Introduction
With over 40 years of opening up and development, China’s industry, especially the manufacturing industry, has achieved considerable development. In 1990, China’s manufacturing industry accounted for only 2.7% of the global manufacturing industry, ranking ninth in the world. By 2000, this proportion had risen to 6%, ranking fourth in the world. In 2010, China’s manufacturing sector accounted for 19.8% of the global manufacturing industry, ranking first in the world. However, volume is not the only standard to measure the development of the manufacturing industry. The efficiency of China’s manufacturing industry is lower than that of Europe, the United States, Japan, or South Korea. While integrating into the global value chain, China mainly engages in labor-intensive industries at the low end of the industrial chain transferred by developed countries (Liu, Reference Liu2016). With the transfer of industry chains with high energy consumption and high pollution, combined with China’s primary focus on development speed with relatively weak pollution control ability, industrial structure imbalance and serious environmental pollution have been incurred in the rapid expansion of the manufacturing industry. Therefore, a shift in the growth mode is needed (Ceglowski & Golub, Reference Ceglowski and Golub2012). Now, China, a major manufacturing country, has entered a new stage of high-quality development. Whether China could transform from ‘large’ to ‘strong’ is dependent upon the fulfillment of high-quality economic development, including mid and high-end development goals, such as green development of the manufacturing industry.
What is green transformation? Green transformation aims to replace the crude development mode of high emission, high pollution, and environmental destruction; reconstruct the harmonious relationship between human and nature through green development approaches, characterized by low emission and low consumption; and promote sustainable development (Zhou, Li, Lu, & Lu, Reference Zhou, Li, Lu and Lu2022). The green transformation of companies, as a key support of economic development and environmental protection, is not only an effective starting point for the country to promote economic green transformation, but also a key concern for the formulation of green economic policies (Dechezleprêtre, Kozluk, Kruse, Nachtigall, & De Serres, Reference Dechezleprêtre, Kozluk, Kruse, Nachtigall and De Serres2019). From the perspective of external regulation of companies, government regulatory behavior is a powerful motivation and source of pressure for the green transformation of manufacturing companies. In terms of regulation, pressure-driven environmental policies such as environmental regulations (Hermwille, Obergassel, Ott, & Beuermann, Reference Hermwille, Obergassel, Ott and Beuermann2017), social responsibility disclosure (Xiang, Liu, Yang, & Zhao, Reference Xiang, Liu, Yang and Zhao2020), and environmental taxation can effectively guide companies’ green behaviors (Fang, Kong, Sensoy, Cui, & Cheng, Reference Fang, Kong, Sensoy, Cui and Cheng2021), raise their environmental awareness, and promote the R&D of low-carbon and emission reduction technologies; in terms of incentives, the government alleviates the environmental pressure brought by regulation through green subsidies and energy-saving consumption incentives (Han, Mao, Yu, & Yang, Reference Han, Mao, Yu and Yang2024), and enhances the willingness of companies to explore green transformation. However, at present, the green transformation of manufacturing companies in China still faces two prominent internal problems: First, insufficient transformation motivation. Companies, as the main profit-seeking body, generally believe that environmental protection investment will crowd out productive investment, lacking the mindsets of integrating the concept of green and low-carbon development into the overall strategy of the company, and are unwilling to take the initiative to invest in green entrepreneurial opportunities. Thus, it is difficult to cope with the contradiction between the environment and economic development, i.e., it is hard to produce effective incentives for green transformation of companies with a low level of orientation to green entrepreneurship. Second, insufficient transformation capability. Green innovation is an important approach for the manufacturing industry to realize green transformation. However, problems such as insufficient innovation resources, weak R&D capacity, low innovation efficiency, and other problems are prevailing in China’s manufacturing companies, which make the development of green innovation lagging behind, i.e., green innovation capacity fails to support green transformation (Peng, Jia, & Zou, Reference Peng, Jia and Zou2022).
With the emergence and application of digital technologies such as cloud computing, artificial intelligence, and the Internet of Things (IoT), digitalization has become a new engine for industrial optimization and upgrading (Wang & Shao, Reference Wang and Shao2024). Empowered by digitalization, the traditional manufacturing industry could improve its R&D and innovation ability, transform the management and operation mode, and improve the efficiency of production factors, etc., and at the same time, it can make the production and operation process more ‘automated’ and ‘refined’, accurately control the material inputs, and reduce waste and consumption of materials in production. In addition, it also dramatically reduces the cost of green transformation, transforming digitalization into a driving force for the green development of manufacturing companies. Does digitization enhance the green transformation of manufacturing companies? How to promote the green transformation of companies through digitization? These are important propositions that need to be addressed urgently, and are also the starting point of this article.
Currently, there are fewer studies on the impact of digitization on the green transformation of manufacturing companies, and the studies on digitization mainly focus on two areas: First, the measurement methods of digitization and, at the micro level, density of industrial robots, digital process valuation (Ferreira, Fernandes, & Ferreira, Reference Ferreira, Fernandes and Ferreira2019), amount of investment in digital transformation projects (Hu, Li, Zhang, & Wang, Reference Hu, Li, Zhang and Wang2022), and text analysis methods used to measure the digitization level of companies (Wang, Shao, Song, Shao, & Wang, Reference Wang, Shao, Song, Shao and Wang2023). The second is the economic effect of digitization, which on the one hand focuses on the internal performance improvement of companies through digitalization, including but not limited to optimizing the ratio of factor inputs (Liu, Liu, & Gong, Reference Liu, Liu and Gong2023), saving labor costs and sales costs (Banalieva & Dhanaraj, Reference Banalieva and Dhanaraj2019), improving innovation performance (Ghasemaghaei & Calic, Reference Ghasemaghaei and Calic2019), improving organizational performance (Wamba, Dubey, Gunasekaran, & Akter, Reference Wamba, Dubey, Gunasekaran and Akter2020), and boosting total factor productivity (Yu, Xu, Zhou, & Chen, Reference Yu, Xu, Zhou and Chen2024); on the other hand, it explores how digitalization drives a company’s external development, such as improving client relationship management (Kamalaldin, Linde, Sjödin, & Parida, Reference Kamalaldin, Linde, Sjödin and Parida2020), easing financing pressure (Li, Lian, & Xu, Reference Li, Lian and Xu2023), and enhancing value chain position (Feng, Zhang, Di, & Li, Reference Feng, Zhang, Di and Li2024).
Previous research on the economic effects of digitalization provides a logical basis for this study, as digitalization has penetrated into all stages of company production and operation, but there are fewer studies that analyze in-depth how digitalization can comprehensively promote the green transformation of companies in production and operation. The value of digitization is not only reflected in the economic effects but also in the environmental value and other noneconomic benefits. Existing microlevel research on the role of digitalization in green transformation focuses on the impact of digitization on green innovation of companies with different research conclusions. On the one hand, companies will form an interconnected industrial chain in digital transformation, optimize the allocation of resource factors, improve production and operation efficiency, and at the same time, smooth information transmission channels, which ultimately ‘empower’ the green innovation of companies (Mubarak, Tiwari, Petraite, Mubarik, & Rasi, Reference Mubarak, Tiwari, Petraite, Mubarik and Rasi2021). On the other hand, while deploying and implementing digital transformation strategies, over-allocation of limited resources to digital transformation may crowd out green innovation R&D funds, which will not be conducive to the green innovation activities of companies (Huang, Wang, Wang, & Yarovaya, Reference Huang, Wang, Wang and Yarovaya2023). In addition, existing studies have applied one-sided indicators such as green innovation and green R&D capability to replace company green transformation indicators, which is difficult to fully reflect the green transformation of companies. Among them, green innovation is only a key capability or intermediate link in the green transformation of companies, but not the equivalent (Zhou et al., Reference Zhou, Li, Lu and Lu2022), and whether it can ultimately realize the win-win outcome of green development needs to be further explored.
This study’s contributions are as follows: We examined the mechanism by which digitalization influences the green transformation of manufacturing companies, investigating four driving pathways: green cognitive ability, green innovation capability, investor attention, and media attention. We approached this analysis from the dual perspectives of internal capacity enhancement and external supervisory pressure, thereby addressing a gap in the literature regarding the impact of digitalization on green transformation. We also performed a heterogeneity analysis based on company ownership, pollution levels, technical characteristics, and other factors, acknowledging the variances among companies in digital transformation, and advocates for tailored digitalization strategies to facilitate the green transformation of companies. Lastly, we introduce a complete green transformation index that overcomes the constraints of current research overly dependent on green innovation indicators, presenting a more holistic evaluation framework for green transformation with substantial practical relevance.
Theoretical Background and Hypotheses Development
To explain the mechanisms through which digitalization empowers the green transformation of manufacturing enterprises, we constructed an analytical framework based on the complementary integration of socio-technical systems theory and stakeholder theory.
Socio-technical systems theory emphasizes the coevolutionary relationship between technical systems and social systems (Trist, Reference Trist1981). This theory posits that organizational performance improvement depends not only on the introduction of technological elements but also on the effective matching and interaction between technology and social elements such as organizational structure, personnel capabilities, and management processes. In the context of digital transformation, this theory provides explanatory power for understanding the reshaping of internal capabilities within enterprises: digital technology, as a core element of the ‘technical system’, promotes the enhancement of corporate green cognitive ability and green innovation capability through its interaction with organizational cognitive structures and innovation systems. However, socio-technical systems theory primarily focuses on the technology–society interaction within organizations, with relatively insufficient attention to the external environment in which enterprises operate. To address this theoretical gap, we introduce stakeholder theory as a complement (Cordeiro & Tewari, Reference Cordeiro and Tewari2015; Freeman, Reference Freeman1984). This theory emphasizes that corporate behavior is not only driven by internal strategic objectives but also subject to continuous supervision and expectation pressure from external stakeholders such as investors and media. Digital transformation enhances the supervisory capacity of external stakeholders over corporate green behavior by improving information transparency and disclosure quality, thereby forming an external pressure mechanism that promotes green transformation.
The complementary logic of the two theories: socio-technical systems theory explains how digitalization reshapes enterprises’ green cognition and green innovation capabilities from the perspective of internal capability enhancement; stakeholder theory explains how digitalization strengthens investors’ and media’s attention and constraints on corporate green transformation from the perspective of external supervision pressure. The two theories respectively focus on the internal and external dimensions of enterprises, forming a complementary dual explanatory pathway, jointly constructing a complete analytical framework of ‘Digital Transformation – Internal Capability Enhancement and External Supervision Pressure – Green Transformation’, as shown in Figure 1.
The theoretical framework of this work
The Direct Promoting Effect of Digitalization on the Green Transformation of Manufacturing Companies
Digital transformation is a process through which enterprises systematically reconstruct resource allocation, process optimization, and organizational strategy. From the perspective of socio-technical systems, the introduction of digital technologies will trigger a chain of adjustments within organizational systems, thereby providing foundational conditions for green transformation.
First, digital technologies, including big data, cloud computing, and artificial intelligence, are inherently innovative, and their application broadens the scope of innovative technologies, thereby enhancing the green R&D and innovation capabilities of companies (Wang, Gan, Zhou, & Wang, Reference Wang, Gan, Zhou and Wang2024). The digitalization of manufacturing companies diminishes information asymmetry, facilitates customer information acquisition, employs big data technology to accurately discern the requirements of potential customers and markets, modifies product specifications, and enhances the eco-friendliness of products to swiftly align with market demands, thereby lowering design trial and error costs and energy consumption. Conversely, R&D personnel can leverage digital technology to more readily access knowledge pertinent to green production and cutting-edge green patent technologies, bolster the capacity for assimilating and internalizing green innovation technologies, amalgamate existing technologies and resources, foster technological and knowledge exchange, and elevate the green development standards of companies. Second, digital technology enables companies to efficiently modify their operational and managerial structures while executing sustainable management practices. Digital technology dismantles temporal and spatial barriers in companies, facilitating the swift and efficient dissemination of strategies, projects, technologies, and other content, thereby lowering management and communication expenses, refining existing management processes to enhance efficiency, and fostering sustainable management (Li & Zhou, Reference Li and Zhou2024). Simultaneously, digitalization mitigates information asymmetry both internally and externally, enhances corporate governance, streamlines organizational structure, diminishes negotiation and oversight expenses, lowers transaction costs, and offers fundamental support for the ecological transformation of companies. Third, digital technology possesses inherent eco-friendly characteristics, including reduced energy usage, diminished pollutants, and a strong compatibility with environmental sustainability (Hao, Liang, Yang, Wu, & Hao, Reference Hao, Liang, Yang, Wu and Hao2024). Companies can leverage digital technology to oversee production processes, diminish energy and raw material consumption, enhance resource utilization efficiency, and optimize resource allocation; monitor pollutant emissions, categorize and treat ‘wastewater’ and ‘solid waste’, and establish a cycle of ‘recycling-utilization-treatment-re-recycling’ to maximize resource usage and minimize emissions; and achieve energy conservation and emission reduction. In summary, digitalization may directly facilitate the green transformation of manufacturing companies through many processes, including technical innovation, process optimization, and efficient resource utilization. Consequently, the subsequent hypothesis is posited:
Hypothesis 1 (H1): Digital transformation can substantially enhance the green transformation of manufacturing companies.
Internal Capability Enhancement Pathways: Based on Socio-Technical Systems Theory
According to socio-technical systems theory, the interaction between technology and organization alters managers’ cognitive structures and organizational innovation capabilities (Lim, Reference Lim2009; Trist, Reference Trist1981). We examine the internal capability enhancement pathways through which digitalization empowers green transformation from two dimensions: green cognitive capability and green innovation capability.
Digitalization, green cognitive capability, and company green transformation
Green cognitive capability refers to the cognitive level of senior management regarding resource and environmental challenges, as well as their cognition and experience of green development (Osborne, Stubbart, & Ramaprasad, Reference Osborne, Stubbart and Ramaprasad2001; Tu, Cao, Goh, & Wang, Reference Tu, Cao, Goh and Wang2024). Socio-technical systems theory posits that the interaction between technical systems and social systems reshapes the cognitive patterns of organizational members. Digitalization improves information openness and knowledge dissemination, enabling senior executives to attain a greater comprehension of green transformation at the strategic level. The green transformation of companies is characterized by significant difficulties and elevated risk, necessitating that companies incur greater financial expenses. The cultivation of environmental consciousness among company management facilitates a more timely and comprehensive understanding of national development policies and environmental protection frameworks, enhances access to information regarding sustainable development, deepens the internal comprehension of green initiatives, encourages proactive engagement with national green development strategies, implements energy conservation and emission reduction measures, and bolsters environmental awareness throughout production processes, thereby promoting the transparency of environmental information by companies. The awareness of ecological environment conservation among senior executives will directly influence the strategic decisions of companies. Consequently, the greater the green cognition of senior executives in companies, the more these companies will prioritize addressing the challenges of substantial investment and elevated risk associated with green innovation, thereby facilitating their green transformation. Consequently, the subsequent hypothesis is posited:
Hypothesis 2 (H2): Digitalization facilitates the green transformation of manufacturing companies by enhancing their green cognitive abilities.
Digitalization, green innovation capability, and company green transformation
Socio-technical systems theory also emphasizes that changes in technical systems will trigger the reconstruction of organizational innovation capabilities. As a general-purpose technology, digitalization not only improves the means of information transmission but also reduces the uncertainty of innovation and accelerates the diffusion and application of green technologies (Singh, Klarner, & Hess, Reference Singh, Klarner and Hess2020; Smith & Beretta, Reference Smith and Beretta2021). First, companies may swiftly acquire market demand through digital technology, enhance the eco-friendliness of products to tailor them for customers, minimize design and R&D expenses, precisely augment green R&D and innovation skills, and facilitate the sustainable transformation of companies. Second, manufacturing businesses consolidate original research and development, innovative resources, and talent through digitalization. They optimize the distribution of green R&D and innovation resources while generating substantial green innovation-related knowledge and patents. Simultaneously, they facilitate training and development for scientific researchers, augmenting their capacity to assimilate and innovate in green technologies, investigate advanced green technologies, and consider clean production methods at the initial stages of R&D alongside pollution control technologies at the final stages, thereby offering technical support for the green transformation of manufacturing companies. Third, digitization can mitigate spatial constraints, enabling diverse locations and sectors to share resources and participate in research and development as well as innovation. Transnational collaboration may introduce novel insights into research and development as well as innovation. The virtual aggregation created by digitalization can enhance collaboration among various areas and industries to bolster R&D and innovation capacities, thus advancing the green transformation of manufacturing companies. Consequently, the subsequent hypothesis is posited:
Hypothesis 3 (H3): Digitalization facilitates the attainment of green transformation in companies by enhancing their green innovation abilities.
External Supervisory Pressure Pathways: Based on Stakeholder Theory
Stakeholder theory points out that a company’s business behavior is not only driven by internal goals but is also subject to the supervision and expectations of external stakeholders (Cordeiro & Tewari, Reference Cordeiro and Tewari2015; Freeman, Reference Freeman1984). We examine the external supervisory pressure pathways through which digitalization empowers green transformation from two dimensions: investor attention and media attention.
Digitalization, investor attention, and company green transformation
Digitalization, by improving the quality and transparency of information disclosure, enables investors to more effectively supervise and evaluate companies’ green performance (Sun, Sun, Wang, & Wang, Reference Sun, Sun, Wang and Wang2025). As green development has become a national plan, external investors are shifting their attention from only short-term economic rewards to including the environmental responsibility of companies in their investment evaluations. The growing desire for green investments among investors has compelled companies to focus more on their own green initiatives. Simultaneously, as the informational conduit of the capital market, analysts employ their expertise to interpret and disseminate corporate information externally, thereby enhancing the likelihood of revealing deceptive green initiatives or environmental inaction by companies. This process mitigates the adverse effects of managerial short-sightedness to some degree and facilitates the green transformation of companies. Furthermore, as a significant overseer in the capital market, the ‘staring effect’ induced by media stories is readily produced. The disclosed companies are subject to public scrutiny. This can elevate the pressure associated with environmental pollution practices, compelling noncompliant companies to heighten their awareness of ecological and environmental stewardship; it can also enhance the reputational capital of green companies, establish product credibility, attract external investment, and expedite green transformation (Ahern & Sosyura, Reference Ahern and Sosyura2015). Consequently, the subsequent hypothesis is posited:
Hypothesis 4 (H4): Digitalization facilitates green transformation in companies by enhancing the attention of investors.
Digitalization, the media attention, and company green transformation
As an important external supervisor in the capital market, the media has a significant ‘watchdog effect’ on company behavior (Ahern & Sosyura, Reference Ahern and Sosyura2015). During green transformation, companies encounter substantial capital investments, elevated risks, and diminished success rates, making them susceptible to market failures such as path dependence and environmental externalities, which frequently impact the transformation process (Chen, Reference Chen2014). The significant focus of the media and public can serve as a powerful external driving force for the green transformation of companies (Chen, Yan, Xie, & Ou, Reference Chen, Yan, Xie and Ou2024). Through digital transformation, companies may enhance information transparency, optimize resource allocation, and demonstrate their capacity for compliant operations and sustainable transformation to external stakeholders. This not only bolsters the environmental reputation of companies in the public and media but also expedites the execution of their green transformation. The media’s supervisory function and public scrutiny of corporate environmental practices can compel companies to enhance their environmental protection initiatives, foster the development of green technologies and optimize production processes, and minimize resource consumption and pollution emissions, thus facilitating the green transformation of companies and securing greater market opportunities and financial backing. Consequently, digital transformation enhances the green transformation process of companies by amplifying media attention.
Hypothesis 5 (H5): Digitalization facilitates the green transformation of companies by enhancing media attention.
Methods
Model Setting
In order to test the impact of digitization on the green transformation of manufacturing companies, we draw on Wang and Guo (Reference Wang and Guo2022) and set up the following benchmark model:
\begin{equation}Gree{n_{it}} = {\alpha _0} + {\alpha _1}Di{g_{it}} + \sum {{\beta _i}{X_{it}} + \sum {Industry} + \sum {Year} + {\varepsilon _{it}}} \end{equation} The explained variable 
$Gree{n_{it}}$ denotes the green transformation degree of company i in year t; the core explanatory variable 
$Di{g_{it}}$ denotes the digitization degree of company i in year t; and 
${X_{it}}$are company control variables, including company size and nature of ownership. Industry and Year denote the industry and yearly fixed effects, respectively.
According to the model setting, if the regression estimate coefficient 
${\alpha _1}$of the core explanatory variable Greenit is significantly positive, it is inferred that the digitization of manufacturing companies positively affects the green transformation of companies, and the deepening of digitization can significantly enhance the green transformation. On the contrary, if the coefficient 
${\alpha _1}$ is significantly negative, it indicates that the promotion of digitization level inhibits corporate green transformation.
Variable Measurement
Explained variable – Corporate green transformation (Green)
At present, there are two main measurement methods in academia regarding the green transformation of companies: one is to measure the green transformation of companies based on the number of green patents or the amount of environmental protection investments, environmental management system certifications, and other single indicators. However, green transformation is a comprehensive overview of the company’s green development mode, including but not limited to the increasing awareness of the managers’ green perception, continuous environmental protection investment, gradually optimized green production process, and innovative green products. The corporate green transformation is not only reflected in the R&D of patented technologies but also in the publicity initiatives, strategic concepts, pollution reduction, and carbon reduction actions. Therefore, a single indicator such as green patents is not enough to fully measure a company's level of green transformation. The second is based on texts through big data recognition technology. In the context of accelerated green development strategy, green transformation-related information will be reflected in the strategic annual reports and Corporate Social Responsibility (CSR) reports of companies, which reflect the green transformation strategy and green transformation achievements of companies. We draw on the research of Loughran and Mcdonald (Reference Loughran and Mcdonald2011) to measure corporate green transformation based on textual information disclosed in annual reports. Compared with the content analysis method, the text analysis method applies computerized natural language processing technology, which can accomplish the precise identification of unstructured text under large samples, greatly reducing the error rate and improving the consistency of judgment. There are two main reasons for choosing annual reports of listed companies as the observation text: one is that green transformation belongs to important strategic information of listed companies, which will be disclosed in the annual report, the most widely publicized document, and is compatible with the summarizing and guiding characteristics of the annual report. Second, annual reports of listed companies are mandatory information disclosure, with strict format requirements and wording specifications, which will greatly improve the efficiency of keyword matching. Therefore, using the number of green transformation words in annual reports of listed companies to measure the green transformation of companies has high feasibility in this aspect. Hart (Reference Hart1995) argues that five key factors, namely, green competence based on products and production processes, employee training and involvement in environmental issues, organizational competence across internal functions, formal environmental management systems and procedures, and strategic planning based on environmental issues, could greatly contribute to the green transformation of companies. On a deeper level, in green transformation, the company needs to produce green products based on green technological innovation, reduce the total amount of pollution emissions, and then improve the performance and sustainable development of companies. Therefore, in accordance with the Environmental Protection Law, the Technical Guidelines for Corporate Environmental Behavior Evaluation, White Papers of Green Manufacturing Standardization, and other policy documents, we selected 113 keywords for corporate green transformation from 5 categories: publicity initiatives, strategic concepts, technological innovations, emission control, and monitoring management. The frequency of each keyword in the annual reports of listed companies is counted to form the word frequency of green transformation, and the natural logarithm of the word frequency plus 1 is used to measure the green transformation level of companies. The larger the value of the indicator, the higher the green transformation level of the company. To validate the index, we compared it with other conventional green transformation indicators, such as the quantity of green patents, and established that the index accurately represents the green transformation of companies. Furthermore, in the robustness test part, we evaluated and examined alternative indicators and various measurement methods to ascertain the robustness and validity of the index.
Explanatory variable – Corporate digitization (Dig)
In this article, we also draw on the methods of Hassan, Hollander, Van Lent, and Tahoun (Reference Hassan, Hollander, Van Lent and Tahoun2019) and Chen, Dasgupta, Huynh, and Xia (Reference Chen, Dasgupta, Huynh and Xia2023) to construct digital transformation indicators of companies based on text analysis. The annual reports of all A-share listed companies in the Shanghai Stock Exchange and Shenzhen Stock Exchange were summarized and organized by Python crawler function, and all text contents were extracted by the Java PDFbox library, which are used as a data pool for subsequent feature word screening. Based on the data pool formed by Python’s text extraction of listed companies’ annual reports, the program searched, matched, and counted the word frequency based on the feature words, and then categorized and collected the word frequency of key technology directions to form the final total word frequency. In order to solve the problem of large standard deviation and right-skewed data that exist in the total word frequency measurement, we added 1 to the word frequency of digitized keywords and carried out logarithmic processing to measure the digitization level of companies (Dig).
Control variables
Based on the ideas of He, Chen, and Zhang (Reference He, Chen and Zhang2021) and Li et al. (Reference Li, Lian and Xu2023), we selected company indicators that may affect green transformation as control variables, including: company size (Size), which is expressed as the natural logarithm of the company’s total assets; company age (Age), which is measured by the logarithmic value of the difference between the year of examination and the company’s establishment year plus 1; return on total assets (Roa), calculated as the ratio of net profit to total assets; debt to asset ratio (Debt), calculated as the ratio of total liabilities to total assets; Duality (Dual), a dummy variable reflecting the governance structure of companies, which takes the value of 1 if the president and the general manager of the company are the same person, and 0 if they are not; shareholding concentration (Share), which is measured by the proportion of shares held by the largest shareholder.
Data Sources
Since 2010, China’s digital technology has advanced considerably, leading to rapid growth in the digital economy and an increased emphasis on digitalization in company development, especially as manufacturing companies face intensified environmental pressures that demand immediate green transformation. The data in this study culminates in 2022, marking a critical moment for China’s manufacturing sector, characterized by increasing environmental concerns and the intensification of green regulations. During this period, digital transformation has increasingly become a vital catalyst for company growth, especially in green transformation, where the utilization of digital technology has markedly intensified. Moreover, the accessibility and trustworthiness of data in 2022 make it a prudent and feasible choice. This study employs manufacturing companies listed on the Shanghai and Shenzhen A-share marketplaces from 2011 to 2022 as the research sample. Thus, the sample is selected based on the following steps: (1) the listed manufacturing companies are screened through the CSMAR database based on the manufacturing classification in the 2012 version of the Industry Classification of the Securities and Exchange Commission; (2) the samples of ST, ST*, PT, and the samples of insolvency are excluded; (3) the samples with main variables missing in three consecutive years are excluded; and (4) the samples with the main variables missing during the sample period are excluded. Finally, the smooth panel data of 710 listed manufacturing companies from 2011 to 2022 are obtained as the research sample. The data on green transformation of manufacturing companies come from the annual reports, social responsibility reports, sustainability reports, and official websites of listed companies in the database of CSMAR, which are categorized, collected, and sorted out, and then be processed through comprehensive arithmetic; the data on digitization of companies, the data on the nature of companies, the financial data, and other control variables, are mainly obtained from the database of CSMAR; the PPI (Producer Price Index) and Price Index of Investment in Fixed Assets are from China Statistical Yearbook. The results of descriptive statistical analysis of the variables are listed in Table 1.
Descriptive statistics of variables
Table 2 presents the correlation coefficients and variance inflation factors (VIFs) for the model’s explanatory variables. The results in Table 2 indicate that the VIFs of the explanatory variables in the model are all below 10, suggesting the absence of significant multicollinearity among the explanatory variables.
Correlation coefficients and variance inflation factors (VIF) for the model’s explanatory variables
*, **, Notes:and *** indicate significance at the level of 10%, 5%, and 1%, respectively.
Results
Benchmark Regression Analysis
In order to verify the impact of digitalization on the green transformation of manufacturing companies, we constructed a fixed effects model for empirical analysis, and Table 3 reports the benchmark regression results. This study employs the industry fixed effect rather than the enterprise fixed effect, primarily due to the substantial disparities in regulations, technologies, market environments, and other factors encountered by various industries. External factors significantly influence the advancement of green transformation in manufacturing companies, and the industry-level fixed effect may effectively manage these heterogeneities. Consequently, the industry fixed effect more accurately captures the influence of digitalization on green transformation compared to the company fixed effect. In the manufacturing sector, the disparities in policy, market demand, and technological advancement among industries are particularly pronounced, rendering the industry fixed effect a more appropriate selection.
Benchmark regression result
*, **, Notes:and *** indicate significance at the level of 10%, 5%, and 1%, respectively; robust standard errors are in parentheses.
In the estimation results of the benchmark regression model in Table 3, column (1) is the estimation result without control variables and fixed effects; column (2) is the estimation result with the inclusion of the fixed effects of industry and time; and the estimation results in both column (1) and column (2) show that corporate digitization (Dig) promotes the green transformation of manufacturing companies with estimated coefficient of digitalization be significant at the 1% level. Column (3) presents the estimation results after controlling for fixed effects with the addition of company variables (company size, company age); column (4) presents the estimation results after adding the control variables of financial indicators, which show that the estimated coefficient of company digitalization is positively significant at the 1% statistical level. On this basis, column (5) is the estimation results including the control variables of corporate governance, and the estimated coefficient of corporate digitization is 0.0399, which is statistically significant at the 1% level. The above regression results indicate that the higher the digitization level of manufacturing companies, the deeper the green transformation is, both positively correlated, regardless of whether fixed effects of industry and time are considered. This indicates that digitization significantly contributes to the green transformation of manufacturing companies, which verifies H1. Current literature posits that the digital process facilitates the green transformation of companies by diminishing information costs and enhancing resource allocation efficiency, ultimately leading to an improvement in green total factor productivity (Wu, Shi, & Wang, Reference Wu, Shi and Wang2024). The fundamental regression findings of this study indicate that digitalization has markedly advanced the green development of manufacturing companies, aligning with the conclusions of pertinent studies on green innovation and green investment. Digitalization facilitates the implementation and dissemination of green technology, enhances overall production efficiency in companies, and decreases energy consumption, hence substantiating the trajectory of microlevel improvements in green total factor productivity. This discovery not only validates the research outcomes at the macro and industrial levels but also offers further evidence at the company level.
Robustness Analysis
A series of robustness tests were conducted to make the benchmark results more reliable.
(1) Adjusting the measurement of variables. The measurements of the explained variable, green transformation, and the explanatory variable, company digitization, were adjusted separately to carry out the robustness test of the research results. Following the method of Liang and Sun (Reference Liang and Sun2024), we selected the number of green patent applications (Green 1) to replace the green transformation indicator, in which the number of green patent applications = the number of green invention patents + the number of green utility patents, and conducts a test through negative binomial regression. In order to eliminate the right-skewed distribution problem of the number of green patent applications, we took the natural logarithm of the number of green patent applications plus 1 (Green 2) to make a robustness test. The estimation results are shown in columns (1) and (2) of Table 4. The comparison shows that the negative binomial regression based on the number of patent applications and the natural logarithm of the number of patent applications plus 1 have little effect on the results, i.e., digitization continues to contribute significantly to green transformation when using different ways to measure the green transformation of companies, and the basic conclusion is robust.
Robustness test results
*, **, Notes: and *** indicate significance at the 10%, 5%, and 1% levels, respectively; robust standard errors are in parentheses.
In addition, we followed Zhao and Fang (Reference Zhao and Fang2023) by measuring the digitization level of companies by the ratio of digital transformation-related intangible assets to total intangible assets at the end of the year as disclosed in the notes to the financial reports of listed companies. Specifically, when the intangible assets item contains the words ‘data’, ‘information technology’, ‘software’, ‘network’, ‘client’, ‘management system’, ‘intelligent platform’, and other keywords related to digital transformation technologies and patents, the item is defined as ‘digital intangible assets’. We then summed up multiple digital technology-related intangible assets of the same company in the same year and calculated its proportion to intangible assets in the current year to measure the digitalization level of the company (Dig 1). The estimation results are presented in column (3) of Table 4, and the regression coefficients remain significantly positive compared to the benchmark regression, indicating that the basic conclusions are robust.
(2) The lagged effect is considered. The lagged effect of listed companies’ annual report data is excluded, and the explanatory variable, company digitization, is lagged by one period (Dig (−1)) to reduce the bias of estimation. The estimation results are shown in column (4) of Table 4 on the following page: the estimated coefficients are significantly positive, which are consistent with the benchmark regression results.
(3) Omitted variables are considered. The benchmark regression in the article only controls the fixed effects of industry and time, and if we do not add the fixed effects of the region to which the company belongs, we may omit the variables that do not change over time at the regional level, leading to bias in the regression. On the basis of the benchmark regression model, we added fixed effects to control the impact of urban factors on the regression results. Column (5) in Table 4 shows the regression results with fixed effects of city, indicating that there is no significant difference in the impact of the provincial and urban digitization level on the green transformation of manufacturing companies.
(4) Alter the model estimation technique. This research further examines the distribution characteristics of the explained variables to assess the robustness of the benchmark regression results. Given that the degree of green transformation is a non-negative variable and may exhibit truncated characteristics, employing ordinary least squares may lead to an underestimation of the model coefficients’ importance. Thus, we utilized established research practices, employed the Tobit model for rigorous regression analysis, and applied the maximum likelihood estimation method for parameter estimation (Lin & Xie, Reference Lin and Xie2025; Xie & Lin, Reference Xie and Lin2025a, Reference Xie and Lin2025b). The estimation outcomes of the Tobit model are presented in the sixth column of Table 4. The findings indicate that the impact of digitalization on the green transformation of organizations remains substantial, suggesting that the conclusions of this article possess considerable resilience.
The results of the above robustness tests can prove the robustness of H1.
Endogeneity Treatment
(1) Instrumental variables. In this article, we argue that the development of digital technologies in China may be influenced by the distribution of early post offices. The distribution of early post offices in China affects the usage habits and potential preferences of digital technologies in the region, which in turn affects the digital transformation of companies. This can fulfill the relevance principle in instrumental variable selection. At the same time, early post offices that aim to provide communication services do not directly affect the green innovation performance of companies at this stage. Based on the above analysis, we draw on the method of He and Chen (Reference He and Chen2025), who selected postal data of each city in 1984 as the instrumental variable for the explanatory variable, company’s digitalization level, and constructed the panel data by cross-multiplying the number of national Internet users in lagged one period with the postal data of each city in 1984, respectively, to obtain the instrumental variable (IV). The instrumental variable (IV) reflects the trend of urban digitalization level over time to a certain extent, and company’s digitalization level is closely related to the digitalization level of its location, which meets the correlation condition; the urban digitalization level has no direct impact on company’s green transformation decision; the green transformation level of a single company will not significantly change the urban digitalization level. Therefore, choosing the urban digitalization level as an instrumental variable for company digitization can satisfy the exclusivity and exogeneity of instrumental variables at the same time. Using the above instrumental variables for the endogeneity test, the estimation results are shown in columns (1) and (2) of Table 5. Column (1) takes company digitization as the explained variable, and takes instrumental variable, urban digitization level, as the regression result of explanatory variable, showing that the regression coefficient of the instrumental variable on the company’s digitization level is 0.0277, significantly positive at the statistical level of 1%, indicating that the company’s digitization is positively affected by urban digitization level. Column (2) takes corporate green transformation as the explained variable, and takes the company’s digitization level estimated by instrumental variables as the explanatory variable, which is still positively significant at the 1% level and the estimated coefficient is significantly enlarged. This implies that company digitization has a significant positive effect on green transformation after controlling for endogeneity, and the marginal effect is even greater. Moreover, in the first stage, the F-test value is 18.9372, and the P-value of Anderson Lagrange Multiplier (LM) test is 0.0001, so the instrumental variable (IV) is considered to be correlated with the endogenous variables, and there is no weak instrumental variable problem, which means that the conclusion is still significant after controlling for endogeneity.
Results of endogeneity treatment
*, **, Notes:and *** indicate significance at the level of 10%, 5%, and 1%, respectively; robust standard errors are in parentheses.
(2) Heckman two-step method. In the article, some listed companies are selected as samples, and from the observed values, it is found that the digitization level of companies has a large gap among different companies, and their parameters, such as company ownership and company size, are also different. In order to overcome the problem of sample selection bias, the Heckman two-step method was constructed to verify the model. In the first step, the dummy variable of company digitization is set to predict the probability of each company, the Probit model is used to select the equation and estimate the probability of company digitization, and the Inverse Mills Ratio (IMR) is calculated with the risk function. In the second step, the IMR obtained in the first step is used as a control variable, being added to the benchmark regression estimation with the addition of the fixed effects of industry and time, which are shown in column (3) of Table 4. The results show that IMR is significant at 1% statistical level, sample selection bias affects the benchmark regression, and after correcting the sample selection bias, the regression results show that the estimated coefficient of company digitization on green transformation is not much different from the benchmark regression, and is positively significant at 1% statistical level. It verifies H1: company digitization significantly promotes the green transformation of manufacturing companies.
Mechanism Analysis
This section will explore the impact mechanism of digitization on the green transformation of manufacturing companies. Drawing on the research of Li and Zhou (Reference Li and Zhou2024), we set up the following model:
\begin{equation}{M_{it}} = {\alpha _0} + {\alpha _1}Di{g_{it}} + {\sum {{\beta _i}X} _{it}} + \sum {Industry + \sum {Year + {\varepsilon _{it}}} } \end{equation} 
${M_{it}}$ is a mechanism variable, including a number of indicators from three aspects including innovation capacity, structural optimization, and resource utilization. The test results of mechanism are shown in Table 6.
Test results of the influence mechanism
*, **, Notes:and *** indicate significance at the level of 10%, 5%, and 1%, respectively; robust standard errors are in parentheses.
We examined the impact of digital empowerment in the green transformation of manufacturing companies from two perspectives: internal capacity enhancement and external regulatory pressure. The enhancement of internal capacity is categorized into green cognitive ability and green innovation ability, whereas external supervisory pressure is divided into investor attention and media attention, to investigate how digitalization facilitates the green transformation of manufacturing companies through internal and external influencing factors.
Mechanism Test Based on Green Perception
Drawing on the research of Duriau, Reger, and Pfarrer (Reference Duriau, Reger and Pfarrer2007), and with green competence perception, corporate social responsibility perception, and external environmental pressure perception as the keywords, we examined the green perception level of companies from three dimensions. In addition, Python was used to mine the information involving green perception in the annual reports of listed companies, and then the information was categorized and pooled to get the total frequency of green perception keywords in the text. The green perception variable (Gco) of companies was obtained through taking the logarithm of ‘the proportion of the total frequency of green perception keywords in the text to the total number of words in the text and plus one’.
Column (1) of Table 5 shows the test results of company’s green perception. The test results show that a company’s digitization can significantly improve the green perception of companies, and then improve the green transformation ability of companies. The test results verify H2. Current literature indicates that company management’s environmental consciousness is a crucial prerequisite for advancing green investment and innovation (Liu & Chen, Reference Liu and Chen2024). Digitalization augments organizations’ capacity to recognize environmental hazards and sustainable opportunities, hence enhancing the quality of decision-making in their green transformation. This aligns with the findings in green investing literature that cognitive improvement facilitates the allocation of funds to environmentally sustainable companies.
Mechanism Test Based on Green Innovation Capacity
From the perspective of innovation output, we selected relevant indicators to measure the green innovation level of companies. Since invention patents and utility model patents are more innovative compared with design patents, they can more substantially reflect the green innovation level of companies. We measured the green innovation level of companies by the sum of green invention patent applications and green utility model patent applications of companies. Considering the right-skewed characteristics of the data, we added 1 and took the natural logarithm in the regression process. Similarly, considering the time lag of green innovation, we chose the sum of green invention patents and green utility model patents filed with one-period lagged in the regression process to measure the green innovation level of companies (Gin).
Column (2) of Table 6 is the test of green innovation capacity of companies, and the results show that the improvement of company digitalization level can significantly enhance the green innovation capacity of companies, thus accelerating the front-end cleaner production technology and back-end pollution prevention technology. The results test the mechanism of green innovation capacity of companies and verify H3. Green innovation is the fundamental catalyst for enhancing green total factor productivity. This study concludes that digitalization markedly enhances the efficiency of company green R&D by optimizing information transmission and resource allocation. This aligns with current studies highlighting the enhancement of green innovation on green productivity and further demonstrates that digitalization is a crucial factor in augmenting green innovation potential (Zhao, Nakonieczny, Jabeen, Shahzad, & Jia, Reference Zhao, Nakonieczny, Jabeen, Shahzad and Jia2022).
Mechanism Test Based on Investor Attention
Some scholars have argued that monitoring pressure from external supporters will motivate companies to take environmental-friendly measures to demonstrate their commitment to protecting the natural environment (Truong, Mazloomi, & Berrone, Reference Truong, Mazloomi and Berrone2021); that is to say, a company’s green transformation requires not only endogenous motivation but also monitoring pressure from external supporters. In this article, we focused on the role of external market concerns from the perspective of investors. For the variable of investor attention (Investorf), we used the logarithm of the annual median of the Baidu search index to measure it, which is mainly obtained through the statistics of searches related to listed companies in the Baidu keywords search.
Column (3) of Table 6 is a test of investor attention, and the results show that the increase of corporate digitization significantly enhances investor attention, which in turn promotes the green transformation of companies, verifying H4. The green investment literature emphasizes the sensitivity of the capital market to the green behavior of companies, while digitalization significantly improves the information transparency of companies, thus attracting more investors’ attention (Sun et al., Reference Sun, Sun, Wang and Wang2025). The results indicate that this attention not only brings capital inflow but also strengthens investors’ financial support for green projects, thus promoting the greening of companies at the capital level, which is highly consistent with the logic of green investment-related research.
Mechanism Test Based on Media Attention
Certain researchers contend that the media, functioning as an external overseer, significantly influences information intermediation and imposes constraints within company governance. Its reports can influence public sentiment and investor anticipations, so compelling companies to adopt enhanced environmental protection measures in response to societal concerns (Graf-Vlachy, Oliver, Banfield, König, & Bundy, Reference Graf-Vlachy, Oliver, Banfield, König and Bundy2020; Luo, Xiong, & Mardani, Reference Luo, Xiong and Mardani2022). The green transformation of companies relies not only on the enhancement of internal capacities but also on the facilitation of external regulatory pressure. This research primarily investigates the influence of external oversight on the green transformation of companies through the perspective of media. We quantified media attention (Media) by employing the natural logarithm of the annual report count of the company as reported by online media, with an increment of one. The pertinent data selects reports from prominent mainstream financial publications, such as East Money, as samples. These reports exhibit extensive coverage and significant authority, providing a more precise reflection of the exposure and regulatory pressure faced by companies in the realm of public opinion. Column (4) of Table 6 assesses media attention. The findings indicate that enhancing the digital capabilities of companies can markedly increase media attention, hence facilitating the green transformation of these companies and verifying H5.
Heterogeneity Analysis
The preceding section confirmed the impact of company digitalization on green transformation. The sample concentrated on the manufacturing sector, revealing a significant disparity among industries. Is there a distinction in the effect of digitalization on green transformation across various enterprises? We utilized company ownership, pollution levels, and technological characteristics as the foundation for heterogeneity analysis. These variables can accurately represent the disparities in external regulatory pressure, market opportunities, and technological innovation capabilities encountered by companies during the execution of digital transformation. For example, there are differences between state-owned enterprises and non-state-owned in terms of strategic objectives, resource acquisition and information circulation, which may lead to different effects of digital transformation. The level of pollution indicates the environmental strain encountered by companies during the green transformation process. Industries with significant pollution may face increased external oversight and investment limitations during their transformation, potentially enhancing their impetus for green transition. We selected these elements as the foundation for the heterogeneity test to enhance our comprehension of how digital transformation influences the green transformation process across various types of companies.
Heterogeneity Test Based on Company Ownership
There are significant differences between state-owned enterprises (SOEs) and non-state-owned enterprises (non-SOEs) in terms of their business objectives, resources, and access to information, so the facilitating effect of digitization on green transformation of companies may be heterogeneous in terms of ownership. Compared with non-SOEs, the effect of digitalization on SOEs’ green transformation is greater. First, in terms of business objectives, SOEs actively respond to the national strategic policy and aim to maximize the overall social benefits, while non-SOEs are more focused on pursuing the economic benefits of themselves. Therefore, SOEs are able to implement national policies such as green transformation faster and better. Second, in terms of resource acquisition, SOEs bear the mission and responsibility given by the state and society, have higher sensitivity and responsiveness to policies, and are able to enjoy policy dividends earlier. Compared with non-SOEs, SOEs have more advantages in R&D financing, talent structure, policy subsidies, etc., and are more qualified and capable of bearing the risks in the process of green transformation. Therefore, SOEs enjoy greater strength in implementing green transformation. Finally, in terms of information acquisition, SOEs are valued and supported by the government, can obtain the latest information from the government at a lower cost, avoid losses due to asymmetric information, and realize the integration of information and resources, which finally promote the green transformation of companies more efficiently.
Is there any difference in the impact of digitization on green transformation among companies of different ownership, and does the digitization of SOEs contribute more to green transformation? To this end, the enterprises are grouped according to the nature of their ownership and regressed in subsamples, and the regression results are shown in columns (1) and (2) of Table 7. The estimation results in column (1) of Table 7 show that the digitization of state-owned manufacturing enterprises has a significantly positive effect on green transformation at the 1% significance level. The estimation results in column (2) show that the digitization of non-state-owned manufacturing enterprises has a significant positive effect on green transformation at the 1% significance level. However, the value of the coefficient of the effect of digitization on green transformation of the non-state-owned manufacturing enterprises is higher than that of SOEs. This means that the facilitating effect of digitization on the green transformation of non-state-owned manufacturing enterprises is higher than that of SOEs. The reason may be that non-state-owned manufacturing enterprises are profit-oriented, and they distribute benefits based on contribution to encourage personnel to enhance efficiency, improve technology, and promote innovation. Some of the non-SOEs have built their foundations in digitalization, making it easier to implement digitalization, stimulate its potential, and boost green transformation. While state-owned manufacturing enterprises are not entirely profit-oriented, the enterprises’ own survival pressure is small. In addition, compared to non-state-owned manufacturing enterprises, its driving force for digitalization is weaker, its integration capability of digitalization is not strong, and it cannot give full play to the role of digitalization. As a result, its digitalization impact on green transformation is lower than that of non-SOEs. In alignment with the findings of prior research regarding the disparities in green innovation capabilities among companies of varying ownership (Liu, Li, Peng, & Lee, Reference Liu, Li, Peng and Lee2020), we offer additional microlevel evidence that digitalization exerts a more pronounced influence on the green transformation of non-SOEs.
Results of heterogeneity analysis
*, **, Notes:and *** indicate significance at the level of 10%, 5%, and 1%, respectively; robust standard errors are in parentheses.
Heterogeneity Test Based on Company’s Pollution Level
A company’s green transformation is also closely related to its pollution emission levels. Heavily polluting companies face relatively greater external environmental regulatory pressure, so their green transformation motivations are greater than non-heavily polluting companies, and their requirements for green transformation are also higher. Therefore, in green transformation, heavily polluting companies are more empowered by digitalization. In addition, under the pressure of environmental regulations, heavily polluting companies need more time and capital investment to realize the digital construction of infrastructure compared with non-heavily polluting companies. Companies can allocate resources more effectively and rationally through digitalization, improve information transparency, and then affect the green transformation. Therefore, environmental regulation exerts a higher impact on the green transformation of heavily polluting companies than that of non-heavily polluting companies, and heavily polluting companies need to develop digital technology in the production process to accelerate their green transformation. Moreover, compared with non-heavily polluting companies, heavily polluting companies receive more attention from the government and are subject to greater penalties, which push the heavily polluting companies to realize green transformation through digital technology.
Heavily polluting companies, characterized by crude development, take the main responsibility for comprehensive ecological improvement at this stage. It is reasonable to believe that the endowment difference between heavily polluting and non-heavily polluting companies is highly likely to affect the green transformation driven by digital transformation. For this reason, we classified the sample companies into heavily polluting companies and non-heavily polluting companies according to the ‘Classification Management List of Listed Companies’ Environmental Verification’ published by the Ministry of Ecology and Environment of the PRC, and identifies the heterogeneity through group regression test, and the results are as shown in columns (3) and (4) of Table 7.
From the estimation results, it can be seen that the company’s digitalization significantly drives green transformation for both heavily polluting and non-heavily polluting companies. The difference is that, from the estimated coefficient value of the core explanatory variables, the effect of digitalization on green transformation of non-heavily polluting companies is higher than that of heavily polluting companies, because heavily polluting companies cause environmental pollution, and in recent years, a series of regulatory measures issued by the relevant departments in China have subjected them to the pressure of environmental regulation and even legitimacy. At the same time, with the gradual improvement of the green financial system, heavily polluting companies face increasing financing difficulty compared with non-heavily polluting companies. Under these regulatory pressures, heavily polluting companies will undergo green transition under these regulatory pressures, which to a certain extent, weakens the driving role of digital transformation. This outcome aligns with current research on corporate green innovation behavior in response to environmental regulation and further enriches the micro evidence within the framework of digital empowerment (Peng & Kong, Reference Peng and Kong2024).
Heterogeneity Test Based on Technical Attributes
There is a huge difference among manufacturing companies with different technological attributes. Relatively speaking, high-tech manufacturing companies have stronger technological innovation capability, higher productivity and lower energy consumption, and it is easier for them to improve production and operation through digitalization. Does the digitalization of manufacturing companies with different technological attributes have the same effect on green transformation? Is it true that the digitization of companies in high-tech industries has a greater impact on green transformation? To this end, according to the Classification of High-Tech Industries (Manufacturing) (2013) issued by the National Bureau of Statistics of China, and with reference to the Guidelines for Industry Classification of Listed Companies issued by the China Securities Regulatory Commission in 2012, we categorized eight types of manufacturing companies as high-tech manufacturing companies, which include the pharmaceutical manufacturing industry, the aerospace and equipment manufacturing industry, and the electronic and communication equipment manufacturing industry. According to the above criteria, the samples are categorized and regressed, and the results are shown in columns (5) and (6) of Table 7. Among them, column (5) is the subsample regression results of high-tech manufacturing companies, and column (6) is the subsample regression results of non-high-tech manufacturing companies. The estimation results show that the regression coefficients of digitization on green transformation for both high-tech and non-high-tech manufacturing companies are positively significant at the 1% statistic level of significance. Meanwhile, the regression coefficients of non-high-tech companies are significantly larger than those of high-tech companies. This means that the digitization of manufacturing companies with different technological attributes can promote green transformation, and the digitization of non-high-tech companies plays a more significant role in boosting green transformation. Possible reasons may be that non-high-tech manufacturing companies are characterized by rough growth, and compared to high-tech companies, they face higher pressure in energy consumption, environmental protection, and other aspects; thus, the promotion effect of digital technologies on non-high-tech manufacturing companies is more striking. While the high-tech manufacturing companies are more energy-saving with less pollution emissions. So compared with high-tech companies, the promotion effect of digital technologies on non-high-tech manufacturing companies is more striking. This finding aligns with the conclusions of prior studies indicating variations in the performance of companies with distinct technical attributes regarding green innovation and productivity enhancement (Tong, Hu, Yang, Liu, & Xu, Reference Tong, Hu, Yang, Liu and Xu2024), and it augments the micro evidence within the framework of digitalization.
Discussion
On the basis of theoretical analysis, we utilized data of A-share listed companies in Shanghai and Shenzhen from 2011 to 2022 to construct green transformation indicators of manufacturing companies, empirically examine whether digitalization advances the green transformation of manufacturing companies, and analyze its impact mechanism. The study confirms that upgrading digitization is conducive to promoting the green transformation of manufacturing companies. The article proposes a four-pronged impact mechanism regarding green perception, green innovation ability, investor attention, and government subsidies, which is of great reference for manufacturing companies in realizing green transformation upgrading and high-quality development.
Based on the above findings, we put forward the following policy recommendations: First, the government should strengthen policy guidance and assistance for the digital transformation of companies, strengthen the integration of digital transformation and green development in industrial policy and development planning, and propose creative and demonstrative solutions for company transformation in green and synergetic approaches. The government should also enhance financial support for digital transformation, ease the financing constraints on R&D investment, reduce the intensity of digital transformation risks, and fully release the amplification, superposition, multiplication, and multiplier effects of digital transformation on the green economy. At the same time, the government should integrate companies, universities, research institutes, and other forces to establish systematic and task-oriented innovation platforms, strengthen cooperation and training in green innovation, create a co-innovation platform for the development of green, low-carbon industrial technologies, identify the focus of digital technologies in empowering ecological environmental protection, construct a green technological innovation system, so as to digitally empower the green development of companies.
Second, the embedding of digital technologies in the production process of companies should be strengthened to explore new approaches of green development. On the one hand, companies should accelerate the application of big data, cloud computing, IoT, and other digital technologies in green manufacturing, ‘Being dyed green’ through the application of digital technologies in procurement, manufacturing, distribution, recycling, and reuse, primarily focus on the construction of green factories, green product design, and green supply chain management, improve the intensification of land use, production cleanliness, waste recycling, and low-carbon energy level, comprehensively reduce the energy consumption in production, and continuously enhance green economic development. On the other hand, companies should actively play a fundamental role in green technological innovation by accelerating the in-depth integration of digital technology with R&D and design, production processes, energy management, green and safe logistics and transportation, resource recycling, and other fields, so as to realize a win–win outcome for companies in both economic benefits and environmental effects.
Third, the heterogeneous characteristics of companies should be fully taken into account, and differentiated digital development strategies should be implemented. Policy preferences and financial support should be provided based on different digital transformation and green needs of companies, so as to maximize the environmental welfare brought by digital transformation. The government should optimize financial support for companies in digital transformation, simplify the administrative approval process of banks and other financial institutions, introduce targeted fiscal and tax policies to help non-SOEs and small- and medium-sized enterprises that lack capital and technology to overcome financial difficulties, and improve the awareness of environmental responsibility and efficiency of green innovation. For heavily polluting companies, especially those with high sulfur dioxide emissions, the government should formulate detailed incentives based on the digital transformation of companies, such as preferential policies and financial support for heavily polluting companies that use IoT technology for quality control in product production, and build a complete green production chain, so as to promote the green and sustainable development of heavily polluting companies.
Data availability statement
The data used in this study are available from the following sources: (1) Financial data, corporate governance data, and ownership data are obtained from the CSMAR Database; (2) Producer Price Index (PPI) and Fixed Asset Investment Price Index are retrieved from the China Statistical Yearbook; (3) Digital transformation-related textual data are collected from the annual reports and social responsibility reports of listed companies disclosed on the CSMAR Database and official company websites. All relevant data and supplementary materials are available from the corresponding author upon reasonable request.
Acknowledgements
We appreciate the help and hard work of the Editor-in-Chief (Xiao-Ping Chen), Lead Guest Editor (Fang Lee Cooke), and Guest Editor (David Ahlstrom). Meanwhile, we are also grateful to the two anonymous reviewers for their insightful comments and suggestions. This work was supported by the Humanities and Social Sciences Research Project of the Ministry of Education of China (25YJC790098) and the Major Program of the National Social Science Foundation of China (22&ZD182).
Di Wang (wangdi@neuq.edu.cn) is a lecturer and graduate supervisor at Northeastern University. His primary research interests lie at the intersection of digital economy, industrial development, and sustainable transformation. He has authored more than 10 peer-reviewed articles in various leading scholarly journals, including Management and Organization Review, International Review of Economics and Finance, Finance Research Letters, and Amfiteatru Economic.
Bin He (hebin7904@163.com) is an associate professor and graduate supervisor at China’s Public Sector Economy Research Centre, Jilin University. He has published widely in top-tier management and environmental journals and led several national-level research projects. His research interests center on the intersection of digital technology, green innovation, and manufacturing upgrading, with a focus on providing practical solutions for the high-quality development of Chinese enterprises.
