium? Evidence from matched employer–employee data. J. Int. Econ. 73 (2),
Environ. Sci. Pollut. Res. 29 (55), 83155–83168. doi:10.1007/s11356-022-21681-1
355–376. doi:10.1016/j.jinteco.2007.04.003
Chen, Y.-S., Lai, S.-B., and Wen, C.-T. (2006). The influence of green innovation
Hu, Q., and Zheng, Y. (2021). Smart city initiatives: a comparative study of
performance on corporate advantage in Taiwan. J. Bus. ethics 67, 331–339. doi:10.1007/
American and Chinese cities. J. Urban Aff. 43 (4), 504–525. doi:10.1080/07352166.
s10551-006-9025-5
2019.1694413
Chen, Z., and Xie, G. (2022). ESG disclosure and financial performance: moderating role of
Huang, M., Li, M., and Liao, Z. (2021). Do politically connected CEOs promote
ESG investors. Int. Rev. Financial Analysis 83, 102291. doi:10.1016/j.irfa.2022.102291
Chinese listed industrial firms’ green innovation? The mediating role of external
Chien, F., Hsu, C., Andlib, Z., Shah, M. I., Ajaz, T., and Genie, M. G. (2022). The role governance environments. J. Clean. Prod. 278, 123634. doi:10.1016/j.jclepro.2020.
of solar energy and eco-innovation in reducing environmental degradation in China: 123634
evidence from QARDL approach. Integr. Environ. Assess. Manag. 18 (2), 555–571.
Huang, W., Luo, Y., Wang, X., and Xiao, L. (2022). Controlling shareholder pledging and
doi:10.1002/ieam.4500
corporate ESG behavior. Res. Int. Bus. Finance 61, 101655. doi:10.1016/j.ribaf.2022.101655
Chu, Z., Cheng, M., and Yu, N. N. (2021). A smart city is a less polluted city. Technol.
Ikram, A., Li, Z. F., and Minor, D. (2019). CSR-contingent executive compensation
Forecast. Soc. Change 172, 121037. doi:10.1016/j.techfore.2021.121037
contracts. J. Bank. Finance 151, 105655. doi:10.1016/j.jbankfin.2019.105655
Dehejia, R. H., and Wahba, S. (2002). Propensity score-matching methods for
Jang, G.-Y., Kang, H.-G., and Kim, W. (2022). Corporate executives’ incentives and
nonexperimental causal studies. Rev. Econ. statistics 84 (1), 151–161. doi:10.1162/
ESG performance. Finance Res. Lett. 49, 103187. doi:10.1016/j.frl.2022.103187
003465302317331982
Jiang, H., Jiang, P., Wang, D., and Wu, J. (2021). Can smart city construction facilitate
Dong, F., Li, Y., Li, K., Zhu, J., and Zheng, L. (2022). Can smart city construction improve
green total factor productivity? A quasi-natural experiment based on China’s pilot smart
urban ecological total factor energy efficiency in China? Fresh evidence from generalized
city. Sustain. Cities Soc. 69, 102809. doi:10.1016/j.scs.2021.102809
synthetic control method. Energy 241, 122909. doi:10.1016/j.energy.2021.122909
Koo, J. E., and Ki, E. S. (2020). Internal control personnel’s experience, internal
Dong, W., Dong, X., and Lv, X. (2022). How does ownership structure affect corporate
control weaknesses, and ESG rating. Sustainability 12 (20), 8645. doi:10.3390/
environmental responsibility? Evidence from the manufacturing sector in China. Energy
su12208645
Econ. 112, 106112. doi:10.1016/j.eneco.2022.106112
Le, H. Q., Vu, T. P. L., Do, V. P. A., and Do, A. D. (2022). The enduring effect of
Fan, Y. (2022). “Research on internal control of yihua company under technology of
formalization on firm-level corruption in Vietnam: the mediating role of internal
artificial intelligence,” in 2022 2nd International Conference on Social Sciences and
control. Int. Rev. Econ. Finance 82, 364–373. doi:10.1016/j.iref.2022.06.021

Tang et al. 10.3389/fenvs.2023.1305539
Li, D., and Shen, W. (2021). Can corporate digitalization promote green innovation? evidence from 253 prefecture-level cities in China. J. Clean. Prod. 269, 122425. doi:10.
The moderating roles of internal control and institutional ownership. Sustainability 13 1016/j.jclepro.2020.122425
(24), 13983. doi:10.3390/su132413983
Su, Z., Guo, Q., and Lee, H.-T. (2022). RETRACTED: green finance policy and
Li, D., Zhou, Z., Cao, L., Zhao, K., Li, B., and Ding, C. (2023). What drives the change enterprise energy consumption intensity: evidence from a quasi-natural experiment in
in China’s provincial industrial carbon unlocking efficiency? Evidence from a China. Energy Econ. 115, 106374. doi:10.1016/j.eneco.2022.106374
geographically and temporally weighted regression model. Sci. Total Environ. 856,
Sun, L., Liu, W., Li, Z., Cai, B., Fujii, M., Luo, X., et al. (2021). Spatial and structural
158971. doi:10.1016/j.scitotenv.2022.158971
characteristics of CO2 emissions in East Asian megacities and its indication for low-carbon
Li, L., Zheng, Y., Zheng, S., and Ke, H. (2020). The new smart city programme: city development. Appl. Energy 284, 116400. doi:10.1016/j.apenergy.2020.116400
evaluating the effect of the internet of energy on air quality in China. Sci. Total Environ.
Sun, L., and Saat, N. A. M. (2023). How does intelligent manufacturing affect the ESG
714, 136380. doi:10.1016/j.scitotenv.2019.136380
performance of manufacturing firms? Evidence from China. Sustainability 15 (4), 2898.
Lima, M. (2020). Smarter organizations: insights from a smart city hybrid framework. doi:10.3390/su15042898
Int. Entrepreneursh. Manag. J. 16 (4), 1281–1300. doi:10.1007/s11365-020-00690-x
Tan, S. Y., and Taeihagh, A. (2020). Smart city governance in developing countries: a
Lin, C., Huang, H.-C., Wu, J.-Y., and Liu, B. (2002). A knowledge management systematic literature review. Sustainability 12 (3), 899. doi:10.3390/su12030899
architecture in collaborative supply chain. J. Comput. Inf. Syst. 42 (5), 83–94. doi:10.
Vázquez, J. L., Lanero, A., Gutierrez, P., and Sahelices, C. (2018). “The contribution of
1080/08874417.2002.11647612
smart cities to quality of life from the view of citizens,” in Entrepreneurial, innovative
Liu, J.-Y. (2018). An internal control system that includes corporate social and sustainable ecosystems: best practices and implications for quality of life (Berlin,
responsibility for social sustainability in the new era. Sustainability 10 (10), 3382. Germany: Springer), 55–66. doi:10.1007/978-3-319-71014-3_3
doi:10.3390/su10103382
Wang, J., and Deng, K. (2022). Impact and mechanism analysis of smart city policy on
Liu, K., Liu, X., Long, H., Wang, D., and Zhang, G. (2022). Spatial agglomeration and urban innovation: evidence from China. Econ. Analysis Policy 73, 574–587. doi:10.1016/
energy efficiency: evidence from China’s manufacturing enterprises. J. Clean. Prod. 380, j.eap.2021.12.006
135109. doi:10.1016/j.jclepro.2022.135109
Wang, K.-L., Pang, S. Q., Zhang, F. Q., Miao, Z., and Sun, H. P. (2022). The impact
Liu, S., Liu, C., and Yang, M. (2021). The effects of national environmental assessment of smart city policy on urban green total-factor productivity: evidence from
information disclosure program on the upgradation of regional industrial structure: China. Environ. Impact Assess. Rev. 94, 106756. doi:10.1016/j.eiar.2022.106756
evidence from 286 prefecture-level cities in China. Struct. Change Econ. Dyn. 58,
Wang, S. S., Zhou, D., Zhou, P., and Wang, Q. (2011). CO2 emissions, energy
552–561. doi:10.1016/j.strueco.2021.07.006
consumption and economic growth in China: a panel data analysis. Energy policy 39 (9),
López-Arranz, M. A. (2019). “The role corporate social responsibility has in the smart 4870–4875. doi:10.1016/j.enpol.2011.06.032
city project in Spain,” in Smart cities and smart spaces: concepts, methodologies, tools,
Wang, X., Wu, W., and Li, H. (2023). The impact of city ranking on industry shifting:
and applications (United States: IGI Global), 1562–1582.
an empirical study. Sustainability 15 (11), 8930. doi:10.3390/su15118930
McCarthy, S., Oliver, B., and Song, S. (2017). Corporate social responsibility and CEO
Wang, Y., Lin, Y., Fu, X., and Chen, S. (2023). Institutional ownership heterogeneity
confidence. J. Bank. Finance 75, 280–291. doi:10.1016/j.jbankfin.2016.11.024
and ESG performance: evidence from China. Finance Res. Lett. 51, 103448. doi:10.1016/
McGuinness, P. B., Vieito, J. P., and Wang, M. (2017). The role of board gender and j.frl.2022.103448
foreign ownership in the CSR performance of Chinese listed firms. J. Corp. Finance 42,
Wen, H., Liang, W., and Lee, C.-C. (2023). China’s progress toward sustainable
75–99. doi:10.1016/j.jcorpfin.2016.11.001
development in pursuit of carbon neutrality: regional differences and dynamic
Menicucci, E., and Paolucci, G. (2022). Board diversity and ESG performance: evolution. Environ. Impact Assess. Rev. 98, 106959. doi:10.1016/j.eiar.2022.106959
evidence from the Italian banking sector. Sustainability 14 (20), 13447. doi:10.3390/
Witkowski, K. (2017). Internet of things, big data, industry 4.0–innovative solutions
su142013447
in logistics and supply chains management. Procedia Eng. 182, 763–769. doi:10.1016/j.
MOHURD (2012). National smart city (district, town) pilot indicator system (trial). proeng.2017.03.197
Available at: https://www.gov.cn/gzdt/2012-12/10/content_2286787.htm (Accessed
Xie, X., Huo, J., and Zou, H. (2019). Green process innovation, green product
July 19, 2023).
innovation, and corporate financial performance: a content analysis method. J. Bus.
MOHURD (2013). The ministry of housing and urban-rural development has Res. 101, 697–706. doi:10.1016/j.jbusres.2019.01.010
released a list of national pilot smart cities. Available at: https://www.gov.cn/jrzg/
Xu, B., and Lin, B. (2016). Reducing carbon dioxide emissions in China’s
2013-08/05/content_2461575.htm (Accessed July 19, 2023).
manufacturing industry: a dynamic vector autoregression approach. J. Clean. Prod.
Peattie, K., and Ratnayaka, M. (1992). Responding to the green movement. Ind. Mark. 131, 594–606. doi:10.1016/j.jclepro.2016.04.129
Manag. 21 (2), 103–110. doi:10.1016/0019-8501(92)90004-D
Xu, G., and Yang, Z. (2022). The mechanism and effects of national smart city pilots in
Praharaj, S., Han, J. H., and Hawken, S. (2018). Urban innovation through policy China on environmental pollution: empirical evidence based on a DID model. Environ.
integration: critical perspectives from 100 smart cities mission in India. City, Cult. Soc. Sci. Pollut. Res. 29 (27), 41804–41819. doi:10.1007/s11356-021-18003-2
12, 35–43. doi:10.1016/j.ccs.2017.06.004
Xu, S., Miao, Li, Zhou, Ma, He, , et al. (2019). What factors drive air pollutants in
Qi, H., Cao, X., and Liu, Y. (2020). The influence of digital economy on corporate China? an analysis from the perspective of Regional difference using a combined
governance: analyzed from information asymmetry and irrational behavior perspective. method of production decomposition analysis and logarithmic mean divisia index.
Reform 4, 50–64. (In Chinese). Sustainability 11 (17), 4650. doi:10.3390/su11174650
Qi, Y. D., Liu, C. H., and Ding, S. L. (2020). Digital economic development, Yan, Y., Cheng, Q., Huang, M., Lin, Q., and Lin, W. (2023). Government
employment structure optimization and employment quality upgrading. Econ. environmental regulation and corporate ESG performance: evidence from natural
Perspect. 11, 17–35. (In Chinese). resource accountability audits in China. Int. J. Environ. Res. Public Health 20 (1),
447. doi:10.3390/ijerph20010447
Qian, L., Xu, X., Zhou, Y., Sun, Y., and Ma, D. (2023). Carbon emission reduction
effects of the smart city pilot policy in China. Sustainability 15 (6), 5085. doi:10.3390/ Yan, Z., Sun, Z., Shi, R., and Zhao, M. (2023). Smart city and green development:
su15065085 empirical evidence from the perspective of green technological innovation. Technol.
Forecast. Soc. Change 191, 122507. doi:10.1016/j.techfore.2023.122507
Qiu, L. (2022). Does the construction of smart cities promote urban green innovation?
Evidence from China. Appl. Econ. Lett. 0 (0), 2688–2696. doi:10.1080/13504851.2022.2103497 Yang, L., Qin, H., Gan, Q., and Su, J. (2020). Internal control quality, enterprise
environmental protection investment and finance performance: an empirical study of
Sharif, A., Bhattacharya, M., Afshan, S., and Shahbaz, M. (2021). Disaggregated
China’s a-share heavy pollution industry. Int. J. Environ. Res. Public Health 17 (17),
renewable energy sources in mitigating CO2 emissions: new evidence from the USA
6082. doi:10.3390/ijerph17176082
using quantile regressions. Environ. Sci. Pollut. Res. 28 (41), 57582–57601. doi:10.1007/
s11356-021-13829-2 Yang, Z. H. (2018). Can smart city improve economic efficiency: quasi natural
experiment based on smart city construction. Sci. Tech. Mgt. Res. 10, 263–266. doi:10.
Shi, D., Ding, H., Wei, P., and Liu, J. (2018). Can smart city construction reduce
3969/j.issn.1000-7695.2018.10.036
environmental pollution. China Ind. Econ. 6, 117–135. doi:10.19581/j.cnki.ciejournal.
2018.06.008 Yigitcanlar, T., and Kamruzzaman, M. (2018). Does smart city policy lead to sustainability
of cities? ’, Land use policy 73, 49–58. doi:10.1016/j.landusepol.2018.01.034
Sohail, M. T., Majeed, M. T., Shaikh, P. A., and Andlib, Z. (2022). Environmental costs of
political instability in Pakistan: policy options for clean energy consumption and environment. Yong, J. (2020). Digital economy empowers manufacturing transformation: from
Environ. Sci. Pollut. Res. 29 (17), 25184–25193. doi:10.1007/s11356-021-17646-5 value reshaping to value creation. Economists (6), 89–94. doi:10.16158/j.cnki.51-1312/f.
2020.06.010
Solana-González, P., Vanti, A. A., García Lorenzo, M. M., and Bello Pérez, R. E.
(2021). Data mining to assess organizational transparency across technology processes: Zhao, X., Wu, L., and Li, A. (2017). Research on the efficiency of carbon trading
an approach from it governance and knowledge management. Sustainability 13 (18), market in China. Renew. Sustain. Energy Rev. 79, 1–8. doi:10.1016/j.rser.2017.05.034
10130. doi:10.3390/su131810130
Zhong, Q., Han, L., and Jin, J. (2022). Do green credit guidelines impact on heavily
Song, Y., Yang, T., Li, Z., Zhang, X., and Zhang, M. (2020). Research on the direct and polluting firms in rent-seeking? Finance Res. Lett. 47, 102747. doi:10.1016/j.frl.
indirect effects of environmental regulation on environmental pollution: empirical 2022.102747

02 - Fenvs-11-1305539

Uploaded by

Copyright:

Available Formats

02 - Fenvs-11-1305539

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

02 - Fenvs-11-1305539

Uploaded by

Copyright:

Available Formats

TYPE Original Research

PUBLISHED 08 January 2024

How do smart city pilots affect the

RECEIVED 02 October 2023

Frontiers in Environmental Science 01 frontiersin.org

Frontiers in Environmental Science 02 frontiersin.org

Frontiers in Environmental Science 03 frontiersin.org

2.4 Mechanism of corporate internet

Frontiers in Environmental Science 04 frontiersin.org

TABLE 1 Variable deﬁnitions.

Symbol Variable Measurement

Lev Financial leverage Total ﬁnancial liabilities/total assets

Roa Return on total assets Corporate net proﬁt/total assets

TobinQ Tobin’s Q value Stock market value of total assets

lnpgdp Per capita GDP Natural logarithm of per capita GDP

rgdp GDP growth rate GDP growth rate

Frontiers in Environmental Science 05 frontiersin.org

Roa 13,923 0.033 0.064 −0.329 0.179

Frontiers in Environmental Science 06 frontiersin.org

TABLE 3 Baseline regression results.

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

Variable HESG HESG HESG HESG HESG HESG

(0.034) (0.033) (0.034) (0.034) (0.051) (0.048)

Size 0.152*** 0.151*** 0.151*** 0.147***

(0.023) (0.023) (0.026) (0.032)

Lev −0.130** −0.128** −0.128* −0.161**

(0.062) (0.062) (0.073) (0.082)

Roa 1.336*** 1.332*** 1.332*** 1.323***

(0.169) (0.169) (0.206) (0.197)

Pcontrol −0.000 −0.000 −0.000 −0.000

(0.001) (0.001) (0.001) (0.001)

Top1 0.005*** 0.005*** 0.005** 0.006***

(0.002) (0.002) (0.002) (0.002)

TobinQ −0.035*** −0.034*** −0.034** −0.042***

(0.010) (0.010) (0.014) (0.013)

Age 0.182 0.180 0.180 −0.041

(0.140) (0.142) (0.195) (0.110)

lnpgdp 0.102* 0.083 0.083 0.106

(0.059) (0.058) (0.072) (0.078)

rgdp −0.005 −0.006 −0.006 −0.005

(0.005) (0.005) (0.006) (0.006)

lnso2 −0.043** −0.037** −0.037 −0.021

(0.019) (0.019) (0.024) (0.028)

lnpm10 −0.029 −0.024 −0.024 −0.007

(0.064) (0.064) (0.101) (0.094)

lnno −0.062* −0.043 −0.043 −0.064

(0.033) (0.033) (0.045) (0.046)

lnsewage 0.124 0.180 0.180 0.145

(0.142) (0.140) (0.228) (0.182)

lnwaste 0.016 0.012 0.012 −0.030

(0.038) (0.037) (0.051) (0.049)

Constant 3.992*** −1.151 3.413*** −1.972 −1.972 0.230

(0.013) (1.396) (0.997) (1.715) (2.284) (1.807)

Firm FE Yes Yes Yes Yes Yes Yes

Year FE Yes Yes Yes Yes Yes Yes

City cluster No No No No Yes No

Adjusted R2 0.457 0.468 0.458 0.469 0.469 0.465

Observations 13,923 13,923 13,923 13,923 13,923 12,068

Frontiers in Environmental Science 07 frontiersin.org

Size 0.152* 0.151* 0.151* 0.147*

Lev −0.130 −0.128 −0.128* −0.161**

Roa 1.336* 1.332* 1.332* 1.323*

Top1 0.005* 0.005* 0.005 0.006*

TobinQ −0.035* −0.034* −0.034 −0.042*

lnso2 −0.043 −0.037 −0.037 −0.021

Constant 3.992* −1.151 3.413* −1.972 −1.972 0.230

Size 0.161* 0.160* 0.184***

Roa 1.138* 1.254* 1.009***

Top1 0.005 0.005* 0.002

TobinQ −0.053* −0.036* −0.056***

Size 0.148* 0.147* 0.147* 0.148* 0.148***

Roa 1.469* 1.469* 1.468* 1.468* 1.468***

Top1 0.005 0.005 0.005 0.005 0.005**

TobinQ −0.040* −0.040* −0.040* −0.040* −0.040***