Impact of Digital Transformation on Green Innovation in Manufacturing under Dual Carbon Targets

Abstract

The development of green innovation in the manufacturing industry is crucial for sustainability, as it can lead to significant environmental and economic benefits. Meanwhile, the impact of digital transformation on green innovation in the manufacturing industry has been proven to be significant in a previous study. To further explore the impact of digital transformation on the development of green innovation in the manufacturing industry under the dual carbon goal, this article selects data from Chinese A-share manufacturing listed companies from 2013 to 2022 as the observation sample, proposes research hypotheses based on stakeholder theory, and conducts empirical analysis. The results indicate that digital transformation can significantly promote the development of green innovation in the manufacturing industry, which is transmitted through corporate environmental responsibility. Corporate environmental responsibility plays a partial mediating role in the impact of digital transformation on the development of green innovation. Meanwhile, media attention can strengthen the positive impact of digital transformation on green innovation. In heterogeneity analysis, it was found that the higher the institutional shareholding and analyst attention, the more likely a company’s digital transformation can promote green innovation. In addition, by comparing and analyzing the data of the two years before and after the proposal of the dual carbon target, it was found that the incentive effect of the dual carbon target did not achieve the expected effect in this article, which may be related to the short time of the proposal of the dual carbon target and the impact of the epidemic.

1. Introduction

Since the reform and opening up, China’s economy has maintained sustained high-speed development, creating economic miracles. However, the extensive development model has made it difficult for China to sustain energy resources, and carbon emissions remain high. Therefore, in 2020, China explicitly proposed the dual carbon target, aiming to peak carbon emissions by 2030 and achieve carbon neutrality by 2060 [1,2,3]. The dual carbon target places higher demands on ecological civilization construction and green low-carbon development, promoting industrial structural upgrading and sustainable development. As a key pillar industry of the country, the manufacturing industry is vital to the national economy, determining the comprehensive strength and international competitiveness of the country [4,5]. Promoting high-quality development of the manufacturing industry is an inherent requirement for constructing a modern economic system [6,7]. Under the backdrop of the dual carbon objectives, the transformation and upgrading of the manufacturing industry to achieve green innovative development are of great significance. Despite its growing significance, green innovation remains understudied in academic research. Although it is an emerging field with significant potential for impact, it has yet to receive the scholarly attention it deserves [8]. This gap highlights the need for more focused studies and in-depth research to advance our understanding of green innovation and its implications across multiple sectors.

Over the years, the continuous breakthroughs in digital technology have given rise to a wave of digital transformation that has swept across the globe [9,10,11]. By introducing advanced digital technologies, manufacturing enterprises can achieve automation and intelligence in production and operations, enhance production efficiency and quality, reduce waste and resource consumption, and lower energy consumption and carbon dioxide emissions. At the same time, digital technology brings more innovation opportunities to the manufacturing industry. “Digital + Manufacturing” has become an inevitable choice for the transformation and upgrading of China’s manufacturing industry, providing a strong impetus for green and high-quality development of the manufacturing industry. Numerous studies have found that digital transformation improves sustainable business performance in social, environmental, and economic dimensions [12]. Additionally, extensive research has investigated how digital technologies drive transformations in various aspects of environmental sustainability, such as pollution control, waste management, sustainable production, and urban sustainability [13]. Despite this wealth of research, there is still a significant gap in the literature regarding the impact of digital transformation on green innovation. This gap highlights the need for additional research to fully understand how digital transformation can advance green innovation, which is critical to achieving comprehensive environmental sustainability [14]. It represents a new developmental model of green and innovation, mainly manifested in the manufacturing industry as the design of new processes conducive to emission reduction, development of environmentally friendly technologies, and manufacture of energy-saving products to improve resource utilization and reduce environmental waste, thereby achieving sustainable development. The development of green innovation in the manufacturing industry relies on the support of digital innovation technologies. Digital transformation is essential in promoting manufacturing enterprises’ energy conservation, emission reduction, green innovation development, and realization of dual carbon goals. Understanding how digital transformation can accelerate green innovation is critical for businesses seeking to align their operations with sustainability goals.

Enterprises, as the most crucial micro subjects in the economic society, undergo digital transformation by extensively applying digital technologies, restructuring various aspects of enterprises, and realizing the restructuring of digital business within enterprises, between enterprises, and across the entire industry chain to reduce resource waste and environmental pollution [15]. Digital transformation can effectively motivate enterprises, especially heavily polluting ones, to fulfill environmental responsibilities. It can also enhance corporate environmental responsibility by improving internal disclosure and increasing external investor attention [16]. Furthermore, corporate environmental responsibility is the foundation of sustainable development, as it promotes the formation of green development concepts within enterprises, encourages managers and employees to embrace environmental awareness [17], and facilitates the development of green innovation within enterprises. Based on the current research, most studies on digital transformation mainly focus on economic performance and influencing factors. However, there is relatively little research on how digital transformation affects the green innovation development of manufacturing enterprises. Therefore, this study selects A-share-listed manufacturing enterprises in China as research subjects, conducts an in-depth analysis of their panel data, explores the inherent connection between digital transformation and green innovation development, further investigates the potential mediating role of corporate environmental responsibility between digital transformation and green innovation development, and studies the moderating effect of media attention.

2. Research Hypotheses

2.1. Digital Transformation (DT) and Green Innovation (GI)

The advent of the digital economy era and the rapid development of digital technologies such as big data have led to a transformation of people’s life concepts and have had significant impacts on the socio–economic landscape. Under the backdrop of the digital economy, enterprise digital transformation represents a way for enterprises to transform and upgrade, seeking new avenues for value creation. Green innovation, on the other hand, is a favorable approach for enterprises to implement national policies under the dual carbon goals and carbon reduction policies.

Digital transformation, through the application of digital technologies, can enhance information-sharing efficiency [18], effectively alleviate information asymmetry [19], optimize resource allocation [20], and enhance corporate governance [21], thereby exerting a positive influence on green innovation [22]. Focusing on the manufacturing industry, digital transformation similarly promotes green innovation [23]. Most studies in academia on the relationship between digital transformation and green innovation show a positive correlation. Digitalization is a core strategy for Chinese enterprises to achieve high-quality development, significantly impacting current production and lifestyle. Concurrently, green sustainable development and balanced development have become themes of the new era. For manufacturing enterprises, digital transformation can facilitate technological innovation, particularly in green technology innovation and application. Based on the above, this study proposes the following hypothesis:

Hypothesis 1: 

Digital transformation positively promotes green innovation in manufacturing enterprises.

2.2. Mediating Effect of Corporate Environmental Responsibility (CER)

The concept of Corporate Environmental Responsibility (CER) began to be studied and emphasized in the 1960s and 1970s. Most scholars consider corporate environmental responsibility as a dimension of corporate social responsibility [24]. This study similarly views it as a dimension of corporate social responsibility and conducts research accordingly.

In academia, studies on the relationship between digitalization and environmental responsibility mostly show a positive influence. High levels of digital transformation in heavily polluting industries such as thermal power and steel lead to higher ESG (Environmental, Social, and Governance) levels [25]. Digital transformation in heavily polluting enterprises can prompt them to protect the ecological environment, fulfill social responsibilities, enhance governance capabilities, and further improve ESG performance [26]. Under the dual carbon goals, corporate environmental responsibility is increasingly emphasized, and digital transformation can promote specific environmental behaviors of enterprises, such as environmental governance costs. Digitalization is advantageous in reducing information asymmetry and transaction costs, enhancing information transparency, aiding companies in fulfilling environmental responsibilities, and improving resource allocation and utilization efficiency, thereby reducing resource wastage. Therefore, digital transformation not only helps mitigate the negative impact on Corporate Environmental Responsibility but also accordingly enhances awareness among manufacturing enterprises.

Corporate green innovation activities not only optimize the allocation of internal and external resources but also reflect a profound understanding of environmental and social responsibility by enterprises. Additionally, they demonstrate a sense of responsibility towards stakeholders. By actively assuming environmental responsibility, enterprises can not only gain recognition and support from stakeholders but also acquire more tangible and intangible resources internally and externally, further promoting the deepening and development of green innovation activities.

CER can promote green innovation through various mechanisms. Firstly, CER can promote green innovation through a shared vision [27]. Secondly, under government subsidy incentives, CER can drive green innovation [28]. Thirdly, CER helps enhance corporate competitiveness and gain innovation advantages, thus fostering green innovation [29]. Existing research indicates that CER has a certain promoting effect on green innovation.

Green innovation is a highly comprehensive and systematic project involving the integration of knowledge and resources from multiple fields, including digital technology and green knowledge. Due to the complexity of green innovation, individual manufacturing enterprises often struggle to possess all the necessary resources for such innovation [30]. From the perspective of resource integration, digital transformation helps integrate resources from various parties, creating an environment conducive to green innovation. From a stakeholders’ perspective, digital transformation plays a crucial role in providing enterprises with more convenient information exchange channels. Enterprises, through the use of digital technology, can bridge the information gap between themselves and stakeholders, enabling stakeholders to have a more comprehensive understanding of the company, assess the fulfillment of CER, and actively contribute to green innovation initiatives [31]. According to stakeholder theory, enterprises must ensure the interests of stakeholders, including shareholders. Enterprises are willing to sacrifice some economic benefits to fulfill CER, prioritize the demands of social groups and other stakeholders, and gain recognition from investors, the public, and the government, thereby obtaining higher-quality resources [32]. Furthermore, this provides enterprises with a broader space for innovative activities, allowing them to gain a deeper understanding and mastery of the knowledge resources, technological frontiers, and development trends required for green innovation. Consequently, enterprises can formulate more proactive green innovation plans, thereby promoting the sustained development of green innovation. Based on this, this study proposes the following hypothesis:

Hypothesis 2: 

Corporate Environmental Responsibility (CER) mediates between enterprise DT and green innovation.

2.3. Moderating Effect of Media Attention

Media serves as a channel for information dissemination. Generally, media is the medium, tool, or carrier used by the public to collect, process, and transmit information. Media attention can create social public opinion by focusing on hot topics and events, thereby influencing enterprise operations, promoting enterprise digital transformation, and green innovation.

As a strategic initiative vigorously promoted by the state, digital transformation in the manufacturing industry receives significant media attention. Media, as a channel for information dissemination, is a critical means for enterprises to convey information to the public. The information reported by the media guides public perception of enterprises, effectively reduces information asymmetry between enterprises and stakeholders [33], enhances enterprise transparency, and influences enterprise decision-making. Negative media coverage can quickly attract public attention and create social public opinion. At such times, stakeholder criticism and societal scrutiny often prompt enterprises to reevaluate and adjust their green innovation strategies. This process of adjustment under negative public opinion pressure may further strengthen the positive impact of enterprise digital transformation on green innovation [34]. When the media extensively covers positive information such as a company’s achievements in green innovation, it sends positive signals to external stakeholders. This positive coverage can enhance the company’s image and reputation in the market, increase market competitiveness, and enhance the trust of investors, financial institutions, and consumers. Consequently, enterprises may attract more investors, consumers, and funding, thereby supporting the advancement of green innovation. Therefore, this study proposes the following hypothesis:

Hypothesis 3: 

Media attention has a positive moderating effect on digital transformation and green innovation.

The specific hypothesis contents are summarized in

Table 1. Research hypothesis contents (source: authors’ own creation).

Research Hypothesis	Contents
Hypothesis 1	Under the dual carbon targets, digital transformation positively promotes green innovation in manufacturing enterprises in China.
Hypothesis 2	Under the dual carbon targets, corporate environmental responsibility (CER) mediates between enterprise digital transformation and green innovation in China.
Hypothesis 3	Hypothesis 3: Under the dual carbon targets, media attention media attention positively moderates digital transformation and green innovation in China.

.

3. Research Design

3.1. Data Source

Recent research has highlighted the significant impact of digital transformation on green innovation in the manufacturing sector. Artificial Intelligence (AI), the Internet of Things (IoT), and big data analytics are transforming pollution control, waste management, sustainable production, and urban sustainability initiatives [13]. These advancements fuel a shift toward greener practices, resulting in more sustainable manufacturing processes.

Studies have shown that digital transformation promotes green innovation by increasing resource investment and lowering debt costs. This dual effect leads to increased patent quality and the adoption of more sustainable innovation practices [10]. Incorporating digital transformation into manufacturing, known as Industry 4.0, provides several key benefits, including increased efficiency, sustainability, customization, and flexibility [35].

While existing studies effectively highlight the potential for digital transformation to promote green innovation, they primarily use qualitative methods such as literature reviews and surveys. In contrast, this current study takes a more comprehensive and rigorous approach, using multiple data sources and conducting a regression analysis from a big data perspective. Specifically, it incorporates implementation data on digitalization (DIG) from annual reports.

This study focuses on Chinese A-share-listed manufacturing companies and uses observational data collected between 2013 and 2022. This time frame was chosen based on data availability in online databases. The sampled companies consist of State-Owned Enterprises (SOEs), Private Enterprises, and Mixed Ownership Enterprises. State-Owned Enterprises (SOEs): in these companies, the major shareholders are the central or local government of China. Private Enterprises: these companies are primarily owned by private investors, including company founders, entrepreneurs, and other private entities. Mixed Ownership Enterprises: these firms feature a blend of state-owned and private investment shares, a model that is relatively common in China. Unlike previous research, which frequently relied solely on online data, this study includes additional implementation data on digitalization (DIG) sourced from corporate annual reports. The relevant data are sourced from several authoritative databases, including the National Intellectual Property Administration of the People’s Republic of China, the CSMAR Database, and the Sino-Securities ESG Database.

This study combines multiple data sources and employs robust analytical methods to provide a more detailed and accurate assessment of how digital transformation drives green innovation in the manufacturing sector.

To achieve the research objectives, improve data accuracy, and reference existing practices, this study applies the following criteria to sample selection:

(1)

Exclude listed companies with ST or *ST trading status for the year;

(2)

Exclude observational samples with significant missing data.

Furthermore, to enhance the accuracy of the research results, this study applies winsorization to all continuous variables at the top and bottom 1% of the samples, resulting in 21,842 observations.

3.2. Variable Measurement

3.2.1. Dependent Variable

Green Innovation (GI). Existing research widely uses the number of green patent applications as a proxy variable for GI. Therefore, this study adopts the sum of green invention patents and utility model patents applied by manufacturing listed companies in the current year to characterize green innovation. Referring to the research methods of scholars such as Tian Haifeng et al. (2023) [36], the number of green patent applications by enterprises is used to measure enterprise green innovation. In the robustness test section, this study selects the number of green patent grants as an alternative indicator to represent green innovation. Additionally, considering the potential lag effects of innovation activities, the number of green patent applications from the previous period is introduced as the dependent variable. Moreover, in light of the impact of the COVID-19 pandemic on enterprise development, this study conducts robustness tests after excluding samples from 2020 and 2021. Considering the timeline of the dual carbon targets, this study compares data from 2018 to 2019 and 2021 to 2022 to explore whether the promotion effects of the dual carbon targets have achieved the expected results.

3.2.2. Explanatory Variable

Digital Transformation (DT). For measuring corporate DT, this study first draws upon important literature on corporate DT research and, referring to Wu Fei et al. (2021) [37], adopts the frequency of key digitalization keywords appearing in corporate annual reports as the metric for assessing DT. This study utilizes Python 3.x software to calculate the frequency of DT keywords in corporate annual reports, strictly excluding keywords irrelevant to business operations or containing negations. Finally, the total sum of the frequency of DT keywords, plus one, and then the natural logarithm, is used as an indicator to measure the degree of corporate DT. The specific Selection of Keywords for Manufacturing Enterprise DT are outlined in

Table 2. Sino-securities ESG Environmental Responsibility Sub-Indicators (source: authors’ own creation).

Primary Indicators	Secondary Indicators	Tertiary Indicators
Environment	Climate Change	Green finance, sponge cities, addressing climate change, carbon reduction pathways, greenhouse gas emissions
	Resource utilization	Water resource consumption, land use, biodiversity, material consumption
	Environmental pollution	E-waste, hazardous waste, industrial emissions
	Environmentally friendly	Renewable energy, green factories, green buildings
	Environmental management	Medical Digitalization, Mobile Wallets, Barcode Payments, Supply Chain Management—E, Sustainable Certification, Environmental Penalties

.

3.2.3. Mediating Variable

Corporate Environmental Responsibility (CER). The existing literature on environmental responsibility measurement mostly considers environmental responsibility as a vital component of corporate social responsibility. The Sino-securities ESG assessment system includes three primary indicators—environment, social, and corporate governance—and 16 secondary indicators, with nine rating levels. This study, drawing upon research by Fan Yunpeng et al. (2023) [38], adopts the CER metric from the environmental responsibility sub-rating in the Sino-securities ESG assessment system. The specific environmental responsibility sub-indicators in the Sino-securities ESG system are outlined in

Table 3. Selection of Keywords for Manufacturing Enterprise DT (source: authors’ own creation).

Variable	Type	High-Frequency Keywords
Digital Transformation (DIG)	Artificial Intelligence (AI)	Artificial Intelligence (AI), Intelligent Warehousing, Semantic Search, Kingdee (EAS), Business Intelligence (BI), Financial Systems, Intelligent Robots, Fine Management, Intelligent Workshops, Expert Systems, Identity Verification, Numerical Control (NC), Facial Recognition, Electronic Design Automation (EDA), Customer Relationship Management (CRM), Enterprise Resource Planning (ERP), Manufacturing Resource Planning System (MRP), Investment Decision Support Systems, Enterprise Resource Management (ERP), Manufacturing Execution Systems (MES), Office Automation (OA), Product Lifecycle Management (PLM), Robotic Process Automation (RPA), Enterprise Resource Planning Software Systems (SAP S/4HANA), Financial Management Systems, Biometric Technology, UFIDA (U9), Machine Learning, Intelligence, Smart Office, Smart Technology, Intelligent Identification, Intelligent Manufacturing, Smart Terminals, Autonomous Driving, Robots, Automation Control, Smart Equipment, Smart Factory, Intelligent Decision-making, Flexibility, Integration, Numerical Control, Computer Vision (CV), Industrial Intelligence, Intelligent Systems, Driverless, Intelligent Operations and Maintenance, Natural Language Processing (NLP)
	Blockchain Technology (BC)	Blockchain, Digital Currency, Virtual Currency, Differential Privacy Technology, Distributed Communication, Consensus Mechanism, Distributed Control Systems (DCS), Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Distributed Computing, Peer-to-Peer Distributed Technology (P2P), Intelligent Financial Contracts
	Cloud Computing Technology (CC)	Cloud Computing, Edge Infrastructure, Neuromorphic Computing, Server Virtualization, Cognitive Computing, Fusion Architecture, Service-Oriented Architecture (SOA), EB-level Storage, Fog Computing, Haze Computing, Edge Computing, Multi-party Secure Computing, Memory Computing, Internet of Things (IoT), Cloud Storage, Cloud Networking, Cloud Platforms, Cloud Migration, Cloud Computing Technologies, Cloud Light, Billion-level Concurrent Systems, Programming Models
	Big Data Technology (BD)	Big Data (BD), Industrial Data, Augmented Reality, Relational Databases, Heterogeneous Data, Credit Reporting, Data Exchange, Virtual Reality, Multi-level Data Compression, Geographic Information Systems (GIS), Cyber-Physical Systems, Mixed Reality, Databases (Oracle), Data Empowerment, Data Visualization, Data Mining, Data Acquisition, Data Cleaning, Data-Driven, Data Storage, Data Industry, Digital Transformation, Data Technology
	Digital Technology Application (DTA)	Medical Digitalization, Mobile Wallets, Barcode Payments, NFC Payments, Internet Finance, Smart Devices, Digital Control, Intelligent Workshops, Smart Terminal Products, Financial Technology, Smart Energy, Network Connection, Quantitative Finance, Smart Energy Conservation and Environmental Protection, Smart Healthcare, Unmanned Retail, Smart Home, Smart Investment Consulting, Smart Cultural Tourism, Smart Power Systems, Digital Retail, Digital Finance, Smart Logistics, Open Banking, Electronic Medical Records (EMR), New Retail, Intelligent Customer Service, B2B, B2C, C2B, C2C, C2M, Online-to-Offline (O2O), Online Retail, E-commerce, Official Accounts, WeChat Mini Programs, Applications (APP), Live Streaming, Microblogs, Mobile E-commerce (M2M), Presale, Online Office, Online Education, Telemedicine, Unmanned Delivery

.

3.2.4. Moderating Variable

Media Attention (Media). The degree of media attention for an event can be measured by the total amount of news related to the event in media reports [39]. existing literature presents various methods for measuring media attention. This study, following the research approach of Mao Zhihong et al. (2024) [40], uses the quantity of media coverage of a company as a measure of media attention and applies a logarithmic transformation to this indicator as a proxy for media attention.

3.2.5. Control Variables

To control for other influencing factors, this study adopts definitions from the existing literature to control variables such as company size (Size), financial leverage (Lev), growth (Growth), operating cash flow (OCF), company age (Age), proportion of independent directors (BInd), ownership concentration (Top1), and executive compensation (Salary). Additionally, industry and annual dummy variables are controlled for. The definitions of each variable are outlined in

Table 4. Sino-securities ESG Environmental Responsibility Sub-Indicators (source: authors’ own creation).

Type	Name	Symbol	Definition
Dependent Variable	Enterprise Green Innovation	GI	Natural logarithm of the total number of patents applied for by the company in the current year
Independent Variable	Enterprise Digital Transformation	DT	Natural logarithm of the sum of frequencies of DIG-related keywords, plus 1
Mediating Variable	Corporate Environmental Responsibility	CER	Environmental responsibility sub-rating in the Sino-securities ESG assessment system
Moderating Variable	Media Attention	Media	Logarithm of the quantity of media coverage received by the company
Mediating Variable	Company Size	Size	Natural logarithm of total assets
	Financial Leverage	Lev	Debt-to-asset ratio (total liabilities divided by total assets)
	Return on Total Assets	ROA	Net profit/total assets
	Growth	Growth	Revenue growth rate
	Operating Cash Flow	OCF	Operating cash flow/total assets
	Company Age	Age	Difference between the current year and the year of establishment
	Proportion of Independent Directors	BInd	Number of independent directors/total board members
	Ownership Concentration	Top1	Largest shareholder/total shares outstanding
	Executive Compensation	Salary	Natural logarithm of the total executive compensation

.

3.3. Model Construction

3.3.1. Main Effects Model

To verify the impact of DT on the GI of manufacturing enterprises in China, this study constructs the following baseline regression model for hypothesis testing:

$${GI}_{it}={\alpha }_0+{\alpha }_1{DT}_{it}+{\alpha }_2{Controls}_{it}+\sum industry+\sum year+\epsilon$$

(1)

Here, GI is the dependent variable representing the green innovation capability of the enterprise, DT is the independent variable representing digital transformation, Controls represent the control variables as described earlier, industry denotes industry fixed effects, year represents time fixed effects, ε denotes the random disturbance term, subscript i indicates the enterprise, and subscript t indicates the year. In the above main effects model, if the expected test result shows that the coefficient α₁ of DT is positive, it indicates that DT has a positive promotion effect on green innovation in manufacturing enterprises, thereby verifying Hypothesis 1.

3.3.2. Mediation Effects Model

Building upon the baseline regression model, this study adopts a mediation effects testing method proposed by Wen Zhonglin and Ye Baojuan (2014) [41] to examine the mediation effects of CER on DT and GI, using a stepwise regression approach. The specific model is as follows:

$${GI}_{it}={\alpha }_0+{\alpha }_1{DT}_{it}+{\alpha }_2{Controls}_{it}+\sum industry+\sum year+\epsilon$$

$${CER}_{it}={\gamma }_0+{\gamma }_1{DT}_{it}+{\gamma }_2{Controls}_{it}+\sum industry+\sum year+\epsilon$$

(2)

$${GI}_{it}={\delta }_0+{\delta }_1{DT}_{it}+{\delta }_2{CER}_{it}+{\delta }_3{Controls}_{it}+\sum industry+\sum year+\epsilon$$

(3)

In the mediation effects testing model, GI is the dependent variable representing the green innovation capability of the enterprise, DT is the independent variable representing digital transformation, CER is the mediator variable representing corporate environmental responsibility, the other control variables remain consistent with previous sections, industry denotes industry fixed effects, and year represents time fixed effects. Model (3.2) primarily aims to verify the impact of DT on the mediator variable CER, while model (3.3) examines whether the mediator variable, CER, acts as a mediator.

3.3.3. Moderation Effects Model

To verify Hypothesis 3 proposed by this study and delve deeper into the moderating effect of media attention (Media), this study constructs the following moderation effects model for testing:

$${GI}_{it}={\alpha }_0+{\alpha }_1{DT}_{it}+{\alpha }_2{Controls}_{it}+\sum industry+\sum year+\epsilon$$

$${GI}_{it}={\beta }_0+{\beta }_1{DT}_{it}+{\beta }_2{Media}_{it}+{\beta }_3{DIG}_{it}\times {Media}_{it}+{\beta }_4{Controls}_{it}+\sum industry+\sum year+\epsilon$$

(4)

In the moderation effects testing model, GI is the dependent variable representing the green innovation capability of the enterprise, DT is the independent variable representing digital transformation, Media is the moderator variable representing media attention, the other control variables remain consistent with previous sections, industry denotes industry fixed effects, and year represents time fixed effects. This study investigates the moderating role of media attention (Media) between DT and GI through the interaction term DT × Media.

4. Results

4.1. Descriptive Statistics and Correlation Analysis

4.1.1. Descriptive Statistics

Using Stata18 software, this study conducted descriptive statistical analysis on variables such as enterprise DT and GI. The results are presented in

Table 5. Descriptive statistics (source: authors’ own creation).

Variables	N	Mean	Median	Std. Dev	Minimum	Maximum
GI	21,842	0.823	0	1.103	0	4.29
DT	21,842	1.345	1.099	1.271	0	4.963
Size	21,842	22.022	21.863	1.19	19.303	27.145
Lev	21,842	0.388	0.374	0.198	0.051	1.034
ROA	21,842	0.04	0.042	0.071	−0.36	0.211
Growth	21,842	0.175	0.108	0.436	−0.653	3.261
OCF	21,842	0.05	0.049	0.069	−0.204	0.255
Age	21,842	2.929	2.944	0.31	1.386	4.174
BInd	21,842	0.377	0.364	0.054	0.3	0.571
Top1	21,842	0.33	0.308	0.14	0.084	0.743
Salary	21,842	15.354	15.328	0.728	13.005	17.403

. From the data in the table, the analysis comprises 21,842 observations. Among all observed enterprises, the mean of GI is 0.823, with a standard deviation of 1.103, minimum value of 0, and maximum value of 4.29. This indicates significant variability in GI performance among manufacturing enterprises, reflecting differing levels of investment and emphasis on GI, showcasing disparities in GI concepts and practices across enterprises. The mean and standard deviation of enterprise DT are 1.345 and 1.271, respectively, with a range from 0 to 4.963, indicating substantial variation in digital transformation among manufacturing enterprises.

4.1.2. Correlation Analysis

Before conducting regression analysis, it is essential to explore potential relationships between the main variables. Through correlation coefficient testing, preliminary insights into variable relationships can be obtained. According to the test results shown in

Table 6. Correlation statistics (source: authors’ own creation).

Variables	GI	DT	Size	Lev	ROA	Growth	OCF	Age	BInd	Top1	Salary
GI	1
DT	0.234 ***	1
Size	0.490 ***	0.117 ***	1
Lev	0.273 ***	0.014 **	0.432 ***	1
ROA	−0.014 **	−0.00500	0.026 ***	−0.410 ***	1
Growth	0.039 ***	0.016 **	0.066 ***	0.040 ***	0.229 ***	1
OCF	−0.00600	−0.0100	0.099 ***	−0.190 ***	0.434 ***	0.015 **	1
Age	0.040 ***	0.019 ***	0.155 ***	0.131 ***	−0.062 ***	−0.057 ***	0.022 ***	1
BInd	−0.00600	0.061 ***	−0.020 ***	−0.00100	−0.029 ***	−0.00900	−0.00400	−0.0110	1
Top1	−0.00600	−0.046 ***	0.088 ***	−0.044 ***	0.160 ***	−0.00600	0.117 ***	−0.070 ***	0.047 ***	1
Salary	0.286 ***	0.209 ***	0.507 ***	0.050 ***	0.215 ***	0.046 ***	0.213 ***	0.104 ***	−0.075 ***	−0.013 *	1

t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

, the correlation coefficient between enterprise DT and GI in manufacturing is significant at the 1% level, with a value of 0.234, providing initial evidence that enterprise DT promotes GI. Additionally, significant correlations are observed between enterprise DT and GI with other control variables. To further investigate the impact of DT on GI in manufacturing enterprises, regression analysis is necessary.

4.2. Regression Results Analysis

4.2.1. Main Effects Test Analysis

Regarding the relationship between enterprise DT and GI, this study employs a stepwise regression method to empirically test this relationship. The test results are presented in

Table 7. Regression analysis of DIG’s impact on GI in manufacturing enterprises (source: authors’ own creation).

Variables	(1)	(2)
	GI	GI
DT	0.166 ***	0.118 ***
	(28.21)	(22.77)
Size		0.398 ***
		(57.45)
Lev		0.396 ***
		(10.03)
ROA		0.391 ***
		(3.57)
Growth		−0.039 ***
		(−2.65)
OCF		−0.349 ***
		(−3.45)
Age		−0.099 ***
		(−4.67)
BInd		−0.0530
		(−0.47)
Top1		−0.199 ***
		(−4.44)
Salary		0.055 ***
		(5.05)
Cons	−0.116 ***	−9.351 ***
	(−3.35)	(−54.40)
Ind	YES	YES
Year	YES	YES
N	21,842	21,842
$\mathrm{Adj}\text{-}{R}^2$	0.121	0.340

t-statistics in parentheses. *** p < 0.01.

. In column (1), the preliminary effect of DT on GI is observed, with a coefficient of 0.166, statistically significant at the 1% level. However, this result does not account for other potential influencing factors, namely the impact of control variables. To more accurately assess the true effect of DT on GI, control variables are included in column (2). After adjusting for control variables, the coefficient of DT becomes 0.118, slightly lower numerically but still significant at the 1% level. This result indicates that even after considering other potential influencing factors, DT continues to have a significant positive impact on GI in manufacturing enterprises.

4.2.2. Mediation Effects Test Analysis

For testing the mediation effect of CER between DT and enterprise GI development, the results are presented in

Table 8. Mediation effects of environmental responsibility (source: authors’ own creation).

Variables	(1)	(2)	(3)
	GI	CER	GI
DT	0.121 ***	0.057 ***	0.116 ***
	(22.68)	(8.57)	(21.90)
CER			0.081 ***
			(14.56)
Size	0.401 ***	0.224 ***	0.383 ***
	(56.00)	(25.19)	(52.96)
Lev	0.407 ***	0.288 ***	0.384 ***
	(9.95)	(5.67)	(9.42)
ROA	0.331 ***	0.589 ***	0.283 **
	(2.84)	(4.08)	(2.44)
Growth	−0.042 ***	−0.188 ***	−0.027 *
	(−2.74)	(−9.89)	(−1.76)
OCF	−0.379 ***	0.324 **	−0.405 ***
	(−3.60)	(2.48)	(−3.87)
Age	−0.111 ***	−0.063 **	−0.106 ***
	(−5.05)	(−2.30)	(−4.84)
BInd	−0.041	−0.244 *	−0.0210
	(−0.35)	(−1.69)	(−0.18)
Top1	−0.224 ***	0.00300	−0.225 ***
	(−4.90)	(0.06)	(−4.93)
Salary	0.054 ***	0.048 ***	0.050 ***
	(4.80)	(3.45)	(4.47)
Cons	−9.372 ***	−3.702 ***	−9.073 ***
	(−52.65)	(−16.79)	(−50.89)
Ind	YES	YES	YES
Year	YES	YES	YES
N	21,015	21,015	21,015
$\mathrm{Adj}\text{-}{R}^2$	0.341	0.0920	0.348

t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

. From column (1), it is evident that DT significantly enhances environmental responsibility. In column (2), the coefficient of CER is 0.057, significantly positive, indicating that environmental responsibility significantly enhances enterprise GI. Moreover, in column (3), the coefficient of DT remains significantly positive, demonstrating the positive impact of environmental responsibility on enterprise GI. However, compared to the coefficient of DT in column (1), there is a slight decrease in column (3), indicating that CER partially mediates the impact of DT on GI in manufacturing enterprises by enhancing CER, thereby promoting the enhancement of enterprise GI.

4.2.3. Moderation Analysis of Media Attention

To examine the moderating effect of media attention on the relationship between DT and GI, we verified this by examining the sign of DT × Media, with the final results presented in

Table 9. Moderating effect of media attention on DT and GI (source: authors’ own creation).

Variables	(1)	(2)
	GI	GI
DT	−0.0160	0.00700
	(−0.54)	(0.26)
Media	0.255 ***	0.000
	(22.67)	(−0.01)
DT × Media	0.030 ***	0.022 ***
	(5.18)	(4.35)
Size		0.391 ***
		(52.84)
Lev		0.389 ***
		(9.78)
ROA		0.362 ***
		(3.27)
Growth		−0.040 ***
		(−2.68)
OCF		−0.395 ***
		(−3.87)
Age		−0.095 ***
		(−4.41)
BInd		−0.0690
		(−0.60)
Top1		−0.200 ***
		(−4.43)
Salary		0.051 ***
		(4.61)
Cons	−1.504 ***	−9.135 ***
	(−22.62)	(−50.45)
Ind	YES	YES
Year	YES	YES
N	28,388	28,388
$\mathrm{Adj}\text{-}{R}^2$	0.179	0.341

t-statistics in parentheses. *** p < 0.01.

. In column (2), the coefficient of DT × Media is 0.022, which is statistically significant at the 1% level. This finding reveals the important role of media attention in enhancing the impact of DT on GI.

4.3. Robustness Checks

4.3.1. Lagged Regression Analysis

Considering that the performance improvement brought about by current-period corporate GI significantly increases the resources a firm possesses, thereby affecting its digital investments, this study needs to address potential endogeneity issues when exploring the impact of DT on manufacturing companies’ GI. For the accuracy of results, this study lagged the explanatory variables by one period. The results are shown in

Table 10. Lagged regression analysis (source: authors’ own creation).

Variables	(1)	(2)
	GI	GI
DTit−1	0.173 ***	0.123 ***
	(26.18)	(21.07)
Size		0.411 ***
		(53.59)
Lev		0.394 ***
		(8.98)
ROA		0.390 ***
		(3.26)
Growth		−0.030 *
		(−1.80)
OCF		−0.432 ***
		(−3.80)
Age		−0.117 ***
		(−4.65)
BInd		−0.0550
		(−0.44)
Top1		−0.209 ***
		(−4.14)
Salary		0.062 ***
		(5.11)
Cons	−0.0240	−9.646 ***
	(−0.64)	(−49.44)
Ind	YES	YES
Year	YES	YES
N	18,326	18,326
$\mathrm{Adj}\text{-}{R}^2$	0.125	0.345

t-statistics in parentheses. *** p < 0.01, * p < 0.1.

. In column (2), the coefficient of lagged DT (DT_it−1) is also statistically significant at the 1% level, being 0.123. Comparative analysis shows no significant differences from the previous findings, thus confirming the robustness of this study’s results.

4.3.2. Replacement of Dependent Variables

This study measures the dependent variable GI as the natural logarithm of the number of green patents applied for by the enterprise. For robustness checks, we use the natural logarithm of the total number of green patents granted to the enterprise to measure corporate GI. The results are shown in

Table 11. Replacement of dependent variables (source: authors’ own creation).

Variables	(1)	(2)
	GI1	GI1
DT	0.122 ***	0.079 ***
	(23.00)	(16.83)
Size		0.345 ***
		(55.01)
Lev		0.382 ***
		(10.70)
ROA		0.127
		(1.28)
Growth		−0.064 ***
		(−4.83)
OCF		−0.207 **
		(−2.27)
Age		−0.095 ***
		(−4.94)
BInd		0.0770
		(0.75)
Top1		−0.198 ***
		(−4.91)
Salary		0.058 ***
		(5.92)
Cons	−0.072 **	−8.239 ***
	(−2.32)	(−53.05)
Ind	YES	YES
Year	YES	YES
N	21,842	21842
$\mathrm{Adj}\text{-}{R}^2$	0.108	0.321

t-statistics in parentheses. *** p < 0.01, ** p < 0.05.

. In column (2), the coefficient of DT is statistically significant at the 1% level, being 0.079. Comparative analysis indicates no significant differences from previous results, thus confirming the robustness of this study’s findings.

4.3.3. Exclusion of the Impact of the COVID-19 Pandemic

Considering the impact of the pandemic on companies’ financing and operational capabilities, which may in turn affect their performance and level of DT, this study excludes samples from the years 2020 and 2021 to address these issues. The results are presented in

Table 12. Exclusion of the impact of the COVID-19 pandemic (source: authors’ own creation).

Variables	(1)	(2)
	GI	GI
DT	0.156 ***	0.113 ***
	(23.14)	(18.73)
Size		0.375 ***
		(47.25)
Lev		0.382 ***
		(8.53)
ROA		0.442 ***
		(3.52)
Growth		−0.0230
		(−1.34)
OCF		−0.358 ***
		(−3.12)
Age		−0.111 ***
		(−4.61)
BInd		−0.00900
		(−0.07)
Top1		−0.181 ***
		(−3.55)
Salary		0.058 ***
		(4.70)
Cons	−0.086 **	−8.841 ***
	(−2.39)	(−45.24)
Ind	YES	YES
Year	YES	YES
N	16,324	16,324
$\mathrm{Adj}\text{-}{R}^2$	0.112	0.320

t-statistics in parentheses. *** p < 0.01, ** p < 0.05.

. From the data in column (2), the coefficient of DT is statistically significant at the 1% level, being 0.113. Comparative analysis shows no significant differences from previous findings, further validating the robustness of this study’s results.

4.4. Heterogeneity Analysis

4.4.1. Analysis Based on Institutional Ownership

Given that both digitalization and corporate innovation are long-term investments that do not yield immediate economic benefits, long-term investors are necessary. Institutional investors, compared to retail investors, can provide more professional investment perspectives to support companies’ future long-term investments. Therefore, we anticipate that digitalization will more effectively promote corporate GI in samples with higher institutional ownership. We group the data based on the median institutional ownership into high and low institutional ownership samples. The results are presented in

Table 13. Analysis based on institutional ownership (source: authors’ own creation).

Variables	(1)	(2)
	GI	GI
	Low Institutional Ownership	High Institutional Ownership
DT	0.114 ***	0.125 ***
	(17.02)	(15.58)
Size	0.335 ***	0.427 ***
	(30.57)	(44.13)
Lev	0.433 ***	0.361 ***
	(8.51)	(5.90)
ROA	0.366 ***	0.494 ***
	(2.71)	(2.75)
Growth	−0.042 **	−0.0320
	(−2.03)	(−1.53)
OCF	−0.437 ***	−0.310 **
	(−3.27)	(−2.02)
Age	−0.114 ***	−0.099 ***
	(−4.01)	(−3.13)
BInd	−0.122	0.00100
	(−0.81)	(0.01)
Top1	−0.416 ***	−0.0920
	(−5.92)	(−1.36)
Salary	0.098 ***	0.027 *
	(6.38)	(1.77)
Cons	−8.422 ***	−9.711 ***
	(−31.93)	(−39.37)
Ind	YES	YES
Year	YES	YES
N	10,910	10,932
$\mathrm{Adj}\text{-}{R}^2$	0.269	0.369

t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

. Regression results reveal that in samples with higher institutional ownership, digitalization more effectively promotes corporate innovation.

4.4.2. Heterogeneity Analysis Based on Analyst Attention

Analysts possess expertise in assessing the impact of DT on future developments, and thus, analyst attention can significantly reduce the financing constraints faced by companies regarding DT, promoting its positive impact. We measure analyst attention by the number of analysts tracking a firm. We group the data based on the median number of analysts into samples with high and low analyst attention. The results are presented in

Table 14. Heterogeneity analysis based on analyst attention (source: authors’ own creation).

Variables	(1)	(2)
	GI	GI
	Low Analyst Attention	High Analyst Attention
DT	0.111 ***	0.131 ***
	(12.87)	(12.40)
Size	0.375 ***	0.421 ***
	(30.23)	(28.22)
Lev	0.551 ***	0.762 ***
	(7.91)	(7.62)
ROA	0.395 *	0.658 *
	(1.91)	(1.89)
Growth	−0.058 **	−0.087 **
	(−2.45)	(−2.39)
OCF	−0.248	−0.825 ***
	(−1.43)	(−3.46)
Age	−0.095 ***	−0.0730
	(−2.71)	(−1.62)
BInd	−0.087	0.0130
	(−0.44)	(0.06)
Top1	−0.165 **	−0.245 ***
	(−2.23)	(−2.64)
Salary	0.038 **	0.058 ***
	(2.04)	(2.60)
Cons	−8.651 ***	−10.352 ***
	(−26.78)	(−28.52)
Ind	YES	YES
Year	YES	YES
N	8126	6124
$\mathrm{Adj}\text{-}{R}^2$	0.299	0.396

t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

. Regression results reveal that in samples with higher analyst attention, digitalization more effectively promotes corporate innovation.

4.5. Incentive Effects of Dual Carbon Targets

To investigate the impact of DT on GI under the context of dual carbon targets, this study uses the year 2020 when China officially proposed dual carbon targets as a temporal point. Further analysis compares data from the years before and after the proposal of dual carbon targets—specifically, 2018–2019 and 2021–2022. The results are presented in

Table 15. Regression analysis of samples from 2018 to 2019 and 2021 to 2022 (source: authors’ own creation).

Variables	(1)	(2)	(3)	(4)
	GIx (2018–2019)		GIy (2021–2022)
DT	0.185 ***	0.123 ***	0.154 ***	0.100 ***
	(14.13)	(10.71)	(14.44)	(10.86)
Size		0.435 ***		0.432 ***
		(27.35)		(34.55)
Lev		0.447 ***		0.501 ***
		(4.82)		(6.81)
ROA		0.337		0.121
		(1.63)		(0.58)
Growth		0.0170		−0.089 ***
		(0.43)		(−3.23)
OCF		−0.576 **		−0.0680
		(−2.42)		(−0.37)
Age		−0.109 **		−0.0500
		(−2.09)		(−1.30)
BInd		−0.229		−0.260
		(−0.88)		(−1.27)
Top1		−0.285 ***		0.0150
		(−2.70)		(0.18)
Salary		0.048 *		0.000
		(1.88)		(−0.01)
Cons	0.0120	−9.857 ***	0.173 ***	−9.223 ***
	(0.20)	(−24.01)	(3.54)	(−28.77)
Ind	YES	YES	YES	YES
Year	YES	YES	YES	YES
N	4434	4434	6224	6224
$\mathrm{Adj}\text{-}{R}^2$	0.110	0.344	0.115	0.372

t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

. The coefficients of DT in columns (1) to (4) are all statistically significant at the 1% level, albeit with differences. Columns (1) and (3) show results without controlling for covariates, while columns (2) and (4) show results after controlling for covariates. In column (4), the coefficient of DT is 0.100, which is lower than 0.123 in column (2), indicating that under the dual carbon context, the driving effect of DT on manufacturing GI seems to fall short of expectations. This finding differs from our initial expectations and could be explained by two factors. Firstly, the incentive effect of dual carbon targets may not be apparent due to its recent introduction, necessitating further examination over time. Secondly, following the proposal of dual carbon targets, the global outbreak of the COVID-19 pandemic led China to adopt silent measures for public safety, causing disruptions in normal business operations during the pandemic and affecting companies’ DT processes and GI.

5. Conclusions and Prospect

5.1. Conclusions

Through a literature review, referencing relevant scholars’ research, and drawing upon innovation theory, resource-based theory, stakeholder theory, and sustainable development theory, this study thoroughly analyzes the impact of DT on the development of manufacturing companies’ GI using CER as a mediator. This research path reveals that DT enhances CER, thereby positively impacting GI. Empirical analysis reveals the following:

(1) The higher the level of DT in manufacturing, the better the development of corporate GI.

Main effect tests show a significant positive impact of DT on corporate GI development. On one hand, the application of digital innovation technologies such as Artificial Intelligence (AI) enables companies to efficiently collect, process, and analyze large amounts of data, reducing energy consumption and waste generation, mitigating environmental risks in production processes, and enabling green production. On the other hand, DT enhances companies’ innovation capabilities; through data analysis and predictive technologies, companies can pinpoint opportunities and demands for GI precisely. DT not only enhances companies’ competitiveness and market position but also drives corporate GI development, aiding companies in achieving environmentally sustainable development.

(2) The better the fulfillment of CER in manufacturing, the stronger the development of corporate GI.

Through an in-depth analysis of environmental responsibility and GI development in manufacturing companies, a positive correlation between these two aspects was discovered. The results of the mediation analysis indicate that CER partially mediates the relationship between DT and GI development, further elucidating the nature of this influence. DT can enhance corporate transparency, reduce information asymmetry, and promote cooperation and sharing among enterprises. By leveraging digital platforms and cloud computing technologies, companies can achieve real-time data sharing and collaboration with suppliers, customers, and other stakeholders. This collaboration promotes resource sharing and recycling, reduces resource waste, reflects the fulfillment of CER, increases stakeholder trust, enhances corporate reputation, and empowers GI development.

(3) Media attention positively moderates the promotional effect of DT on GI development.

The media holds significant influence and dissemination power in society, capable of conveying a company’s DT and GI initiatives to a broader audience. When a company’s DT and GI efforts receive media attention and coverage, they gain greater exposure and recognition, enhancing corporate visibility and image and attracting more attention and investment. This attention can bring additional resources and support for GI development, further driving its progress.

(4) Differences exist in the promotional effects of DT on GI development based on varying institutional ownership proportions and analyst attention.

When manufacturing companies have a higher proportion of institutional investors, these investors typically pay more attention to and value the company’s DT and GI initiatives. Institutional investors often possess more resources and expertise to assess a company’s strategies and sustainable development. They are more inclined to invest in companies with GI potential and support their DT efforts to achieve higher returns and long-term growth. Analysts typically conduct in-depth research on and analysis of manufacturing companies, providing relevant reports and recommendations. When analysts give a manufacturing company a higher rating and attention, it attracts more investors and stakeholders. Positive evaluations and attention from analysts can enhance a company’s reputation and image, increasing the promotional effect of DT on GI development.

5.2. Management Insights

For a long time, people’s over-reliance on traditional economic growth models has resulted in serious ecological problems. Overexploitation of resources and high pollution emissions have led to pollution of air, water, and soil; reduced biodiversity; and exacerbated climate change. GI, as a path of innovation with an environmental orientation at its core, has profound implications for promoting balanced development between the economy and ecology. Actively promoting GI not only effectively addresses resource wastage and environmental pollution issues arising from economic development but also promotes sustainable economic development while protecting the ecological environment. Research findings indicate that manufacturing firms in China play a promoting role in GI development through the process of DT. Furthermore, this promotional effect can be transmitted through CER. Additionally, media attention positively moderates this relationship. These research findings provide the following management insights for manufacturing firms in advancing DT, enhancing environmental responsibility awareness, and realizing GI development in practice.

(1) Government Perspective

The government plays a crucial role in promoting DT and GI. Firstly, it is essential to improve digital infrastructure and related facilities and increase policy support and investment in enterprise DT, thereby promoting sustained and healthy economic and social development. Secondly, the government should enact corresponding policies and regulations to ensure corporate compliance with CER. Establishing environmental responsibility and penalty mechanisms is necessary to penalize companies that violate environmental regulations and standards, holding them accountable for their environmental responsibilities. Thirdly, the government should establish differentiated subsidy policies tailored to the characteristics of manufacturing enterprises to promote DT and GI.

(2) Enterprise Perspective

Enterprises should stay abreast of the times, immerse themselves in the wave of DT, and continuously enhance their level of digitalization. Improving digitalization will better equip enterprises to adapt to market changes, explore new business opportunities, and enhance their core competitiveness. Secondly, enterprises should enhance their fulfillment of CER. CER is a crucial pathway for enhancing GI during the DT process. With increasing societal attention to environmental protection, enterprises need to actively fulfill environmental responsibilities through green production and operations using digital technologies. Thirdly, enterprises should prioritize GI development. As indicated by descriptive statistical analysis, there are varying levels of emphasis on GI among manufacturing enterprises in China. Some companies still have a relatively low emphasis on GI and continue to use traditional production methods and technologies, leading to resource wastage and increased environmental burdens. Enterprises should conscientiously enhance environmental awareness, actively engage in DT and GI to adapt to changing market demands and drive Chinese manufacturing enterprises towards greener and more sustainable development.

5.3. Research Limitations and Prospect

This study explores the impact of DT on enterprise GI development using panel data from China’s manufacturing sector from 2013 to 2022, incorporating environmental responsibility as a mediating variable and media attention as a moderating variable. The empirical results offer insights for other researchers and enterprises, yet there are some limitations that future research could address for further improvement.

Firstly, concerning panel data, this study only covers the years 2013 to 2022 and does not encompass the latest data from all listed manufacturing enterprises in China. Additionally, due to limited time post-proposal of dual carbon targets, relevant enterprise data are sparse. Future research could expand the timeframe of data collection to include more recent years, thereby increasing the sample size and further analyzing the specific impact of DT on GI under the dual carbon context.

Secondly, in practical applications, although this study provides a reference for China’s manufacturing enterprises in DT and GI, specific practical applications require detailed analysis and operations tailored to individual enterprise circumstances. Therefore, future research could delve into the specific application of DT and GI in different types, sizes, and regions of manufacturing enterprises, comprehensively revealing the mechanisms of enterprise GI development and offering more specific and practical guidance for enterprises.