Can Digital Transformation Drive Green Innovation in China’s Construction Industry under a Dual-Carbon Vision?

Abstract

Against the backdrop of increasing global environmental pollution and energy consumption, green innovation is necessary to achieve green transformation. As an industry with a huge demand for resources and energy consumption, the construction industry shoulders the mission of the times to promote green innovation to enhance the ability of sustainable development. Digital technology provides new opportunities for green innovation in the construction industry. However, the impacts and mechanisms of digital transformation driving green innovation have not been thoroughly studied. In this paper, 121 listed companies in China’s construction industry are selected as a sample from 2011 to 2021, and a total of 1331 annual observations are obtained, and the impact and mechanism of digital transformation on construction enterprises’ green innovation are empirically analyzed by establishing regression models. The study indicates that digital transformation can facilitate green innovation in construction companies by enhancing corporate risk-taking and improving corporate governance. Compared with non-state-owned enterprises, state-owned enterprises have more endogenous incentives for green transformation based on multiple pressures, which to some extent weakens the driving role of digital transformation. The driving effect of enterprises’ digital transformation is more significant when the intensity of regional environmental regulation is high. This paper examines how the digitization of construction enterprises can lead to new greening ideas from the perspective of green innovation. It provides an important theoretical basis and decision-making reference to support the construction industry in its digital transformation and realize the goal of “dual carbon”.

1. Introduction

The escalating global environmental pollution poses a grave threat to the Earth’s ecosystem and sustainable human development [1]. To combat the growing severity of the climate crisis, nations globally have established “carbon neutral” goals that will guide their carbon mitigation tactics over the next three decades. China first suggested a “dual-carbon” objective at the United Nations General Assembly’s 75th session., aiming to peak its carbon dioxide emissions by 2030 and achieve carbon neutrality by 2060 [2]. Following this, the “14th Five-Year Plan” solidified the objective of “dual carbon” and made it abundantly evident that it will encourage the greening of significant sectors and regions and accomplish the green and sustainable growth of businesses. In March 2022, the 14th Five-Year Plan for the Development of Building Energy Efficiency and Green Buildings outlined a roadmap to reduce carbon emissions in the construction industry. The plan called for accelerated implementation of digitalization, enhancement of low-carbon innovation capacity for enterprises, and upgrading low-carbon technology levels [3].

Under the “dual-carbon” goal, how to realize green, low-carbon sustainable development has become a key issue, in which green innovation is the key. Green innovation is a key tactic for creating a win-win situation for social and economic advancement. It may help companies balance economic and environmental benefits [4]. Meanwhile, digital transformation has become a primary driver pushing building industry growth due to the ongoing integration of information technologies including big data, blockchain, cloud computing, and artificial intelligence [5]. The enhancement of construction enterprises’ ability to innovate in green technology cannot be disconnected from the support of the internal organizational elements. The current digital information technology has profoundly changed the combination of production factors of construction enterprises. Also, it provides a valuable opportunity and environment for construction enterprises to innovate in green technology [6]. The building business is more energy-intensive and has a higher negative environmental effect than other sophisticated ecologically favorable industries [7]. Thus, digital technology is instrumental in facilitating construction enterprises’ transition from traditional high-energy-consuming and high-emission development practices to a green, low-carbon, and sustainable development mode. The construction industry performs an enormous part in the digital transformation of enterprises, making it imperative to promptly and accurately evaluate the industry’s actual progress. Additionally, it is essential to examine the impact of digitalization on green innovation within the construction industry at a micro level.

Can the digitalization efforts undertaken by Chinese construction businesses facilitate the advancement of green innovation within the framework of the “dual-carbon” strategy? What are the underlying mechanisms? Is there a differential influence of digital transformation on green innovation in construction enterprises in different contexts? The examination of these matters is unquestionably of major theoretical and practical import. Based on this, this study constructs a mechanism for the effects of digital transformation on the green innovation of construction enterprises under the “dual-carbon” vision and empirically tests a sample of listed enterprises in China’s construction industry from 2011 to 2021. The mediating influences exerted by digital transformation and green innovation in construction enterprises are examined from two dimensions: corporate governance capabilities and the level of corporate risk-taking. Further examine the variation in the above relationships under two different external conditions, namely different property rights and environmental regulation. The marginal contributions of this study are primarily in the following areas: first, we propose a theoretical and analytical framework to examine the link between digital transformation and green innovation within construction enterprises. By providing empirical evidence, we aim to contribute to the existing body of research on the determinants of green innovation in the construction industry. Second, we found that enterprise governance capacity and the level of risk-taking of enterprises are two important channels that influence the digital transformation of construction enterprises on improving the level of green innovation. These findings offer compelling evidence in support of promoting the digital transformation of construction enterprises as a means to foster the advancement of green innovation. Third, our analysis reveals that the advancement of green innovation within construction enterprises via digital transformation is subject to the varied impact of property rights characteristics and the level of regional environmental regulations. This result offers novel insights into the strategies that construction enterprises can adopt to enhance the promotion of green innovation through digital transformation. Finally, the outcomes of this study offer a significant theoretical foundation and decision-making guidance for the government in devising and executing digital transformation and green innovation strategies for construction enterprises. These strategies will facilitate the advancement of the construction industry’s green transformation and achieve the “dual carbon” objective.

Firstly, there are fewer studies on the micro-mechanisms of digital transformation affecting green innovation in enterprises; secondly, at the micro level, studies on the impact of digital transformation on green innovation have mainly focused on exploring the impact of digital transformation on green innovation in manufacturing industry, while there is very little literature focusing on the perspective of the construction industry. Therefore, this paper fills this gap by exploring the impact of the digital transformation of construction enterprises on green innovation and its mechanism. It provides an important theoretical basis and decision-making reference for supporting digital transformation and green innovation in the construction industry and realizing the sustainable development of the construction industry under the dual-carbon goal.

2. Literature Review and Research Hypotheses

2.1. Literature Review

2.1.1. Green Innovation

Green innovation encompasses a range of improvements across several domains, encompassing product and process innovation, management strategies, and technology innovations in areas such as energy efficiency, pollution avoidance, waste recycling, green product design, and environmental management within the business sector. It aims to promote environmental sustainability while catering to the needs of the consumers [8]. Enterprise green innovation is a complex activity of enterprises, specifically referring to the innovative activities of enterprises through the optimization of resource utilization and the use of new technologies or processes, thereby reducing environmental pollution and promoting management and sustainable development [9]. Enterprises’ continuous strengthening of green innovation initiatives and proactive participation in the implementation of green innovation strategies can effectively improve environmental performance [10]. Focusing on domestic and international research on enterprises’ green innovation, relevant studies are categorized into two main areas: one is an examination of the factors influencing enterprises’ green innovation, with one perspective focusing on enterprises’ external conditions, such as environmental regulation, market demand, and stakeholder pressure [11], and the other focusing on enterprises’ culture [12], enterprise capabilities (operational, risk-resilient, and financial) [13], and entrepreneurship [14]. The second is to explore the mode and path of green innovation of enterprises, on the one hand, to explore the Green innovation ecological model and its application strategy in enterprise management [15], on the other hand, to explore the path of the realization of green innovation of enterprises from the aspects of sustainable development, multivariate configuration, and policy-driven [16].

2.1.2. Digital Transformation and Green Innovation for Enterprises

The pervasive integration of digital technology has led to the comprehensive incorporation of digital transformation in enterprises, including several facets such as production, decision-making, and operational interaction. Optimizing technological elements is the key to the sustained competitiveness of enterprises [17]. Digital transformation has changed the way redundant resources are utilized in enterprises, helping them access more resources and obtain more flexible ways of cooperation [18]. Through the thorough incorporation of digital technology in diverse enterprises’ production management processes, the ability to integrate resources and process information can be notably enhanced [19], and then realize the core business of the enterprise “technology penetration”. It has been observed through existing literature that digital development is not only the primary focal point of enterprises’ strategic transformations but also a progressively crucial source of innovative vitality for enterprises. Regarding R&D efficiency, enterprises rely on building a digital construction platform to achieve internal information sharing and knowledge integration, strengthen environmental protection cooperation among enterprises, optimize green innovation resources, and then improve enterprise green innovation efficiency [20]. Regarding costs and expenses, increased transparency of information resulting from digital transformation mandates that enterprise managers elevate their operational quality. This action facilitates the creation or optimization of both digital analytical decision-making and management control systems, which ultimately leads to a reduction in the company’s agency costs [21]. For resource allocation, digital technology has expanded the data mining space for more precise strategic target decomposition, investment and financing decisions, and market segmentation strategies, which enhances overall operational efficiency for enterprises [22]. Digital technology can allocate land, capital, and labor to areas that effectively generate user value. It also enables timely adjustment and optimization of factor allocation based on feedback results [23]. Although digital transformation can facilitate innovation activities in enterprises, there is no academic consensus regarding the impact of such transformation on green innovation. As Li and Jia [24] note, digital technology advancements can prompt enterprises to purchase new production equipment. However, during the transitional phase, enterprises may increase resource extraction and energy loss to increase production, potentially reducing their green innovation practices rapidly. The study of Ghasemaghaei and CalicG [25] also pointed out that the impact of data volume of big data in improving the innovation performance of enterprises is not significant.

2.1.3. Green Innovation in Construction Enterprises

In the realm of building enterprises, scholarly investigations have been undertaken on green innovation to examine the subject matter from two primary vantage points. The initial perspective centers on the exploration of the determinants that propel environmentally sustainable innovation within construction companies. For instance, Yin, Dong [26] and Li, Liu [27] analyzed the effect of green innovation in construction enterprises from the perspective of network, management, and environmental policy, and conducted empirical evidence. Secondly, Chen, Wang [28] conducted a study that specifically examined the green innovation mechanism of construction businesses. The study concentrated on three key parts: target-level components, elements, and network-level elements. The researchers aimed to investigate and understand the green innovation mechanism employed by construction enterprises. Based on an exploratory single-case study, Chen and Wang [29] refined the resource orchestration theory in the Chinese context at the actor and micro-path levels. They utilized this framework to examine the green innovation behavioral mechanisms of Chinese construction enterprises participating in the urban carbon neutrality process. Ref. [30] utilized evolutionary game theory as a basis to conduct numerical simulations that assess the effect of various determining factors on selecting cooperative strategies for green technology innovation inside building company settings.

So brief, the examination of the correlation between digital transformation and green innovation under the framework of “dual-carbon” is a subject matter that elicits divergent perspectives within the academic discourse. Evidence is scarce about the micro-enterprise level. Furthermore, there exists a lack of scholarly material about the correlation between these two entities as observed within the construction sector. Current research has not yet conducted a detailed analysis of the inherent mechanisms underlying digital transformation and green innovation in construction firms, nor have there been sufficient empirical studies to validate their driving effects and intrinsic mechanisms across a large sample of construction enterprises. Furthermore, there is a notable absence of comprehensive analysis of the internal viewpoint of enterprises. Therefore, there is a necessity to further enrich and expand the mechanism and scope of the study. Simultaneously, it is essential to engage in broader discourse regarding possible discrepancies in the influence of digital transformation on green innovation inside construction enterprises across various scenarios.

2.2. Research Hypotheses

2.2.1. Digital Transformation and Green Innovation

Green innovation in enterprises is a complex activity. Compared to conventional innovation, the process of green innovation entails collaboration, multilevel interaction, and knowledge sharing [31], necessitates higher financial investment, and poses greater risks. Digital transformation significantly improves the resource integration and information processing capabilities of enterprises through the in-depth application of digital technology to various production management processes and improves the innovation level of enterprises while optimizing their innovation resources [32]. Enterprises need to rely on a variety of internal elements, including their technological capabilities, to innovate and maintain a competitive advantage [33], and digital transformation provides the relevant foundation for enterprises to engage in green innovation [34,35]. First, new technologies represented by artificial intelligence, blockchain, cloud computing, big data, etc. can help construction enterprises provide joint innovation systems, platforms, and tools for research and development and innovation [36]. By leveraging these technologies, enterprises can enhance the efficiency of knowledge reorganization during the innovation process, thereby augmenting their capacity for green innovation [37]. Secondly, digital transformation can promote data integration, sharing, and value release, enhancing the operational capabilities and management efficiency of green innovation activities in construction enterprises. Digital transformation can help optimize the internal processes of construction enterprises, optimize resource allocation and scheduling, and reduce the sunk costs and failure risks of innovation [38]. Under the influence of digital transformation, the financial and human resources required for green innovation in construction enterprises are better optimized and allocated [39]. Simultaneously, the in-depth application of digital technology improves the disclosure level and information transparency of construction enterprises, promotes the refined management of construction enterprises, and accumulates a good material foundation for green innovation. Third, the digital transformation of enterprises can optimize the allocation of internal and external resources and stimulate the growth potential of enterprises, thereby providing endogenous power for green innovation activities [40]. The digital platform can integrate information on green building materials and advanced technologies, promote the wide application of environmentally friendly materials and energy-saving technologies in construction projects, and support the green transformation of the construction industry [41]. The in-depth application of digital technology helps to enhance the interconnection and comprehensive integration of information systems, improve enterprises’ ability to integrate upstream and downstream resources in the supply chain as well as their ability to respond quickly to changes in market demand, and at the same time refine new information and knowledge. This enables construction enterprises to jump out of the one-dimensional growth model and stimulate higher-quality green innovation performance [42]. Based on the above analysis, the following research hypotheses are proposed:

H1. 

Digital transformation can promote green innovation in construction enterprises.

2.2.2. The Mediating Effect of Enterprise Governance Capacity

Enterprise digital transformation can enhance its management process, decrease management redundancy, optimize the organizational structure, and eliminate information barriers in internal management. This leads to significant improvements in management resilience and governance while creating more scientific and efficient management of green innovation resources, ultimately resulting in an improved resource utilization rate and smoother transformation of green outputs [36]. Thus, the efficiency of a company’s green innovation can be enhanced. On one hand, digital technology can reshape the management process of construction enterprises, and innovate and restructure the internal competition mode, cooperation mechanism, and management boundaries of construction enterprises, to effectively improve the management capacity of construction enterprises [43]. It can also enhance the sensitivity of construction enterprises to the green innovation external market [44], ensuring that construction enterprises can seize the opportunities of green innovation and avoid the risks of green innovation, thus increase the success rate of green innovation in construction enterprises. Another aspect, digital transformation has the potential to optimize the organizational structure of construction enterprises [45], resulting in a flatter and more flexible structure. This could increase the efficiency of enterprise governance, allowing construction enterprises to uncover potential customer needs better and quickly respond to supplier feedback in complex external environments. To effectively control the risk of green innovation in construction enterprises to enhance the success rate of green innovation and reduce the waste of innovation inputs so that the output of green innovation under the same inputs increases significantly [46]. Based on the above analysis, the following research hypotheses are proposed:

H2. 

Digital transformation in construction enterprises can contribute to increased green innovation by improving corporate governance.

2.2.3. The Mediating Effect of Enterprise Risk-Taking Level

The implementation of digital transformation has the potential to foster an increased capacity for risk-taking among organizations. Simultaneously, green innovation, being a high-risk investment endeavor, is more likely to be promoted by enterprises that exhibit a greater capacity towards risk-taking [47]. The level of enterprises’ risk-taking is influenced by their ability to acquire resources. On the one hand, digital transformation facilitates the integration of beneficial resources from all stakeholders and alleviates the issue of information asymmetry [48]. Relying on digitization, construction enterprises form a cross-field networked collaborative innovation platform to make green R&D, design, and green construction management more precise by sharing, integrating, utilizing, and recreating information, and knowledge to effectively avoid and cope with unknown risks [49]. On the other hand, digital transformation reduces agency conflicts within construction enterprises by exerting governance effects [50], thereby increasing the level of enterprise risk-taking. Digital transformation enhances process and information transparency in construction enterprises, curbs opportunistic behavior, and effectively mitigates agency conflicts between shareholders and management [51]. At the same time, digital transformation enhances the external market’s supervision of construction enterprises, greatly improves the information disclosure ability of construction enterprises, increases the channels of external supervision, and improves the strength and effectiveness of the external market’s supervision of enterprises [52]. Based on the above analysis, the following research hypotheses are proposed:

H3. 

Digital transformation in construction enterprises can contribute to increased green innovation by improving enterprise risk-taking.

The Hypothetical research model see Figure 1.

3. Research Design

3.1. Samples Data

Firstly, this study samples enterprises engaged in the construction of engineering projects, technology research and development, sales of construction materials, engineering planning and design, and engineering project management from A-share listed companies. Further, it should be noted that most of the samples did not have signs of digital transformation and green innovation before 2011 after data collection, which is not meaningful for the study. Therefore, this study selects Chinese A-share listed construction enterprises from 2011–2021 as the research sample. The financial data of the enterprises are from the CSMAR database and wind database, and the data related to green patents are from the CNRDS database. The initial data are preprocessed as follows: (1) All ST and *ST samples are excluded. (2) Remove the missing core data samples. (3) The missing data are linearly interpolated. (4) To reduce the impact of extraneous outliers, the continuous variables are shrink-edged by 1% up and down.

3.2. Model Setup

3.2.1. Main Model

Benchmark regression modeling enables explicitly establishing the relationship between independent and dependent variables, facilitating direct testing of research hypotheses. In further analyses, the researcher may include interaction terms, nonlinear terms, or other complex model structures to investigate the relationships between variables in more detail. In studies of green innovation and digital transformation, benchmarking regressions can help identify the direct impact of digital transformation on green innovation while controlling for other known influences. This paper presents a benchmark regression model to verify the impact of construction enterprise digital transformation on green innovation, based on the research of most scholars [53,54]. The model is constructed as follows:

$${Gpf}_{it}=\alpha +\beta { \times Digi}_{it}+\mathsf{\gamma }\times {controls}_{it}+{\lambda }_t+{\mu }_i+{\epsilon }_{it}$$

(1)

$${Gps}_{it}=\alpha +\beta { \times Digi}_{it}+\mathsf{\gamma }{ \times controls}_{it}+{\lambda }_t+{\mu }_i+{\epsilon }_{it}$$

(2)

$${Gp}_{it}=\alpha +\beta \times {Digi}_{it}+\mathsf{\gamma }\times {controls}_{it}+{\lambda }_t+{\mu }_i+{\epsilon }_{it}$$

(3)

where Gpf_it, Gps_it, and Gp_it represent green innovation indicators as explanatory variables, Digi_it is an enterprise digital transformation indicator, which is the core explanatory variable in this paper; controls_it is a set of control variables; λ_t represents a time-fixed effect; μ_i represents an individual fixed effect; and ε_it represents a random error. According to the previous theoretical model, if the coefficient β of Digital is significantly positive, it indicates that digitalization mainly promotes green innovation in enterprises.

3.2.2. Intermediation Models

The previous section of this paper presents an empirical model demonstrating that the digital transformation of construction enterprises can promote green innovation. The next step is to further clarify the mechanism of action between the two. This paper focuses on two perspectives: enterprise governance capacity and risk-taking level. To verify the hypotheses based on the three perspectives, this paper utilizes Ning, Jiang’s [12] research and constructs the following mediation models:

$$Y_{it}={\alpha }_4+{\beta }_4\times {Digi}_{it}+{\mathsf{\gamma }}_4\times {controls}_{it}+{\lambda }_t+{\mu }_i+{\epsilon }_{it}$$

(4)

$${Med}_{it}={\alpha }_5+{\beta }_5\times {Digi}_{it}+{\mathsf{\gamma }}_5\times {controls}_{it}+{\lambda }_t+{\mu }_i+{\epsilon }_{it}$$

(5)

$$Y_{it}={\alpha }_6+{\beta }_6\times {Digi}_{it}+\mathsf{\delta }\times {Med}_{it}+{\mathsf{\gamma }}_6\times {controls}_{it}+{\lambda }_t+{\mu }_i+{\epsilon }_{it}$$

(6)

Med_it denotes the mediator variable; Y_it denotes the explanatory variables Gpf_it, Gps_it, and Gp_it; and the other variables are as above. First, we examined the effect of digital transformation on green innovation by observing the coefficient β₄ in Equation (4). Second, we examined the effect of digital transformation on mediation by observing the coefficient β₅ in Equation (5). Third, we examined the effect of digital transformation and mediation on green innovation by observing the coefficients β₆ and δ in Equation (6). The mediating effect must meet the following requirements: (1) β₄ is statistically significant; (2) β₆ is not significant when both β₅ and δ are significant, then mediation is sufficient, and partial mediation exists if β₆ is significant but less than β₄. The model also controls for the effects of industry fixation and year fixation.

3.3. Variable Definitions

Explanatory variable: Digital transformation of enterprises. Within the contemporary period characterized by rapid advancements in digital technology, organizations have increasingly turned to digital transformation as the principal way to achieve substantial progress and growth. The yearly reports of enterprises are more likely to reflect the informational nature of digital growth since they serve as comprehensive and directive documents. Researchers frequently utilize “the number or percentage of digital-related keywords in companies’ annual reports” to measure digital transformation [55,56]. Hence, the primary objective of this research is to assess the extent of digital transformation inside organizations by an examination of the frequency at which phrases related to digital development appear in annual reports. Simultaneously, to address the issue of a right-skewed distribution, the frequency of the term “digital transformation” is incremented by one and subsequently transformed using the natural logarithm function, resulting in the creation of the core explanatory variable, denoted as “Digi”.

Explained variable: Green Innovation. There are multiple benefits associated with using green patent filings as a metric for assessing the extent of green innovation. First, this strategy possesses quantitative attributes and effectively captures the outcomes of enterprises’ green innovation activities in a manner that is easily comprehensible. Second, patents offer a distinct advantage over measures like R&D investment, as they can be categorized. Categorization studies based on various attributes can effectively reflect the implications and value contributions of different types of innovative activities [57,58]. In this research, the level of green innovation is assessed by incorporating the overall number of Green Patents filed in a given year, which is augmented by 1 and then subjected to a natural logarithmic transformation. This approach is adopted to better align with the real-world circumstances of enterprises. Further, it is divided into patents for green inventions (Gpf) and patents for green utility models (Gps). In the robustness test, the share of green patents in total patents acts as an explanatory variable.

Control variables: Enterprise characteristics and capital structure may influence green innovation. As an enterprise gains experience, its management becomes more efficient and skilled, and its contribution to green innovation increases. Therefore, enterprise age (Age) may have a potential impact on green innovation. Enterprises of different sizes possess significantly different green innovation capabilities [59]. According to [60], we use the logarithm of total assets as Size. Return on assets (Roa) is a financial indicator that assesses the profitability of an enterprise. The higher the profitability, the greater the ability to engage in green innovation [61]. Based on previous literature, we define Roa as the net profit after tax divided by the average balance of total assets at the beginning and end of the period, expressed as a percentage. Ref. [62] confirm a significant association between equity concentration (Top1) and green innovation. Top1 is measured using the percentage of shareholding of the largest shareholder as a proxy variable. In addition, this paper also draws on the studies of scholars such as [63], and [64] to introduce cash flow from operating activities (Cash), financial leverage (Lev), growth (Gro), director structure (Boa), and two positions (Dual) of the enterprise as control variables. The specific variable definitions are listed in

Table 1. Defining Variables.

	Variable Symbol	Variable Definition
Explanatory variable	Digi	LN (Word frequency of keywords related to digitization of annual reports of enterprises + 1)
Explanatory variable	Gpf	LN (Number of patent applications for green inventions + 1)
	Gps	LN (Number of green utility model patent applications + 1)
	Gp	LN (Number of Green Patents filed + 1)
Control variable	Size	LN (Total assets at year-end)
	Lev	Total liabilities/total assets
	Roa	Net profit/(the average balance of total assets at the beginning and end of the period)
	Gro	Sales revenue growth rate
	Top1	The shareholding ratio of the largest shareholder
	Age	LN (Year of observation—year of listing)
	Dual	Whether the chairman and general manager are combined, if combined, take 1, otherwise take 0.
	Soe	State-owned companies are assigned a value of 1, while non-state-owned companies are assigned a value of 0.
	Cash	Net cash flow from operations/total assets
	Boa	Number of independent directors/total number of board members

.

3.4. Descriptive Statistics

Table 2. Descriptive statistics.

Variable	N	Mean	Med	Stddev	Min	Max
Gpf	1331	0.133	0	0.463	0	4.836
Gps	1331	0.159	0	0.520	0	4.727
Gp	1331	0.223	0	0.648	0	5.288
Digi	1331	0.807	0	1.068	0	5.829
Size	1331	23.73	23.44	1.634	16.18	28.50
Lev	1331	0.678	0.715	0.167	0.0280	1.347
Roa	1331	0.0220	0.0220	0.0420	−0.431	0.256
Cash	1331	0.0100	0.0130	0.128	−3.224	0.736
Gro	1331	12.85	0.118	411.1	−4.216	14883
Top1	1331	0.394	0.380	0.165	0.0450	0.819
Age	1331	2.581	2.773	0.670	0	3.401
Boa	1331	0.384	0.364	0.0690	0.222	0.800
Dual	1331	0.134	0	0.340	0	1

displays the statistical data for the variables. The construction enterprises exhibit a range of green patent applications, with the maximum count being 197 and the minimum count being 0. The mean value of green patents is 1.156, indicating a very limited extent of green innovation within the respective domain. Furthermore, there exist notable variations in degrees of innovation among enterprises. According to the Digital Transformation Index of construction enterprises, the range of keyword disclosures about the digital development of enterprises during the sample period varied from 0 to 339, with an average of 4.516. This observation aligns with the 2019 Digital Transformation Index of Chinese Enterprises, which indicated that a mere 9% of total disclosures were related to the substantial digital transformation of enterprises. These results suggest that China’s current pace of digital transformation remains sluggish.

4. Empirical Results and Explanations

4.1. Benchmark Regression Analysis

To determine hypothesis H1 of this study, an empirical test is conducted according to Equation (1), and

Table 3. Benchmark regression results.

Variable	Explained Variable: Green Innovation
	(1) Gpf	(2) Gps	(3) Gp	(4) Gpf	(5) Gps	(6) Gp
Digi	0.056 ***	0.082 ***	0.092 ***	0.056 ***	0.079 ***	0.090 ***
	(0.015)	(0.017)	(0.020)	(0.014)	(0.016)	(0.019)
Constant	0.081 ***	0.076 **	0.134 ***	0.483	−0.097	0.180
	(0.030)	(0.032)	(0.040)	(0.422)	(0.508)	(0.590)
Obs	1331	1331	1331	1331	1331	1331
R-squared	0.039	0.061	0.059	0.079	0.083	0.089
N	121	121	121	121	121	121
year fix	YES	YES	YES	YES	YES	YES
Id fix	YES	YES	YES	YES	YES	YES
Controls	YES	YES	YES	YES	YES	YES

Note: In parentheses, coefficients are clustered using the t-statistic. The symbols ** and *** are used to denote significance levels of 10% and 5%, respectively. Same below.

shows the regression results of digital transformation and green patents in construction enterprises. Columns (1) to (3) present the projected results of the fixed-effects model, accounting for individual and temporal effects, without the inclusion of control variables. In column (3), the computed regression coefficient between corporate digital transformation and green patents is 0.092, indicating statistical significance at the 1% level. We also find a significant positive impact of enterprises’ digital transformation on green inventions and utility patents as well. Upon including control variables in columns (4)–(6), the regression coefficient for the digital transformation of firms and green patents is estimated to be 0.090, which is statistically significant at the 1% level. In addition, digital transformation also remains capable of boosting the number of green invention patents and green utility model patents in enterprises. The results obtained from the benchmark regression analysis indicate that the implementation of digital transformation in construction enterprises has a substantial impact on promoting green innovation. This effect remains consistent, irrespective of the inclusion or exclusion of fixed factors in the analysis. This implies that enterprises that have undergone digital transformation are more inclined to increase their investment in green innovation initiatives to align with the emerging requirements of the digital economy and sustainable development, hence providing support for Hypothesis 1.

4.2. Robustness Check

4.2.1. Substitution of Variables

Firstly, to address the varying degrees of innovation among enterprises, we conducted a reevaluation of the assumptions and employed the proportion of green patents relative to the overall number of patent applications as a metric for quantifying the extent of green innovation inside these enterprises. The outcomes are presented in columns (1) through (3) of

Table 4. Robustness test results (1): replacement of explanatory variables.

Variable	Explained Variable: Green Innovation
	(1) Gpfrio	(2) Gps-rio	(3) Gprio	(4) Gpf	(5) Gps	(6) Gp
Digi	0.017 ***	0.019 ***	0.017 ***
	(0.006)	(0.004)	(0.005)
L. Digi				0.019	0.042 **	0.041 *
				(0.015)	(0.019)	(0.022)
Constant	0.208	0.149	0.277	0.243	−0.094	−0.040
	(0.229)	(0.154)	(0.187)	(0.428)	(0.632)	(0.685)
Obs	1331	1331	1331	1210	1210	1210
R-squared	0.031	0.051	0.049	0.046	0.053	0.053
N	121	121	121	121	121	121
Year fix	YES	YES	YES	YES	YES	YES
Id fix	YES	YES	YES	YES	YES	YES
Controls	YES	YES	YES	YES	YES	YES

Note: The coefficients of the control variables are omitted; The symbols *, **, and *** are used to denote significance levels of 10%, 5%, and 1%, respectively.

. Secondly, to better control for the individual heterogeneity characteristics of enterprises, this paper lags the dependent variable digital transformation (Digi) by one period while using a fixed-effects model, and the regression results are denoted as L. Digi. The outcomes are presented in columns (4) through (6) of Table 4. The comprehensive results show that digital transformation can still significantly promote green innovation in enterprises after considering various influencing factors, thus confirming the reliability of the basic conclusions of this study. However, there is no significant association between lagging digital transformation and patents for green inventions. Based on prior literature [65], this suggests that enterprises may prioritize technological innovations that require less technical complexity for strategic reasons, ultimately hindering their ability to improve technological innovation quality.

4.2.2. Substitution of Explanatory Variables

This study considers the Digi indicators as continuous variables. It aims to reconstruct the indicators measuring the amount of digitalization in firms using the following methods: (1) drawing on Yang, Yuan [66], we replaced the explanatory variables of enterprises’ digital transformation with dummy variables. Subsequently, the digital transformation indicators were segregated based on the median criterion. Enterprises with a high degree of digital transformation obtain a value of 1 if their word frequency exceeds the median. Conversely, enterprises with a low level of digital transformation, below the median, are assigned a value of 0. The variable is entered as Digi_Adj. (2) The terminology of enterprise digitization can be categorized into five main dimensions: cloud computing, the Internet, artificial intelligence, big data, and the Internet of Things. However, there are differences in the number of terms covered in the lexicon for each dimension, which may affect the ability of the digitization terminology lexicon to capture enterprise digitization in different dimensions. Based on this, this study first constructs digital segmentation indicators for each dimension separately, i.e., Digital 1–5; then, according to equation (4), the five segmentation indicators are standardized by a sub-annual deviation to eliminate the magnitude, and the processed indicators are accumulated to obtain the new enterprise digitalization indicator, which is denoted as Digi_Std. (3) The segmented metrics’ five dimensions underwent principal component analysis, retaining factors with eigenvalues exceeding 1. The new digitized metric was called Digi_PCA. The results of the testing are presented in

Table 5. Robustness test results (3): replacement of explanatory variables.

Variable	Explained Variable: Green Innovation
	(1) Gpf	(2) Gps	(3) Gp
DigiAdj	0.058 ***	0.075 ***	0.091 ***
	(0.018)	(0.021)	(0.027)
Digisd	0.152 **	0.267 **	0.288 **
	(0.076)	(0.105)	(0.113)
DigiPCA	0.022 *	0.042 **	0.044 **
	(0.013)	(0.018)	(0.019)
Obs	1331	1331	1331
N	121	121	121
Year fix	YES	YES	YES
Id fix	YES	YES	YES
Controls	YES	YES	YES

Note: The three indicators with replacement are regressed separately against the explanatory variables, omitting the control variable coefficients, and combined into a table. The symbols *, **, and *** are used to denote significance levels of 10%, 5%, and 1%, respectively.

. Regardless of the specific measurement approach employed to assess the extent of digital transformation in enterprises, Digi_x’s coefficient exhibits a statistically significant positive effect. This finding suggests that digital transformation has a substantial and beneficial impact on the green innovation capabilities of enterprises. Thus, it can be determined that the primary conclusion of this study is robust.

$${Digital}_{\_si}={\displaystyle \frac{{Digital}_i-\mathrm{m}\mathrm{i}\mathrm{n}\left({Digital}_i\right)}{\max \left({Digital}_i\right)-\mathrm{m}\mathrm{i}\mathrm{n}\left({Digital}_i\right)}}$$

(7)

$${Digital}_{std}={\sum }_{i=1}^5{Digital}_{\_si}$$

(8)

4.3. Endogeneity Test

4.3.1. Instrumental Variable

Considering unobservable omitted variables and possible reverse causation problems, this paper constructs instrumental variables and applies 2SLS for regression.

First, in fundamental terms, the degree of digital development in the area where the companies lie may have some effect on the degree to which their operations are digital, meeting the relevance condition. On the other hand, the exogenous condition can be satisfied because the digital development stage of other companies in the identical area has no direct impact on the green innovation of companies. Drawing on LI, LAN [67], the study used the average of the degree of digital transformation of other enterprises in the same region of the enterprises as an instrumental variable for the level of digitization of the enterprises in that year (L.IV_lnDigital_Envir ^C).

Second, we selected postal and telecommunication data as instrumental variables for each city in 1984. The communication methods utilized by enterprises’ location in previous developmental stages, considering the level of progress and social preferences, impact the adoption and integration of information technology by said enterprises during the observed period, which satisfies the relevancy criteria. Furthermore, infrastructure such as postal and telecommunications primarily cater to public communication services. Thus, it does not make a direct contribution to the green innovation process of companies, which is exogenous. Additionally, the 1984 data on post and telecommunications for each city are cross-sections and unsuitable as instrumental variables for panel data, this study opted for the cross-multiplier term that combines the total regional postal and telecommunications businesses per 10,000 people with the number of Internet broadband ports in each province in 1984 as an instrumental variable for measuring the level of digital transformation of enterprises during that period (L.IV_lnPost_Internet ^C).

Table 6. Endogeneity Test.

Variable	(1)	(2)	(3)	(4)
	Digi	Gp	Digi	Gp
L.IV_lnDigital_Envir C	0.033 ***
	(10.52)
L.IV_lnPost_Internet C			0.187 ***
			(7.38)
Digi		0.288 ***		0.438 ***
		(4.59)		(4.37)
Controls	YES	YES	YES	YES
_cons	0.352	0.453 ***	0.362	0.385 **
	(1.47)	(2.68)	(1.49)	(2.20)
N	1331.000	1331.000	1327.000	1327.000
r2_a	0.094	0.095	0.058	0.058
Kleibergen-Paap rk LM	102.519 ***		53.310 ***
Cragg-Donald Wald F	102.055		47.356
Stock-Yogo	16.38		16.38

The symbols ** and *** are used to denote significance levels of 10% and 5%, respectively.

displays the results of the regressions using the instrumental variables method. The instrumental variable regression results using two-stage least squares indicate that the Kleibergen-Paap rk LM statistics are both significant at the 1% level, rejecting the initial hypothesis that the instrumental variables are under-identified; The Cragg-Donald Wald F statistics exceed the Stock-Yogo weak instrumental variable identification F-test threshold at the 10% significance level, rejecting both the initial hypothesis of weak instrumental variable and demonstrating that the selection of the instrumental variable is valid. The initial step of regression modeling reveals that the digital transformation of firms has a statistically significant positive relationship with the instrumental variable, as indicated by the significance level of 1%. The findings from the second stage of columns (2) and (4), demonstrate that the coefficients associated with the digital transformation of enterprises remain statistically significant after addressing potential endogeneity concerns. This indicates that technological changes in construction enterprises continue to have a strong positive impact on green transformation. Consequently, the main conclusions drawn in this paper are deemed robust.

4.3.2. Propensity Matching Method

The utilization of propensity score matching (PSM) can serve as a partial solution for endogenous bias that emerges due to issues related to sample selection. The samples are categorized into two distinct groups, namely the treatment group and the control group, based on the degree of digital transformation seen inside the construction enterprises. The demarcation line for categorizing the treatment and control groups in the construction sector is determined by the mean yearly digital transformation level. The treatment group consists of those entities that have a digital transformation level above the sample mean, while the control group comprises entities with a digital transformation level below the mean. Propensity score matching with Size, Lev, Roa, Gro, Top1, Age, Dual, Soe, Cash, and Boa as covariates.

Table 7. Endogeneity test: sample balance test results.

Variable	U/M	Treatment Group	Control Group	Bias (%)	Reducing Bias (%)	t	p
Size	U	24.028	23.669	22.0	91.2	3.08	0.002
	M	24.028	24.06	−1.9		−0.20	0.842
Lev	U	0.68365	0.67693	4.1	92.9	0.56	0.575
	M	0.68365	0.68413	−0.3		−0.03	0.975
Roa	U	0.02078	0.02261	−3.9	−27.9	−0.61	0.543
	M	0.02078	0.01844	5.0		0.54	0.588
Cash	U	0.01325	0.00972	3.2	97.1	0.39	0.700
	M	0.01325	0.01335	−0.1		−0.01	0.989
Gro	U	8.2217	13.853	−1.7	−41.4	−0.19	0.848
	M	8.2217	0.25869	2.4		1.00	0.316
Top1	U	0.37191	0.39856	−16.8	78.1	−2.26	0.024
	M	0.37191	0.37773	−3.7		−0.40	0.687
Age	U	2.5287	2.592	−9.7	46.5	−1.32	0.188
	M	2.5287	2.5625	−5.2		−0.56	0.578
Boa	U	0.38277	0.38484	−3.0	18.2	−0.42	0.674
	M	0.38277	0.38446	−2.5		−0.29	0.776
Dual	U	0.15612	0.12888	7.8	76.8	1.12	0.264
	M	0.15612	0.16245	−1.8		−0.19	0.851
Soe	U	0.46835	0.59232	−25.0	94.9	−3.51	0.000
	M	0.46835	0.46203	1.3		0.14	0.890

Note: The symbol U denotes the state of “Before matching”, whereas the symbol M indicates the state of “After matching”.

demonstrates that some covariates have significant t values before matching, whereas the t values are no longer significant after matching with the nearest neighbor matching technique. Furthermore, the standardized deviation of all covariates exhibits an average decrease of 66.4%, indicating a significant reduction in the standard deviation of the covariates after propensity score matching. This suggests that the endowment characteristics of digital transformation across different levels have been eliminated. In this paper, we estimate the average treatment effect (ATT) of green innovations driven by digital transformation of enterprises based on the new sample after matching, and the results are shown in

Table 8. Endogeneity test: average treatment effect estimates.

Variable	(1) Gpf	(2) Gps	(3) Gp
ATT	0.275 ***	0.354 ***	0.407 ***
standard error	0.032	0.036	0.045
t	8.67	9.86	9.26

Note: The symbols *** is used to denote significance levels of 1%, respectively.

. The average treatment effect of digital transformation in enterprises on green innovation is positive at a significant level of 1%, indicating that this process can substantially enhance sustainable development in the construction sector. Moreover, the fundamental conclusion of this article is robust.

4.4. Mechanism Test

4.4.1. Enterprise Governance Capacity Mechanisms

Drawing on existing literature [68], this paper uses enterprises’ total asset turnover (Emc) as a proxy variable for their governance capacity. If an enterprise has a higher total asset turnover, it indicates that the principal-agent cost in the enterprise is less and the governance of the enterprise is stronger. The findings presented in Column (1) of

Table 9. Estimation Results for Mechanism Testing (2)—Governance Capacity of Enterprises.

Variable	(1)	(2)	(3)	(4)
	Emc	Gp	Risk	Gp
Digi	0.032 ***	0.068 ***	0.006 ***	0.083 ***
	(0.008)	(0.016)	(0.002)	(0.019)
Emc		0.665 ***
		(0.109)
Risk				1.138 **
				(0.564)
Constant	1.118 **	−0.563	0.177 **	−0.021
	(0.475)	(0.609)	(0.088)	(0.607)
Obs	1331	1331	1331	1331
R-squared	0.142	0.146	0.151	0.094
N	121	121	121	121
Year fix	YES	YES	YES	YES
Id fix	YES	YES	YES	YES
Controls	YES	YES	YES	YES

The symbols ** and *** are used to denote significance levels of 10% and 5%, respectively.

indicate that the rate of digital transformation exhibits a coefficient of 0.032 and achieves statistical significance at the 1% level. This suggests that the implementation of digital transformation within construction enterprises plays a role in enhancing corporate governance. Specifically, a 1% increase in the level of digital transformation within enterprises corresponds to a 0.032 unit increase in corporate governance. Column (2) of Table 9 shows that enterprise governance capacity has a significant positive contribution to green innovation in enterprises. It appears that the presence of effective corporate governance skills plays a vital role in facilitating digital transformation and green innovation inside enterprises. The implementation of digitalization in construction enterprises has caused the power to optimize resource allocation, improving governance capacity. The latter, in turn, leads to an increase in the utilization of green innovation resources and a significant improvement in production benefits. Consequently, the efficiency of green innovation is promoted, thereby confirming hypothesis H2.

4.4.2. Risk-Taking Level Mechanisms

Increased risk-taking by enterprises implies greater uncertainty about their cash flows in future periods, and the volatility of corporate earnings is the most widely used measure of risk-taking [47]. This study employs the metric of profitability volatility as a means to assess the extent of risk exposure undertaken by enterprises (Risk). The method of calculation involves adjusting the earnings return rate by the industry average, followed by selecting the observation period of t − 1~t + 1 years to calculate the standard deviation of the earnings return rate. The influence of digital transformation on the risk-taking level of construction enterprises is examined according to Equation (7), and the regression results are detailed in column (3) of Table 9. The findings demonstrate a statistically significant and positive association between digital transformation and the level of risk-taking, suggesting that the implementation of digital transformation initiatives can increase the risk-taking capacity of enterprises. Columns (4) of Table 9 provide a deeper examination of the influence of enterprises’ risk-taking behavior on green innovation. The outcomes show that an increase in the level of risk-taking by enterprises can significantly contribute to their green innovation. The digital transformation of construction enterprises forms a networked collaborative innovation platform for the sharing of information, knowledge, and innovation resources, which makes green R&D, design, and green construction management more precise, thus effectively avoiding and coping with unknown risks, and improving the level of enterprise risk-taking and the success rate of green innovation. Hypothesis H3 is verified.

4.5. Heterogeneity Analysis

4.5.1. Nature of Property Rights

The incentive for green innovation among enterprises exhibits heterogeneity, potentially influenced by the characteristics of their property rights. Against the backdrop of active promotion of policies aimed at reducing carbon emissions, non-state-owned firms exhibit a greater inclination towards pursuing green transformation through the utilization of digital technology, as compared to their state-owned counterparts. To examine the possible impacts of property rights, we divided the sample into two categories: enterprises that are under state ownership and those that are not. Subsequently, we conducted group tests. The findings show that digital transformation still promotes green innovation in enterprises in both non-SOEs (non-state-owned enterprises) and SOEs (state-owned enterprises) samples. Overall, however, digital transformation has been more significant in promoting green innovation in non-state enterprises. These results are consistent with the existing literature [69]. A possible reason for this is that SOEs themselves have more social responsibility and should be more responsible for environmental protection through green technology innovation under the “dual carbon” goal. The potential impact of digital transformation on green innovation in state-owned enterprises may be comparatively smaller when compared to non-state-owned enterprises. Specifically, the responsiveness of state-owned enterprises to digital transformation about green innovation is lower than that observed in non-state-owned enterprises (See

Table 10. Heterogeneity results from property rights.

Variable	Non-SOEs			SOEs
	(1) Gpf	(2) Gps	(3) Gp	(4) Gpf	(5) Gps	(6) Gp
Digi	0.061 ***	0.091 ***	0.108 ***	0.049 **	0.064 ***	0.066 **
	(0.017)	(0.022)	(0.024)	(0.023)	(0.021)	(0.027)
Constant	−0.127	0.261	0.122	1.342 *	0.203	1.019
	(0.518)	(0.628)	(0.692)	(0.691)	(0.752)	(0.935)
Obs	572	572	572	759	759	759
R-squared	0.151	0.140	0.161	0.059	0.086	0.078
N	52	52	52	69	69	69
Year fix	YES	YES	YES	YES	YES	YES
Id fix	YES	YES	YES	YES	YES	YES
Controls	YES	YES	YES	YES	YES	YES

The symbols *, **, and *** are used to denote significance levels of 10%, 5%, and 1%, respectively.

).

4.5.2. Environmental Regulation

Due to their profit-oriented nature, enterprises usually ignore the protection of the ecological environment while pursuing maximum profits. Environmental regulation initiated by governments with incentives or mandatory means is an important way to curb ecological degradation and promote enterprises to realize green transformation for sustainable development. It can reasonably be predicted that the extent to which enterprises’ digital transformation drives green transformation will differ based on regional variations in the stringency of environmental regulations. This study is based on previous scientific research [23], the comprehensive index method was used to evaluate the environmental regulatory intensity, and the environmental regulatory intensity comprehensive index (Enr) was determined by industrial wastewater emissions per unit of output value, industrial SO₂ emissions per unit of output value, and industrial soot emissions per unit of output value. The computation process: to begin, it is necessary to standardize each indication. Subsequently, the entropy approach was employed to derive the weights of the elements, and subsequently, the environmental regulation composite index was computed using the obtained weights and standardized values. A greater Enr score indicates a higher level of pollutant emissions and a lower intensity of environmental management.

Additionally, we categorize the sample of businesses into two groups based on the level of environmental regulation: a high-environmental regulation group and a low-environmental regulation group. The dividing line is determined by the median value of each province in each year. This division allows us to examine the variations in the influence of digital transformation on promoting green transformation. In

Table 11. Results of Environmental Regulation Heterogeneity.

Variable	Low Environmental Regulation			High Environmental Regulation
	(1) Gpf	(2) Gps	(3) Gp	(4) Gpf	(5) Gps	(6) Gp
Digi	0.019	0.049	0.043	0.067 ***	0.090 ***	0.104 ***
	(0.014)	(0.041)	(0.037)	(0.017)	(0.018)	(0.022)
Constant	1.073	0.535	0.633	0.305	−0.493	−0.165
	(0.728)	(1.797)	(1.808)	(0.528)	(0.532)	(0.681)
Obs	252	252	252	1079	1079	1079
R-squared	0.132	0.076	0.093	0.083	0.093	0.096
N	32	32	32	108	108	108
Year fix	YES	YES	YES	YES	YES	YES
Id fix	YES	YES	YES	YES	YES	YES
Controls	YES	YES	YES	YES	YES	YES

The symbols *** is used to denote significance levels of 1%, respectively.

, a noteworthy positive correlation persists in the samples where there are elevated levels of environmental regulation in columns 4 to 6, but it fails to reach significant levels in the samples where there are low levels of environmental regulation. Through analysis, it becomes evident that the influence of digital transformation on environmental activities inside organizations exhibits discernible variations across different cohorts, with a more pronounced impact identified in regions characterized by more stringent environmental regulations.

5. Discussion

The globalization of the digital economy has led to significant progress in resource allocation and productivity in many countries. This change has provided a strong impetus for transitioning from traditional development models to a low-carbon, green approach. Digital transformation has emerged as a new economic dividend, promoting sustainable value creation for Chinese businesses. Modern information and communication technologies, such as BIM, the Internet of Things, and artificial intelligence, play a vital role in the economy, environment, and society [9]. For construction enterprises, digital transformation removes barriers to information for construction enterprises, allowing them to quickly identify green development opportunities and foster green creativity. Green innovation in construction enterprises covers energy consumption, green construction, and environmental pollution prevention. On the one hand, digital technology can be applied to the process of green innovation in construction enterprises. Digital-based operation and management can carry out real-time, accurate, and full-area collection of energy consumption generated during the construction and building process, establish a closed-loop channel for energy information collection and feedback, realize the refined management of energy consumption of the energy system, and change the high-energy, high-emission, and low-efficiency mode of production. Secondly, digital technology can play the advantage of synchronous sharing of data and information, which is conducive to providing specific digital support for construction enterprises to effectively integrate technology, resources, and team management related to green innovation, and catalyzing continuous innovation in the whole-cycle process of green construction. On the other hand, digital organizations and environments can promote the sustainable development of green innovation in construction enterprises. Digital transformation can break down information barriers by changing the digital culture and bringing in digital talent, encouraging construction enterprises to enhance communication and draw on resources, promoting the sharing of results, and promoting green innovation in enterprises. Concurrently, the digital environment can facilitate enterprises to work together to overcome problems encountered during digital transformation, such as construction methods and quality assurance, thus creating an enabling industrial ecosystem and activating green innovation potential. This finding is consistent with the results of Ye, Ouyang [70] and Liu, Wang [71], and extends their studies. Their sample in question focuses on manufacturing and heavily polluting enterprises. However, our sample extends to construction enterprises that are not included.

The study demonstrates that enterprise governance capacity and level of risk-taking are mediating pathways through which digital transformation affects green innovation, revealing the intrinsic mechanisms through which the former influences the latter. First, digital transformation improves enterprise governance capacity, which drives green innovation. Green innovation requires enterprises to consider resource consumption, environmental impacts, production costs, and economic benefits comprehensively. This places higher demands on their governance capacity. Digital technology provides efficient enterprise management infrastructure and platforms, which can effectively improve their governance capacity [72]. Good governance capacity helps reduce firms’ transaction costs, and discourages opportunistic behavior, thereby increasing the willingness and quality of green innovation. In addition, effective corporate governance helps companies identify changes in complex environments, proactively take strategic measures to mitigate potential risks, promote the smooth progress of business model innovation, and ultimately improve the quality of green innovation. This finding coincides with the findings of Gao, Feng [73]. Second, digital transformation enhances the level of risk-taking in enterprises, which in turn promotes the development of green innovation. The use of digital technology has not only changed how businesses compete and position themselves strategically, but it has also created opportunities to improve organizational structures and governance systems. This allows businesses to more effectively identify and prevent risks during their operations, ultimately increasing their ability to withstand risks [74]. Therefore, during the process of digital transformation for improving green innovation performance, the high level of risk-taking by enterprises can increase their willingness and confidence to carry out green innovation activities. This paper reaffirms this view based on data from listed companies in China’s construction industry.

We also find that private enterprises and enterprises with high environmental regulation have a greater impact on green innovation through the use of digital technology than SOEs and enterprises with low environmental regulation. On the one hand, privately owned construction enterprises typically have greater flexibility in their organizational structures and decision-making processes and can respond more quickly to market changes and technological innovations [75]. This flexibility has enabled private construction enterprises to rapidly adopt and apply digital technologies, leading to breakthroughs in green innovation. On the other hand, as environmental regulations tighten, enterprises need to manage their resources more carefully to reduce waste and pollution [76]. The application of digital tools such as BIM, cloud computing, and IoT can provide real-time monitoring and streamlined decision-making to support the implementation of green innovations.

6. Conclusions and Policy Implications

6.1. Conclusions

In the context of the “dual-carbon” strategy, it is now a necessary trend for the construction industry to pursue green innovation activities. This study focuses on Chinese A-share listed businesses in the construction sector between 2011 and 2021 as the research sample. It aims to provide a complete analysis of the impact of digital transformation on green innovation within construction enterprises. Additionally, it investigates the mediating mechanism and the heterogeneity connected to this relationship. The primary conclusions are as follows:(1) digital transformation of building enterprises can significantly enhance green innovation, and the conclusions are valid after the robustness test as well as the endogeneity test. (2) The mediating roles of enterprise governance capacity and enterprise risk-taking level are apparent between the digital transformation and green innovation of construction enterprises. Digital transformation of construction enterprises can accelerate green innovation by enhancing enterprise governance capacity and improving enterprise risk-taking levels. Among these channels, the transmission effect of enterprise governance capacity is more prominent. (3) Differences in property rights and environmental regulations can have an asymmetrical effect on the construction industry’s digital transformation-driven green innovation process. SOEs are subject to more regulatory constraints and environmental responsibilities in comparison to non-SOEs. As a result, SOEs possess a stronger internal motivation for implementing green transformation, which somewhat diminishes the impact of digital transformation. Meanwhile, with the context of highly rigorous regional environmental rules, the digital transformation of construction enterprises significantly amplifies their ability to green innovation.

6.2. Theoretical Contributions

First, this paper confirms that digital transformation has an important impact on green innovation in construction enterprises. Previous studies have mostly focused on the impact of digital transformation on green innovation at the regional and industry-wide levels, but have not delved into construction enterprises to investigate whether digital transformation can have an impact on green innovation. This study fills this research gap, responds to the call for “expanding the research dimension of digital transformation on green innovation”, and enriches the research on digital transformation and the factors affecting green innovation.

Second, the research on the impact of digital transformation on green innovation in construction enterprises has been deepened based on the mechanism perspective. Existing studies mostly theoretically deduce green innovation influencing factors and implementation paths from a static perspective, neglecting the process mechanism of how to effectively utilize digital transformation to achieve green innovation. The mediating role of enterprise governance capacity and risk-taking in digital transformation and green innovation of construction enterprises, as well as the related heterogeneous impacts, further opens the “black box” of the mechanism of digital capacity on green innovation of manufacturing enterprises, which can help to make up for the insufficiency of the existing literature on the process mechanism of digital transformation and green innovation, and provide new ideas for subsequent studies. This study provides new ideas for subsequent research.

6.3. Policy Implications

This study provides strong evidence that digital transformation incentivizes green innovation in construction enterprises and suggests the following policy implications based on the previous empirical analysis:

(1) The study will evaluate and assess the new position of green transformation and development within the framework of “dual-carbon”, highlighting the crucial role of digital transformation in promoting green innovation among construction enterprises and providing guidance to accelerate the pace of digital transformation. Construction companies must enhance the thorough amalgamation of the emerging technologies of BIM and the Internet of Things with green construction and organizational management. This will spur green technological innovation and accelerate the green, low-carbon transformation and upgrading of the construction industry. Government departments should provide greater support for the digital transformation of construction enterprises and expedite the creation of sound policies to bolster the advancement of enterprise digitization. The government should thoroughly promote the integration and application of intelligent construction equipment, core software, and the Internet in the construction industry so that digital technology can play the biggest role in creating green value for enterprises. Furthermore, actively promote collaborative research and development of technology, advance the construction of digital platforms, leverage the role of leading enterprises in the construction industry to provide demonstration and leadership, and establish an open industrial structure for the construction industry.

(2) Utilize the transmission effects of enterprise governance capacity, and risk-taking level to establish the dynamic transmission mechanism of “digital transformation—enterprise governance capacity/risk-taking level—green innovation”. Supported by digital technology, aims to promote the optimization of the synergy and efficiency of the governance of various market entities within the chain group, realize the transformation from the simple addition of various entities, data, resources, platforms, networks, systems, technologies and functions to the deep fusion, intelligent fusion, intelligent governance, and intelligent services, and improve the governance capacity and effectiveness of the governance of construction enterprises. Relying on digital development, it promotes the transformation of construction enterprises’ internal management model, expands the width and depth of enterprise management’s access to information, rationally allocates green innovation resources within the enterprise, and creates a favorable environment for improving the enterprise’s risk-taking level. Ultimately, digital development is employed to construct intelligent building systems and proficient management systems that promote green innovation in a holistic, coordinated, and sustainable manner.

(3) Following the principle of differentiation, specific measures targeted towards promoting digital transformation in different enterprises are taken, and enterprises are guided to choose a green development model according to their endowments. Firstly, support state-owned companies in carrying out digital green innovation activities by optimizing market competition, providing environmental protection subsidies, and purchasing services from the enterprises, formulate a green innovation evaluation index system for state-owned companies, and strengthen the evaluation of the allocation of green research and development funds for state-owned companies, to improve the efficiency and quality of green innovation of state-owned enterprises. Secondly, private enterprises’ heterogeneity must be fully considered to reduce the “one-size-fits-all” approach to generalized policies and increase guidance and support for them. We can accomplish it by establishing a financing guarantee system for private construction enterprises’ digital transformation, protecting property rights and knowledge, and providing interest-free loans, ultimately fostering green innovation in private enterprises. Thirdly, there is a need to enhance the judicious application of environmental regulatory tools while fully utilizing the respective merits of command-and-control and market-incentive tools. Additionally, it is crucial to reinforce the support and monitoring of green construction practices and actively explore alternative regulatory approaches to foster green behavior among construction enterprises.

7. Limitations and Directions for Future Research

This study requires further attention regarding its limitations for future research. Firstly, while the annual reports of the listed construction enterprises have gradually included information on digital transformation at the level of the enterprise, this may not adequately reflect the specific characteristics of Chinese construction enterprises as a whole due to the relatively small number of listed enterprises. Therefore, future research should aim to gather more information on digital transformation at the enterprise level and conduct more in-depth analyses at the enterprise level based on improved data availability. Second, this paper’s sample is limited to listed companies in the construction industry. There is no research focused on listed companies in various professional categories or non-listed companies. Future research should comparatively analyze the relationship between digital transformation and green innovation in listed companies across different professions to explore differences in this relationship. Third, the dynamic mechanism of how enterprises’ digital transformation affects green innovation requires continuous monitoring in the future. The interplay between enterprise governance capacity and the level of enterprise risk-taking in the context of digital transformation may have an impact. Further exploration of the mechanism of action can be carried out in the future with a configuration perspective.