The Impact of Digital–Real Economy Integration on ESG Performance: The Moderating Role of Organizational Inertia

Abstract

Against the backdrop of Chinese modernization and high-quality development, how to enhance corporate sustainability through the integration of digital and physical technologies has become a central concern for both academia and practice. Based on panel data from Chinese A-share listed companies spanning 2009 to 2023, comprising 11,034 firm-year observations, this study systematically examines the impact of digital–real technology integration on corporate ESG performance. It further introduces organizational inertia as a moderating variable in this relationship. Anchored in dynamic capability theory, an analytical framework is developed and empirically tested. The results demonstrate that digital–physical convergence significantly improves ESG performance, while organizational inertia negatively moderates this positive effect. Heterogeneity analyses reveal that the promoting effect is more pronounced in non-state-owned enterprises, firms in central and western regions, and non-high-pollution industries, but less evident in state-owned enterprises, firms in eastern regions, and high-pollution industries. Further tests on ESG sub-dimensions show significant improvements in environmental (E) and governance (G) aspects, but no evident impact on social (S) responsibility. This study uncovers the mechanisms and contextual boundaries through which digital–physical convergence fosters corporate sustainability, providing empirical evidence and policy implications for the coordinated advancement of digital transformation and green governance.

1. Introduction

Against the backdrop of Chinese modernization and high-quality development, the economy and society are simultaneously confronted with the dual challenge of digital upgrading and green transformation. The Report of the 20th National Congress of the Communist Party of China explicitly calls for accelerating the development of the digital economy, promoting the deep integration of digital technologies with the real economy, and advancing green and low-carbon development to build a modern society in harmony with nature. This strategic direction indicates that the advancement of digital technologies in enterprises must be guided by the principle of sustainable, low-carbon development, while high-quality growth increasingly relies on digitalization and intelligent transformation as key enablers. The two dimensions are thus interwoven, becoming critical pillars in fostering new productive forces. At the firm level, these macro-level imperatives are often embodied in environmental, social, and governance (ESG) performance, which has gradually supplanted financial indicators as the core standard for assessing corporate sustainability [1]. This means that firms must not only accelerate along the digitalization pathway but also demonstrate greater responsibility in green transition and social contribution. Amid the global trend of corporate sustainable development and digital transformation, a universal contradiction persists: despite high expectations for digital technology, its application often falls into the traps of “technology islands” and “instrumentalization”—meaning a disconnection from core business processes and strategic goals. This disconnect prevents digital technologies from being systematically translated into substantive improvements in Environmental, Social, and Governance (ESG) performance. The resulting ESG transformation dilemma, stemming from this technology-business decoupling, forms the core theoretical question this research aims to explore. Against this backdrop, China’s national strategies promoting the “deep integration of the digital and real economies” and “green and low-carbon development” provide a large-scale, policy-driven, typical context for addressing this universal problem. Therefore, utilizing this context and adopting a micro-level perspective of technology penetration, this paper systematically investigates whether corporate digital-real technology integration can serve as an effective pathway to overcome the aforementioned dilemma and enhance ESG performance.

The academic literature on ESG performance has produced rich findings, generally focusing on three streams. The first emphasizes external institutional and policy pressures, such as government regulation, capital market scrutiny, and media oversight, as key drivers of ESG engagement [2]. The second highlights internal governance and resource bases, including managerial characteristics, corporate culture, and innovative capacity, in shaping ESG practices [3]. The third examines technological enablers, particularly how digital transformation enhances transparency, stakeholder participation, and resource allocation efficiency, thereby improving corporate sustainability [4]. However, existing research tends to stress “digitalization” per se, while overlooking the convergence of digital technologies with real economic activities. In practice, firms often invest heavily in digital platforms and systems without embedding them into production, governance, and strategic processes, thereby creating a “value vacuum.” Only when digital technologies are firmly anchored in core business and production processes can digital–real technology convergence realize its enabling potential, driving systemic improvements in ESG performance and laying a solid foundation for cultivating green productive forces.

Despite its strategic importance, the mechanisms through which digital–real convergence influences ESG performance remain underexplored. In the Chinese context, institutional pressures, capital market preferences, and social oversight jointly complicate firms’ pathways toward sustainability, creating uncertainty regarding whether convergence can effectively translate into ESG improvement. Moreover, prior research has not sufficiently addressed the role of internal organizational characteristics. This study introduces “organizational inertia” as a moderating factor, grounded in real-world conditions. Over time, firms develop structural, strategic, and procedural rigidities that hinder timely resource reconfiguration and strategic adjustment when faced with new technologies and external demands [5,6].

Based on these considerations, this paper addresses three research questions: (1) Does digital–real technology convergence significantly improve corporate ESG performance? (2) How does organizational inertia influence this relationship? (3) Do these effects vary across different types of enterprises? To answer these questions, we draw on dynamic capability theory as the primary analytical lens, complemented by path dependence theory and stakeholder theory, to develop a conceptual framework linking digital–real convergence, ESG performance, and organizational inertia. The framework is then empirically tested using panel data from Chinese listed firms.

This study contributes in three main ways. First, it addresses a gap in the literature on the intersection of digitalization and sustainability by incorporating digital–real convergence into the ESG research framework, thereby extending the dialogue between the digital economy and corporate sustainability. Second, it uncovers the moderating role of organizational inertia in shaping the convergence–ESG nexus, enriching the understanding of how dynamic capabilities and path dependence interact in corporate transformation contexts. Third, by examining heterogeneity across ownership types, industries, and regions, the study delineates the contextual boundaries of convergence-driven ESG enhancement, offering targeted implications for policymakers and corporate managers, and providing new empirical evidence for the cultivation of green productive forces and the construction of low-carbon economic systems.

Subsequent research will primarily focus on Section 2 (Literature Review and Hypotheses Development), Section 3 (Research Design), Section 4 (Empirical Tests and Results), and Section 5 (Conclusions and Implications).

2. Literature Review and Hypotheses Development

2.1. Digital–Real Technology Convergence

Digital–real technology convergence refers to the deep integration of digital technologies—such as big data, cloud computing, artificial intelligence, and the Internet of Things—with physical business processes, product systems, and industrial chain activities, thereby enabling the efficient coordination of information, capital, and material flows [7,8]. Unlike traditional informatization upgrades or automation reforms, digital–real convergence at the firm level emphasizes the embedding and application of digital technologies in core functions such as production, R&D, management, and services. As the micro-level carrier of the “digital–real integration” process, firms serve as the key arena where this transformation unfolds. Essentially, digital–real convergence represents the complementarity and recombination of digital and physical industrial technologies within enterprises: on the one hand, it expands knowledge boundaries and improves innovation quality; on the other, it enhances firms’ absorptive capacity for the diffusion of integrated technologies, thereby strengthening competitive advantages [8,9]. In empirical studies, firm-level digital–real convergence is commonly measured through two main approaches: secondary data methods, such as constructing convergence indices based on patent data and R&D investment [9], and survey-based methods that capture managers’ perceptions of technological, process, and organizational integration [10]. It should be noted that the patent-based measurement approach primarily captures integration at the technological knowledge layer. While offering objective and comparable advantages, it may not fully encompass the depth of integration in non-technical dimensions such as business processes and organizational structures. This represents a boundary condition of this methodology.

In practice, digital–real convergence has been shown to generate substantial economic and managerial benefits. By improving production efficiency and optimizing resource allocation, it reduces operating costs and enhances firms’ resilience to market volatility [7,10]. The deep coupling of digital and physical business processes enables greater predictability and precision in supply chain management, quality control, and customer services, while also accelerating new product development and market responsiveness [7,11]. From a sustainability perspective, the transparency and traceability afforded by digital–real convergence help improve energy efficiency, reduce resource waste, and promote the adoption of green production models [12]. Although these benefits provide both technological and managerial foundations for ESG engagement, the mechanisms through which convergence systematically enhances ESG performance remain underexplored—particularly with regard to heterogeneity across organizational contexts.

Existing research further highlights that the outcomes of digital–real convergence are jointly shaped by external and internal conditions. Externally, factors such as the maturity of digital infrastructure, regulatory and policy environments, and the digitalization of upstream and downstream partners affect implementation costs and coordination efficiency. Internally, firms’ resource endowments, digital capabilities, organizational culture, and management systems play a more fundamental role in determining success [13]. Among these, organizational responsiveness to change is critical to unlocking the potential of digital–real convergence. In technology-driven strategic transformation, organizational inertia often emerges as a major barrier: rigidities in structures, processes, and resource allocation [14] can delay decision-making, hinder business reconfiguration, and constrain governance innovation [6]. Thus, examining organizational inertia as an internal contingency in the convergence–ESG relationship not only advances scholarly understanding of contextualized effects but also offers practical insights into how firms can prevent sustainability gains from being undermined during technological integration.

2.2. ESG Performance

With the deepening of digital–real economy integration, its potential impact on corporate environmental, social, and governance (ESG) performance has attracted growing attention from both academia and practice. ESG, as a comprehensive framework for assessing corporate sustainability, emphasizes the coordinated advancement of environmental responsibility, social value creation, and corporate governance. Unlike traditional performance evaluations centered on financial indicators, ESG incorporates carbon reduction, resource efficiency, employee welfare, social responsibility, and governance transparency into a unified system, reflecting a long-term balance across economic, social, and environmental dimensions [15,16]. In terms of measurement, most studies rely on third-party rating systems such as the China Securities Index (CSI) ESG rating or the Wind ESG disclosure rating, while others construct text-based indicators from annual reports and corporate social responsibility disclosures [17,18]. However, differences in dimension weighting, data sources, and cross-industry comparability often pose challenges to the consistency and generalizability of findings.

The drivers of corporate ESG performance are complex and multifaceted. Externally, institutional environments and policy orientations are widely recognized as critical forces: stringent environmental regulations, disclosure requirements, and fiscal incentives can significantly motivate firms to improve their ESG performance [17,19,20]. Capital market attention also exerts both disciplinary and incentive effects, as investor preferences, lending policies, and sustainability demands from supply chain partners influence corporate decision-making through financing costs and market access [2]. Furthermore, media oversight and public opinion generate reputational pressure, encouraging firms to strengthen their social responsibility and governance transparency [21]. Internally, ESG engagement is shaped by firms’ strategic orientation: managerial values and strategic cognition directly affect the prioritization of ESG investment [22], while resource endowments and technological capabilities determine the effectiveness of environmental management, social contribution, and governance innovation. Corporate culture, organizational structure, and incentive mechanisms further influence the persistence and depth of ESG practices [23,24].

It is important to note, however, that these drivers do not operate uniformly across all contexts. The alignment between organizational characteristics and external conditions may alter both the pathways and outcomes of ESG improvement. Factors such as industry competition, market volatility, and corporate life-cycle stage shape the efficiency with which external pressures are translated into internal actions [25]. Similarly, organizational flexibility in responding to change and reallocating resources may determine whether technological applications effectively support sustainability objectives [26]. When firms are constrained by strong organizational inertia, their capacity for timely strategic adjustment and resource reconfiguration is limited, often resulting in delayed responses to external sustainability pressures and, ultimately, weaker ESG improvements.

2.3. Organizational Inertia

Organizational inertia refers to the tendency of organizations to maintain stability in their structures, strategies, processes, and cultures, while responding slowly to changes in the external environment [5]. This phenomenon stems from path dependence and institutionalized arrangements developed during long-term operations, leading firms to rely on established patterns rather than undertaking major adjustments under environmental uncertainty [27]. Scholars typically distinguish between two primary types of inertia: structural inertia, which emphasizes rigidity in formal institutions, power allocation, and resource distribution; and strategic inertia, which highlights firms’ persistence in strategic choices, innovation trajectories, and market positioning [6]. More recent studies have extended these categories to include technological, cultural, and cognitive inertia, capturing the multiple constraints that hinder firms from adapting effectively to external changes [28]. In terms of measurement, early studies often relied on indicators reflecting organizational size and structural stability—such as changes in employee numbers, asset scales, and firm age—to capture inertia levels [27,29]. This approach, based on financial and operational data, remains one of the most widely used in academia due to its simplicity and practicality, as it directly reflects organizational stability in resource allocation and structural arrangements.

Mechanistically, organizational inertia is often viewed as a critical contingency factor influencing strategic adjustment and firm performance. As a moderator, inertia may weaken the positive effects of external opportunities or internal capabilities on performance outcomes [30,31]. Within the sustainability domain, high levels of structural and procedural stability can slow the adoption of green technologies, thereby undermining the environmental performance gains expected from policy incentives and market demand [32,33]. In the context of organizational innovation, inertia can constrain flexibility in new product development or business model transformation, limiting the extent to which technological resources and innovation capabilities are translated into market advantages [34]. Conversely, moderate inertia may help firms maintain strategic consistency and integration efficiency, preventing the excessive costs and strategic drift associated with frequent adjustments. This “double-edged sword” nature suggests that the interaction between inertia and organizational capabilities must be analyzed in context [35,36,37].

2.4. Hypotheses Development

2.4.1. Digital–Real Technology Integration and ESG Performance

Dynamic capability theory suggests that firms can sustain competitive advantages in uncertain environments by sensing, seizing, and transforming opportunities and resources [38]. Digital–real technology integration, which embeds digital technologies into core physical business processes, is an important way for firms to strengthen such capabilities [9]. On the one hand, real-time data collection and intelligent analytics allow firms to identify risks and opportunities in environmental and social areas earlier, helping them respond proactively to policy changes, market demands, and resource constraints [39,40]. On the other hand, tools such as digital twins, algorithm optimization, and smart scheduling improve energy and resource efficiency while embedding low-carbon and social responsibility principles into operations, product design, and supply chain management [41,42]. Through such data-driven reconfigurations, firms also achieve greater transparency and inclusiveness in governance, making responsibility fulfillment, performance monitoring, and disclosure more sustainable. In this way, digital–real technology integration strengthens firms’ responses across environmental, social, and governance dimensions, and builds an internal driver for continuous ESG improvement.

Yet internal capabilities alone are not enough. For ESG improvements to be recognized, they must be validated through external trust, perception, and resource allocation [43]. Put differently, digital–real technology integration gives firms the “tools” to transform, but whether these tools generate institutional and market rewards depends on stakeholders. From a stakeholder theory perspective, firms must build legitimacy and co-create value with governments, investors, supply chain partners, and the public [44]. Integration supports this by improving transparency through data visualization, credible traceability systems, and digital platforms for multi-party participation [39]. These tools lower the cost for stakeholders to access and verify ESG information, while increasing trust in corporate responsibility.

Therefore, dynamic capabilities provide an internal “toolkit” for ESG improvement, while stakeholder engagement serves as an external “transmitter” for value realization. Digital-real technology integration collectively drives ESG performance enhancement by simultaneously strengthening these internal and external mechanisms.

Based on this reasoning, we propose the following hypothesis:

H1. 

Digital–real technology integration improves firms’ ESG performance.

2.4.2. The Moderating Role of Organizational Inertia in the Relationship Between Digital–Real Technology Integration and ESG Performance

Through digital–real technology integration, firms enhance their dynamic capabilities, which in turn support systematic improvements across environmental, social, and governance (ESG) dimensions. However, the release of dynamic capabilities does not occur uniformly in all organizational contexts. Internal constraints play a critical role in shaping this process. Among them, organizational inertia—defined as the stability and rigidity of structures, processes, and strategies—is considered a key contextual factor that limits the transformation of dynamic capabilities into actual performance outcomes [5,45].

At the structural level, inertia is reflected in firms’ reliance on existing resource allocations and process arrangements. This reliance makes organizations continue to follow established routines even when digital–real integration offers greater transparency and efficiency [27]. Such structural inertia slows down the extent and speed at which dynamic capabilities can enhance environmental and social governance. At the strategic level, inertia manifests in path dependency. Over time, firms develop strategic cognitions and action patterns that are difficult to change, even when new opportunities emerge from technology integration [45,46]. Consequently, firms may pursue incremental adjustments instead of rapid responses to sustainability pressures, thereby diluting the benefits of digital–real technology integration [47]. At the governance level, inertia further exacerbates the problem. Rigid institutional arrangements and disclosure mechanisms prevent firms from fully leveraging the transparency gains enabled by digital platforms. While in theory such platforms could strengthen stakeholder engagement, inflexible governance structures confine disclosure practices to traditional modes, limiting external oversight and social recognition [48].

In sum, when organizational inertia is high, firms fail to promptly reconfigure resources and processes in response to digital–real integration. This weakens the potential of dynamic capabilities and reduces the positive effect of digital–real technology integration on ESG performance. Accordingly, we propose the following hypothesis:

H2. 

Organizational inertia negatively moderates the relationship between digital–real technology integration and ESG performance, such that the positive impact of integration is weaker under higher levels of organizational inertia.

3. Research Design

3.1. Sample and Data Collection

This study selects the period 2009–2023 as the research window. On the one hand, this period corresponds to the critical stage when China’s green transition and digitalization strategies advanced in parallel. On the other hand, it aligns with the gradual establishment of ESG rating systems in China, allowing us to better capture the dynamic impact of digital–real technology integration on corporate sustainable development. The sample data are mainly drawn from three sources. First, the patent database, from which we extract information on firms’ invention patents, including applications and disclosures filed with the China National Intellectual Property Administration (CNIPA), IPC, and patent citations. Second, the CSMAR database, which provides basic information and financial indicators of A-share listed firms. Third, the Wind database, which contains ESG ratings issued by China Securities Index Co., Ltd. After constructing the relevant variables, we matched the three databases and applied the following screening procedures: (1) excluded firms with missing key variables; (2) excluded firms in the financial industry; (3) excluded firms under “ST” or special treatment status; (4) excluded firms with only single observations during the study period; and (5) winsorized continuous variables at the 1% level in both tails. The final dataset consists of 11,034 firm-year observations for the period 2009–2023. The main variables are presented in

Table 1. Overview of major variables.

VarName	Variable Type	Degree Basis
ESG	Dependent variable	Huazheng Index Information Service Co., Ltd. (Huazheng)
TechCon	Independent variable	Measured by the flow of digital technologies into physical industry innovations, using patent citation data.
OI	Moderating Variable	A composite index of firm size, age, and total assets (standardized).
Size	Control variable	total assets in the current year
Lev	Control variable	Ratio of total liabilities to total assets.
ROE	Control variable	Return on equity.
Indep	Control variable	Proportion of independent directors.
Board	Control variable	Natural logarithm of board size.
Growth	Control variable	Current year revenue/prior year revenue-1
ListAge	Control variable	Ln(Years since IPO+1)
TobinQ	Control variable	Tobin’s Q
BM	Control variable	Book value/market value

.

3.2. Variables and Measurements

3.2.1. Explained Variable: ESG

The dependent variable in this study is corporate ESG performance, measured using ESG ratings provided by Huazheng Index Information Service Co., Ltd. (Huazheng), a China-based index provider. As one of the most authoritative third-party rating agencies in China, the Huazheng ESG evaluation system draws on international mainstream frameworks while incorporating localized adjustments to reflect the disclosure practices of Chinese listed firms. Compared with other rating systems such as Bloomberg and Commerce Green, the Huazheng system offers broader coverage, higher update frequency, and longer historical traceability, and has been widely applied in both academic research and investment analysis in China [49,50].

The rating system adopts a multi-layered structure, comprising three primary dimensions—environment (E), social (S), and governance (G)—which are further divided into 16 secondary indicators, 44 tertiary indicators, and over 300 basic indicators. These indicators are constructed through advanced techniques such as natural language processing (NLP) and semantic analysis, extracting and quantifying information from firms’ public disclosures. After standardization and weighted aggregation, a comprehensive ESG score is generated. The final scores are classified into nine rating levels: C, CC, CCC, B, BB, BBB, A, AA, and AAA. For empirical analysis, these categories are converted into numerical values ranging from 1 (C) to 9 (AAA), where higher values indicate better ESG performance [49].

Since Huazheng ESG ratings are released on a quarterly basis, this study uses the annual average of the four quarterly ratings as the measure of a firm’s ESG performance, ensuring both timeliness and comparability across firms.

3.2.2. Explanatory Variable: Digital–Real Technology Integration (TechCon)

Following Huang [9], this study measures firms’ digital–real technology integration by capturing the flow of digital technologies into physical industry innovations through patent citation information. Specifically, based on the Classification of Core Digital Economy Industries and Corresponding International Patent Classification (IPC) Codes (2023) issued by the China National Intellectual Property Administration, patents are first identified as digital technology patents. If a patent’s IPC primary classification does not belong to the digital industry, but its cited patents include digital technologies, this patent is classified as an instance of digital–real technology integration. At the firm–year level, the number of such patents is aggregated, and the measure of digital–real technology integration (TechCon) is constructed as the natural logarithm of (1 + the aggregated number).

3.2.3. Moderating Variable: Organizational Inertia (OI)

The measurement of organizational inertia has been approached in two main ways. Some studies employ survey-based scales, while others adopt proxy variables due to the limited maturity of organizational inertia research and concerns about scale validity. Kelly and Amburgey [27], for example, used firm size and firm age as proxies. Building on this approach, this study incorporates an additional dimension—firm assets—to better capture inertia. Specifically, organizational inertia is calculated by standardizing and summing three indicators: (1) firm size, measured as the natural logarithm of (1 + total number of employees); (2) firm age, measured as the natural logarithm of (1 + years since establishment); and (3) firm assets, measured as the natural logarithm of (1 + year-end total assets). A higher value of this composite index indicates a higher level of organizational inertia.

3.2.4. Control Variables

To account for other potential influences on ESG performance, this study includes a set of firm-level control variables (ContVars): firm size (Size, natural logarithm of total assets), leverage (Lev, ratio of total liabilities to total assets), return on equity (ROE, ratio of net profit to shareholders’ equity), proportion of independent directors (Indep, ratio of independent directors to board size), board size (Board, natural logarithm of total number of directors), firm growth (Growth), listing age (Listage), Tobin’s Q (TobinQ), and the book-to-market ratio (BM, book value divided by market value).

3.3. Research Model

To determine the appropriate panel model, this study first conducted a Hausman test. The test results significantly rejected the null hypothesis at the 1% level, indicating the superiority of the fixed effects model over the random effects model. Consequently, this study employs a two-way fixed effects panel model to test the hypotheses derived from the theoretical section, ensuring more consistent and efficient estimation results for the core variables.

$${\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{T}\mathrm{e}\mathrm{c}\mathrm{h}\mathrm{C}\mathrm{o}\mathrm{n}}_{\mathrm{i},\mathrm{t}}+\sum \mathrm{C}\mathrm{V}\mathrm{s}+\sum \mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+\sum \mathrm{I}\mathrm{n}\mathrm{d}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(1)

$${\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{T}\mathrm{e}\mathrm{c}\mathrm{h}\mathrm{C}\mathrm{o}\mathrm{n}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\alpha }}_2{\mathrm{O}\mathrm{I}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\alpha }}_3{\mathrm{T}\mathrm{e}\mathrm{c}\mathrm{h}\mathrm{C}\mathrm{o}\mathrm{n}}_{\mathrm{i},\mathrm{t}}\times {\mathrm{O}\mathrm{I}}_{\mathrm{i},\mathrm{t}}+\sum \mathrm{C}\mathrm{V}\mathrm{s}+\sum \mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+\sum \mathrm{I}\mathrm{n}\mathrm{d}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(2)

To examine the impact of digital–real technology integration on corporate ESG performance, this study first establishes the baseline regression model (1), where i and t denote firm and year, respectively; represents the ESG performance of firm i in year t; ${\mathrm{T}\mathrm{e}\mathrm{c}\mathrm{h}\mathrm{C}\mathrm{o}\mathrm{n}}_{\mathrm{i},\mathrm{t}}$ measures the degree of digital–real technology integration; $\mathrm{C}\mathrm{V}\mathrm{s}$ is a vector of control variables; $\mathrm{I}\mathrm{n}\mathrm{d}$ and Year capture industry and year fixed effects, $\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{l}\mathrm{y}; \mathrm{a}\mathrm{n}\mathrm{d}{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$ is the error term. To further test the moderating effect of organizational inertia, we construct model (2), where ${\mathrm{O}\mathrm{I}}_{\mathrm{i},\mathrm{t}}$ denotes the level of organizational inertia of firm i in year t, and the interaction term ${\mathrm{T}\mathrm{e}\mathrm{c}\mathrm{h}\mathrm{C}\mathrm{o}\mathrm{n}}_{\mathrm{i},\mathrm{t}}\times {\mathrm{O}\mathrm{I}}_{\mathrm{i},\mathrm{t}}$ captures its moderating effect. Other variables remain consistent with model (1).

4. Empirical Test and Results

4.1. Descriptive Statistics

Table 2. Descriptive statistics.

VarName	Obs	Mean	SD	Min	Max
ESG	11,034	4.229	0.918	1.000	7.250
TechCon	11,034	0.269	0.616	0.000	6.225
OI	11,034	0.106	1.921	−3.849	5.245
Size	11,034	22.173	1.178	20.113	25.633
Lev	11,034	0.408	0.189	0.060	0.869
ROE	11,034	0.071	0.115	−0.500	0.349
Indep	11,034	0.376	0.053	0.333	0.571
Board	11,034	2.232	0.170	1.792	2.708
Growth	11,034	0.164	0.299	−0.432	1.583
ListAge	11,034	2.046	0.746	0.000	3.296
TobinQ	11,034	2.088	1.218	0.864	7.732
BM	11,034	0.595	0.236	0.129	1.157

presents the descriptive statistics of the key variables. The average ESG performance of firms is 4.229, with a relatively large standard deviation and a wide gap between the maximum and minimum values. This indicates that most firms perform below the overall average, while a few leading firms significantly pull up the overall level of ESG performance, which is consistent with the current situation of Chinese listed companies. The mean value of digital–real technology integration (TechCon) is 0.269, also with considerable variation, suggesting substantial heterogeneity among listed firms in China. While some companies are at the forefront of digital transformation, others have not yet implemented relevant initiatives.

4.2. Baseline Regression

To empirically test the proposed hypotheses, we adopt a stepwise fixed-effects regression framework. The analysis first establishes the direct impact of digital–real technology integration (TechCon) on ESG performance through the baseline model and then introduces organizational inertia (OI) and its interaction term to examine the moderating effect.

The results of the baseline regressions are reported in columns (1)–(3) of

Table 3. Baseline Regression.

	(1)	(2)	(3)
	ESG	ESG	ESG
TechCon	0.101 ***	0.143 ***	0.045 ***
	(7.13)	(9.67)	(3.06)
Size			0.252 ***
			(22.17)
Lev			−0.957 ***
			(−16.83)
ROE			1.000 ***
			(11.73)
Indep			1.664 ***
			(8.78)
Board			0.265 ***
			(4.28)
Growth			−0.118 ***
			(−3.93)
ListAge			−0.136 ***
			(−9.74)
TobinQ			−0.032 **
			(−2.54)
BM			−0.163 **
			(−2.20)
_cons	4.202 ***	4.191 ***	−1.799 ***
	(441.70)	(443.06)	(−6.99)
Ind	NO	Yes	Yes
Year	NO	Yes	Yes
N	11,034	11,034	11,034
F	51 ***	94 ***	139 ***
r2_a	0.004	0.034	0.135

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

, with estimation results summarized in Table 3. Column (1) presents the regression of TechCon on ESG performance without additional controls, showing a significantly positive correlation at the 1% level. Column (2) controls for industry and year fixed effects, and the significance remains unchanged. Column (3) further includes firm-level control variables alongside industry and year fixed effects; the estimated coefficient of TechCon is 0.0448, still significant at the 1% level. These findings provide preliminary support for Hypothesis H1, indicating that digital–real technology integration significantly improves firms’ ESG performance.

4.3. Robustness Checks

To verify the robustness of the baseline results, we conduct several additional tests focusing on sample period adjustments, estimation methods, and alternative measures of the dependent variable. Results are reported in columns (1)–(4) of

Table 4. Robustness Test.

	(1)	(2)	(3)	(4)
	ESG	ESG	ESG	WindESG
TechCon	0.046 ***	0.045 ***	0.039 ***	0.044 **
	(3.19)	(3.19)	(2.69)	(1.96)
_cons	−1.528 ***	−1.799 ***	−2.435 ***	2.517 ***
	(−5.74)	(−6.80)	(−9.18)	(8.88)
Ind	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
Control	Yes	Yes	Yes	Yes
City	No	No	Yes	No
N	9889	11,034	11,034	5654
F	127 ***	136 ***	138 ***	33 ***
r2_a	0.137	0.135	0.210	0.125

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

.

First, to exclude potential distortions caused by special events, we remove observations from 2020, when the COVID-19 pandemic had a profound impact on firms’ operations, supply chains, and CSR activities, potentially introducing abnormal ESG fluctuations. The re-estimated results (Table 4, column 1) remain significantly positive at the 1% level, consistent with the baseline model.

Second, we adjust for potential estimation bias caused by individual heterogeneity or serial correlation by clustering robust standard errors. The results (Table 4, column 2) are consistent with the baseline regression, confirming the robustness of our findings.

Third, considering that regional differences in policy environments, economic development, and digital infrastructure may systematically affect ESG outcomes, we introduce city fixed effects into the regression. The results (Table 4, column 3) still show a significantly positive relationship at the 1% level, further validating the robustness of the baseline conclusion.

Finally, to address concerns about measurement dependence on a single ESG rating agency, we replace Huazheng’s ESG rating with Wind’s ESG rating as the dependent variable. The results remain significantly positive and robust, consistent with the baseline findings.

4.4. Endogeneity Test

A potential concern is endogeneity due to reverse causality between digital–real integration and ESG performance. To mitigate this issue, we use lagged values of the key explanatory variable as instruments. Specifically, we re-estimate the model using one-, two-, and three-period lags of TechCon. This approach is justified because past integration levels can strongly predict current integration, while they are unlikely to be directly affected by current ESG outcomes, satisfying both relevance and exogeneity conditions. Moreover, the cumulative and lagged nature of ESG improvements makes this approach theoretically consistent.

The results, reported in columns (1)–(3) of

Table 5. Endogeneity Test.

	(1)	(2)	(3)
	ESG	ESG	ESG
L. TechCon	0.051 ***
	(3.20)
L2. TechCon		0.056 ***
		(3.39)
L3. TechCon			0.055 ***
			(3.23)
_cons	−1.921 ***	−2.346 ***	−2.829 ***
	(−6.59)	(−7.48)	(−8.19)
Ind	Yes	Yes	Yes
Year	Yes	Yes	Yes
Control	Yes	Yes	Yes
N	8834	7530	6312
F	114 ***	114 ***	102 ***
r2_a	0.136	0.152	0.159

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

, show that regardless of whether a one-, two-, or three-period lag is used, the coefficients remain significantly positive at the 1% level. This not only confirms the robustness of our baseline findings but also highlights the persistent and lagged effects of digital–real integration on ESG performance, consistent with the gradual nature of corporate green transformation and governance improvement. These findings further support Hypothesis H1.

4.5. Moderating Effect

Building on the baseline regression, we further examine the moderating role of organizational inertia. Specifically, we include the interaction term between TechCon and OI in the model. As shown in

Table 6. Moderating Effect.

	(1)
	ESG
TechCon	0.060 ***
	(3.57)
OI	0.061 ***
	(6.80)
OI_TechCon_c	−0.012 **
	(−2.18)
_cons	−0.129
	(−0.36)
Ind	Yes
Year	Yes
Control	Yes
N	11,034
F	121 ***
r2_a	0.138

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

, the coefficient of TechCon is 0.0604 and remains significantly positive at the 1% level, indicating that digital–real technology integration has a beneficial impact on ESG performance. However, the interaction term between TechCon and OI is −0.0124 and significant at the 5% level, suggesting that organizational inertia negatively moderates this relationship.

This result confirms Hypothesis H2: when organizational inertia is high, the positive effect of digital–real integration on ESG performance is weakened. In other words, while integration enhances ESG performance by enabling data-driven decision-making, process reconfiguration, and governance improvements, high levels of organizational inertia hinder the full realization of these benefits.

4.6. Further Analyses

4.6.1. Heterogeneity by Firm Type

This study conducts heterogeneity analysis from three dimensions: ownership type, industry attribute, and regional environment, to systematically examine the contextual impact of digital-real technology integration on corporate ESG performance. The selection of these three dimensions stems from their collective role in shaping the key contexts that drive differentiated responses among Chinese enterprises to technological and sustainable development requirements. First, ownership type determines a firm’s core objectives and resource dependencies. The ESG performance of state-owned enterprises (SOEs) is often driven more by policy mandates and administrative assessments; they tend to have stronger resource endowment but may also exhibit greater organizational inertia. In contrast, non-SOEs, operating under market competition and financing constraints, rely more on technological innovation to enhance efficiency and legitimacy. Second, industry environmental sensitivity shapes differing external pressures and improvement potential. Heavily polluting industries face the strongest compliance pressure under the “dual carbon” goals, and their ESG behaviors often exhibit passive compliance characteristics. Conversely, non-heavily polluting industries are more likely to engage in ESG practices as an active strategic choice, where the marginal effect of technology integration may be more pronounced. Finally, regional development disparities create systematic differences in digital infrastructure, policy support, and market environments between central/western and eastern regions. These differences directly affect firms’ technology absorption capacity and motivation for sustainable development. Through these three layers of heterogeneity testing, this study comprehensively reveals the multi-level boundaries of the effect of digital-real integration on corporate ESG performance, providing a precise basis for targeted policy interventions.

The empirical results regarding ownership heterogeneity are presented in columns (1) and (2) of Table 6. As shown in column (1), the regression coefficient of digital-real technology integration on corporate ESG performance is insignificant in the SOE subsample. In contrast, column (2) shows a significantly positive coefficient in the non-SOE subsample. These results indicate that ownership type serves as an important contextual moderator in the relationship between digital-real technology integration and ESG performance. The differences in institutional attributes lead to significant divergence in the underlying mechanisms driving this relationship.

The empirical results for heterogeneity based on whether a firm belongs to a heavily polluting industry are shown in columns (3) and (4) of

Table 7. Heterogeneity by Firm Type.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
	ESG	ESG	ESG	ESG	ESG	ESG	ESG
TechCon	0.030	0.056 ***	0.025	0.046 ***	0.009	0.165 ***	0.089 ***
	(1.33)	(2.87)	(0.78)	(2.81)	(0.49)	(3.69)	(2.79)
_cons	−2.518 ***	−1.428 ***	−2.128 ***	−1.791 ***	−1.836 ***	−1.173 *	−2.567 ***
	(−6.00)	(−4.04)	(−3.86)	(−5.99)	(−5.61)	(−1.80)	(−4.40)
Ind	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Control	Yes	Yes	Yes	Yes	Yes	Yes	Yes
SOE	1	0
Pollute			1	0
Region					East	West	Mid
N	3351	7683	2729	8305	7449	1383	2201
F	49 ***	93 ***	28 ***	119 ***	100 ***	15 ***	28 ***
r2_a	0.157	0.134	0.110	0.155	0.142	0.122	0.144

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

. The result in column (3) indicates that the regression coefficient of digital-real technology integration on ESG performance is insignificant in the heavily polluting industry subsample. Conversely, column (4) shows a significantly positive relationship in the non-heavily polluting industry subsample. This finding deviates somewhat from conventional expectations, suggesting that digital-real integration has not been effectively translated into ESG improvements in heavily polluting industries, whereas it plays a more prominent facilitative role in non-heavily polluting industries.

The empirical results for regional heterogeneity are displayed in columns (5) to (7) of Table 7. The result in column (5) shows an insignificant impact of digital-real technology integration on ESG performance in the subsample of listed firms in the eastern region. In contrast, the results in columns (6) and (7) show a significantly positive relationship at the 1% level in the central and western region subsamples. However, the coefficient is larger for firms in the western region (0.165) than for those in the central region (0.089). This indicates that a one-unit increase in TechCon improves ESG performance by 0.165 units for western firms, but only by 0.089 units for central firms. This finding suggests that the institutional environment and external constraints faced by firms in different regions lead to differentiated manifestations of the integration mechanism.

4.6.2. ESG Sub-Dimensions

To further uncover the mechanisms through which digital–real technology integration influences corporate sustainability, we conduct separate tests on the three ESG pillars. Results reported in columns (1)–(3) of

Table 8. ESG Sub-Dimensions.

	(1)	(2)	(3)
	E	S	G
TechCon	0.080 ***	−0.016	0.067 ***
	(4.47)	(−0.69)	(3.84)
_cons	−3.725 ***	−1.948 ***	0.284
	(−11.86)	(−4.75)	(0.93)
Ind	Yes	Yes	Yes
Year	Yes	Yes	Yes
Control	Yes	Yes	Yes
N	11,034	11,034	11,034
F	124 ***	60 ***	215 ***
r2_a	0.141	0.219	0.250

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

show that integration has significantly positive effects on environmental (E) and governance (G) performance, both at the 1% level, whereas its impact on social responsibility (S) is not significant. This finding indicates that the contribution of digital–real integration to ESG performance is not uniform across dimensions but instead displays clear heterogeneity.

Specifically, digital–real integration enhances energy efficiency and emission management through real-time data collection, intelligent production scheduling, and optimized supply chain coordination, generating direct improvements in the environmental dimension. At the same time, the adoption of digital platforms and big data analytics strengthens transparency and traceability in governance processes, facilitating greater accountability and institutional upgrading. In other words, integration naturally supports improvements in green production capacity and governance transparency, explaining its consistently positive effects on E and G.

By contrast, improvements in the social dimension appear more limited. This is partly because social responsibility practices are often shaped by corporate values, cultural orientation, and resource allocation preferences, factors not entirely driven by technological tools. Moreover, social outcomes tend to be long-term and indirect and thus are less likely to be captured in the short term. These findings suggest that while digital–real integration provides solid support for environmental and governance performance, achieving significant progress in social responsibility requires complementary institutional arrangements and value-driven initiatives.

5. Conclusions and Implications

Based on data from Chinese A-share listed companies spanning 2009 to 2023, this study systematically examines the impact of firms’ digital-real technology integration on their ESG performance. It further introduces organizational inertia as a moderating variable to explore its mechanism within this relationship. A series of heterogeneity tests reveal differentiated effects depending on firm attributes and contextual characteristics.

The findings indicate that, at the aggregate level, digital-real technology integration significantly enhances corporate ESG performance. This suggests that the deep coupling of digital technologies with physical business operations can strengthen firms’ data-driven capabilities and resource reconfiguration abilities, leading to comprehensive improvements in environmental responsibility, social responsibility, and corporate governance. This conclusion aligns with the logic of dynamic capability theory and stakeholder theory. Furthermore, additional moderation effect analysis shows that organizational inertia exerts a significant negative influence on this relationship. When firms exhibit strong inertia, the flexibility of internal resource allocation and strategic adjustment decreases, hindering the effective realization of the positive effects of digital-real integration and thereby weakening its improvement effect on ESG performance.

Further analysis reveals the differentiated effects of digital-real technology integration across various contexts, aiming to provide unique guidance for firms of different types and characteristics. Heterogeneity analysis based on ownership shows that digital-real integration significantly promotes ESG performance in non-state-owned enterprises (non-SOEs), but not in state-owned enterprises (SOEs). This reflects that non-SOEs rely more heavily on technological integration to enhance legitimacy and market competitiveness. This differentiated phenomenon may stem from three key factors. Regarding resources and institutions, SOEs benefit from policy support and resource guarantees [51], and their ESG performance is primarily driven by administrative assessments, leaving limited marginal room for contribution from technology integration. In contrast, non-SOEs face market competition and financing constraints, making them more reliant on technology integration to improve operational efficiency and governance transparency. Organizationally, SOEs generally exhibit structural rigidity and managerial inertia, which slows the release of technological potential [52], whereas non-SOEs, with their organizational flexibility, can integrate technology into business processes more rapidly [53]. Finally, from a legitimacy-building perspective, SOEs already possess a high base of institutional trust, while non-SOEs need to leverage digital-real integration to shape a responsible market image to gain recognition from investors and the public. Consequently, the enabling effect of technology on ESG improvement is more pronounced for them.

The results on regional heterogeneity indicate that the effect of digital-real integration is more pronounced among firms in the central and western regions, while it is insignificant in the eastern region. This suggests that regional development disparities lead to a divergence in marginal effects. Firms in the eastern region started earlier and have achieved higher levels in both digitalization and ESG management. Consequently, the marginal improvement effect of digital-real integration is easily diluted by their established foundation, resulting in weaker statistical significance. In contrast, firms in the central and western regions exhibit significant shortcomings in digital infrastructure and ESG practices. Once technological integration is implemented, it can quickly address these management gaps, leading to notable improvements in areas such as green production and social responsibility. Furthermore, faced with regional development bottlenecks, firms in the central and western regions are more motivated to leverage digital-real integration as a strategic tool to break through resource constraints and build regional competitiveness [54]. Their stronger willingness and intensity in implementing technological transformations consequently lead to more evident ESG enhancement effects.

The heterogeneity tests across industries yield findings that deviate from conventional expectations: digital-real technology integration shows no significant effect in heavily polluting firms but exhibits a significant positive effect in non-heavily polluting firms. This indicates that a compliance-oriented approach under stringent policy pressure leaves limited room for improvement in heavily polluting enterprises. The ESG behaviors of these firms are primarily driven by regulatory compliance, with their environmental investments often being passive responses. This results in limited marginal improvement potential from digital-real integration. Furthermore, such firms are typically large in scale and complex in hierarchy, exhibiting strong organizational inertia that hinders the deep embedding of new technologies into production and governance processes [54]. In contrast, non-heavily polluting industries, operating under a relatively relaxed regulatory environment, rely more on market competition and stakeholder monitoring to shape their ESG performance. Here, digital-real integration effectively enhances resource efficiency, increases information transparency, and directly translates into ESG competitive advantages through improved external interactions, thus demonstrating a more pronounced facilitative effect.

Further regression results on the individual ESG pillars also reveal differences: digital-real technology integration has a significant positive effect on the environmental (E) and governance (G) dimensions, but an insignificant effect on the social responsibility (S) dimension. The likely reason is that digital-real integration directly enhances environmental performance through data-driven optimization of energy efficiency and emission control. Simultaneously, it improves governance levels by utilizing digital platforms to enhance information disclosure and process traceability. In contrast, improvements in the social dimension are more limited, as they depend more heavily on non-technological factors like corporate culture and values, and the outcomes are often difficult to quantify in the short term. This indicates that while technological empowerment can strongly support the environmental and governance dimensions, comprehensive improvement in social responsibility still requires synergistic drives from institutional and cultural factors.

5.1. Theoretical Implications

This study contributes to the emerging literature on digital transformation and corporate sustainability in several ways.

First, it reverses the conventional causal perspective by positioning digital–real technology integration as a driver of ESG performance, rather than merely a response to ESG pressure. This perspective integrates dynamic capability theory and stakeholder theory, showing that digital integration enables firms to sense opportunities, seize them, and reconfigure internal resources to achieve sustainable value creation.

Second, by introducing organizational inertia as a moderating mechanism, this study extends dynamic capability theory into the organizational behavior domain. The findings highlight that technological capability alone is insufficient for sustainability transformation; instead, the institutional and cognitive flexibility of firms determines whether digital integration can translate into real ESG improvement.

Third, the heterogeneity findings enrich contextualized research in the Chinese setting. Differences across ownership types, regions, and industries reveal how institutional environments shape the boundary conditions of technology-driven sustainability. Specifically, the weaker effects in SOEs and heavily polluting firms indicate that policy-driven compliance may crowd out innovation-driven sustainability efforts.

Finally, the sub-dimensional results demonstrate that digital–real integration has stronger impacts on environmental and governance performance than on social responsibility, suggesting that technology-oriented approaches are more effective in areas requiring data transparency and operational efficiency than in value-driven domains such as social welfare and employee well-being.

5.2. Managerial Implications

The findings offer several actionable insights for firms and policymakers.

First, firms should treat digitalization not merely as a technological upgrade but as a strategic transformation that embeds digital technologies—such as big data, AI, and IoT—into production, supply chain management, governance structures, and information disclosure. Only through deep integration can firms shift from “technology empowerment” to “business reconfiguration,” enabling systemic improvements in ESG performance. Policymakers should also recognize the critical role of digital–real integration in enhancing corporate sustainability by improving data governance, expanding green finance mechanisms, and strengthening ESG disclosure standards.

Second, the moderating effect of organizational inertia underscores the importance of organizational flexibility. Managers must accompany technology adoption with structural and cultural transformation to avoid the “active investment but passive execution” paradox. Reducing hierarchical rigidity, promoting learning-oriented cultures, and empowering forward-looking leadership can help firms reconfigure resources efficiently and internalize digital transformation into ESG practices.

Third, based on the heterogeneity findings across ownership types and regions, this study proposes targeted and differentiated policy implications to establish a precise “policy toolkit”: For state-owned enterprises (SOEs), the policy focus should shift from external administrative constraints to internal governance activation. Regulators could incorporate the depth and application effectiveness of digital-real integration into the ESG performance evaluation system for SOE executives. Furthermore, SOEs should be encouraged to establish dedicated digital innovation funds and green technology incubators to deeply align technological integration with strategic transformation. For non-state-owned enterprises, the core is to lower transformation barriers and strengthen market incentives. The government could introduce specific “Digital-Real Integration-ESG” subsidized loan programs and tax credit policies to directly reduce their financial costs of transition. Simultaneously, financial institutions should be encouraged to develop ESG-linked green credit products, providing market-based financing solutions. For firms in central and western regions, the key lies in addressing foundational capability gaps. Policy should focus on creating regional industrial digitalization public service platforms to offer accessible data, computing power, and technical consulting support. Additionally, fiscal transfers for digital infrastructure construction should be prioritized in these regions, and leading eastern firms should be incentivized to engage in cross-regional technical collaboration through a “digital enclave” model.

Fourth, in heavily polluting industries, the lack of significant ESG improvement underscores the need for stronger external incentives. Policymakers should embed digital governance tools—such as intelligent monitoring and transparent disclosure—into environmental regulations, while capital markets and public oversight can exert additional legitimacy pressure. Combining external incentives with internal adjustments can help such industries escape their high-pollution, high-consumption development path.

Finally, the empirical findings on ESG sub-dimensions offer insights for both practice and evaluation system refinement. Digital-real integration demonstrates significant effects on environmental (E) and governance (G) performance, but its impact on the social responsibility (S) dimension remains limited. This suggests that companies need to incorporate more culture- and value-driven mechanisms into their social responsibility practices. It also indicates room for improvement in how current ESG rating systems measure social performance. Future ESG rating standards should better capture corporate contributions in the digital context, for instance, by introducing quantifiable metrics such as employee rights protection and supply chain responsibility management, to refine the assessment accuracy of corporate social responsibility.

5.3. Limitations and Future Research Directions

Despite its contributions, this study has several limitations that suggest directions for future research.

First, the measurement of digital–real technology integration relies primarily on patent data. While objective, it may not fully capture the embeddedness of digitalization in firms’ operations or management practices. Future studies could integrate textual analysis of corporate reports, survey data, or case studies to construct more comprehensive indicators.

Second, organizational inertia is represented through quantitative proxies, which may not fully reflect its cognitive or cultural dimensions. Future research could employ qualitative or mixed-method approaches to better understand how inertia operates across different organizational contexts.

Third, the limited impact of digital–real integration in heavily polluting industries may reflect time lags or slow technological diffusion. Longitudinal analyses or dynamic modeling could help capture the delayed effects of digitalization on ESG outcomes.

Finally, this study focuses on the Chinese context, where institutional and policy environments play a distinct role in shaping firm behavior. Future comparative studies across different institutional settings could test the generalizability of these findings and deepen our understanding of how digital transformation interacts with sustainability governance worldwide. Regarding model specification, this study lacks more rigorous tests for endogeneity issues. Future research could identify more suitable instrumental variables and further investigate the driving pathways through which digital-physical technology integration enhances ESG performance.