Data Asset Information Disclosure and Supply Chain Resilience: Evidence from China with a Sustainable Development Perspective

Abstract

With global supply chain restructuring and the rapid development of the digital economy, improving supply chain resilience has become a critical issue for achieving sustainable development. This study empirically investigates the impact of data asset information disclosure on supply chain resilience using a sample of Chinese A-share listed companies from 2012 to 2024. The findings reveal that data asset information disclosure significantly enhances corporate supply chain resilience. This conclusion remains robust after a series of robustness tests (excluding impression management risk, alternative measurement of the core explanatory variable, and sample exclusion of special industries and shock periods) and endogeneity treatments (lagged regression, instrumental variable approach, and propensity score matching). Mechanism analysis indicates that disclosure enhances resilience by promoting technological innovation and alleviating financing constraints. Heterogeneity analysis shows that the positive effect is more pronounced in non-state-owned enterprises, firms in core supply chain positions, enterprises with high supply chain concentration, regions with a higher degree of marketization, and industries with low competition intensity. This study contributes to the literature on sustainable supply chain governance and offers practical implications for strengthening supply chain resilience and achieving high-quality sustainable development.

1. Introduction

Considering significant restructuring of global industrial and supply chains and escalating uncertainties, bolstering supply chain resilience has emerged as a fundamental concern for high-quality corporate advancement and sustainable economic and social progress. The 20th National Congress of the Communist Party of China and the Third Plenary Session of the 20th Central Committee have explicitly underscored the necessity of enhancing the resilience and security of industrial and supply chains, emphasizing the strategic significance of stable supply chains for national security and sustainable development. Nonetheless, amidst various shocks such as escalating geopolitical tensions, deglobalization trends, and recurrent public health crises, along with increasingly intricate supply chain networks and a sophisticated division of labor, persistent issues including production stagnation, logistical disruptions, and significant demand fluctuations have arisen. A 2023 study from the World Economic Forum indicates that 92% of multinational corporations have encountered significant supply chain interruptions over the past five years, resulting in an average annual revenue decline above 6%, hence emphasizing the critical need for resilient supply chains [1]. Supply chain resilience denotes a system’s capacity to swiftly recuperate, adapt dynamically, and enhance itself following internal and external disruptions, hence underpinning the sustainable functioning of supply chains [2]. The current advancement of corporate supply chain resilience encounters several obstacles: dependence on critical resources diminishes shock resistance, information asymmetry between supply and demand induces the bullwhip effect, and tight network connections exacerbate cross-entity risk transmission [3]. Simultaneously, informational obstacles between upstream and downstream partners further impede joint responses and swift recovery. Consequently, the construction of a secure, stable, and sustainable supply chain system has emerged as a significant issue of mutual interest for both academia and industry.

In the digital economy, the resource-based view, dynamic capability theory, and other frameworks offer significant theoretical insights into the mechanisms underlying supply chain resilience. Current research has systematically delineated the factors influencing supply chain resilience from both internal and external perspectives: internally, corporate digital transformation, supply chain integration, inventory management, and organizational learning have been demonstrated to significantly enhance resilience [4,5,6]; externally, policy support, the quality of collaborative relationships, industry competition, and the developmental level of new regional quality productive forces also exert substantial influence [7,8,9]. Furthermore, supply chain finance offers crucial financial assistance for enhancing resilience by mitigating financing limitations and optimizing resource allocation [10]. While the studies provide varied insights into resilience enhancement, they predominantly concentrate on conventional production elements and governance frameworks, inadequately addressing the dual imperatives of long-term supply chain sustainability and risk mitigation, and are unlikely to meet the practical requirements of environmentally sustainable, low-carbon development and high-quality growth in the digital economy era.

The enactment of policies like the Interim Provisions on the Accounting Treatment of Enterprise Data Resources has officially established the legal status of data assets as essential intangible assets, leading to heightened focus on the valuation and standardized governance of data elements [11,12]. A data asset is a data resource that is legally owned or controlled by an entity, has the potential to generate future economic advantages, and exists in either digital or electronic form [13]. Data asset information disclosure pertains to the practice whereby enterprises reveal details regarding the ownership, application contexts, value realization, and risk management of data assets via annual reports, announcements, and other mediums in compliance with institutional standards, serving as a crucial link between internal data value and external stakeholder perception [11]. Current research indicates that the disclosure of data asset information can significantly diminish information asymmetry and convey favorable signals concerning technological capability and developmental potential. In the capital market, such information enhances pricing efficiency and diminishes stock price synchronization [14]. Regarding company operations, it can attract institutional investors, alleviate financing constraints, and enhance innovation input and investment efficiency [15,16]. In terms of stakeholder interactions, transparency fosters the trust of financial institutions, enabling companies to secure increased credit assistance and reduced financing costs [17].

A limited number of studies have commenced examining the cross-enterprise spillover effects of data asset information disclosure, revealing that customers’ data asset disclosure positively influences suppliers, while enhanced information transparency mitigates overall supply chain risks [18,19]. Nonetheless, certain studies indicate that unstructured and voluntary data sharing may serve as a mechanism for impression management, and inappropriate disclosure could result in the leaking of trade secrets and diminished competitive advantages, hence intensifying supply chain vulnerability [20]. Standardized and transparent disclosure of data asset information can theoretically provide reliable support for collaborative decision-making between upstream and downstream entities, optimize production planning, ordering strategies, and credit allocation, and significantly contribute to the construction of supply chain resilience [1]. Nonetheless, this mechanism still requires stringent empirical validation. Can the disclosure of data asset information genuinely enhance supply chain resilience? What are its internal transmission mechanisms? Does the impact demonstrate considerable variety based on variations in enterprise characteristics, industry conditions, and regional institutions? Addressing these inquiries holds substantial theoretical and practical importance for enhancing data element governance and fostering sustainable supply chain development.

Although existing studies have identified the key influencing factors of supply chain resilience from multiple dimensions, and some literature has explored the value effects of data asset information disclosure in capital markets, corporate operations, and stakeholder relationships, there remains a clear research gap at the intersection of these two fields. On the one hand, research on supply chain resilience has largely focused on corporate digital transformation, supply chain finance, and external policy environments, with insufficient attention paid to the emerging signaling mechanism of data asset information disclosure. On the other hand, studies on data asset information disclosure have primarily centered on capital market pricing and firm characteristics, with limited extension to supply chain collaboration and resilience building. To address this gap, this study takes Chinese A-share listed companies from 2012 to 2024 as the research sample, combining manually collected data on data asset information disclosure with supply chain-related indicators from CSMAR. It systematically examines the impact, transmission mechanisms, and boundary conditions of data asset information disclosure on supply chain resilience, providing empirical support for sustainable supply chain development.

The marginal contributions of this study are primarily evident in four aspects. First, it extends research on the influencing factors of supply chain resilience by the endogenous perspective of information disclosure, revealing the role of data asset signals in supply chain collaboration, risk mitigation, and long-term sustainable operations, thereby offering a new pathway for resilience building and sustainable transformation. Second, it broadens the investigation into the economic consequences of data asset information disclosure, shifting the analytical focus from capital markets and internal corporate performance to the supply chain context, enriching the micro-evidence of how data elements facilitate sustainable supply chain development. Third, it systematically elucidates the transmission mechanisms through two pathways—technological innovation and financing constraints—and conducts heterogeneity analysis integrating enterprise, industry, and market dimensions, thereby refining the theoretical logic of how data elements drive supply chain resilience and sustainable development. Fourth, it provides actionable empirical references and decision-making support for enterprises to optimize supply chain management and for policymakers to improve the data asset disclosure system and the construction of data factor markets, contributing to the high-quality sustainable development of supply chains.

2. Theoretical Analysis and Research Hypotheses

2.1. The Direct Impact of Data Asset Information Disclosure on Supply Chain Resilience

In the context of extensive global supply chain restructuring and sustainable development transformation, information asymmetry and inadequate collaboration are the primary obstacles hindering the enhancement of supply chain resilience. This paper posits that data asset information disclosure, as an innovative governance mechanism in the digital economy, can systematically bolster supply chain resilience by optimizing the information environment, enhancing resource coordination, and improving risk response capabilities. Drawing on information asymmetry theory, resource dependence theory, and organizational resilience theory, we argue that such disclosure supports the long-term stability of supply chain operations.

Initially, data asset information disclosure significantly improves the information transmission environment within the supply chain. It dismantles communication barriers among participants and establishes a robust foundation for risk mitigation. Traditional supply chains suffer from widespread information asymmetry between core enterprises and their upstream and downstream partners. Suppliers struggle to accurately understand the actual demand and inventory levels of downstream customers, resulting in a misalignment between production and market demand. Downstream enterprises lack complete visibility into suppliers’ production capacities and potential risks, hindering timely responses to supply disruptions. Financial institutions, limited by insufficient overall supply chain operational information, face constraints in credit decision-making, which indirectly impedes capital allocation for resilience enhancement [21]. The disclosure of data asset information substantially reduces information asymmetry through standardized and transparent information dissemination, thereby providing reliable decision-making foundations for all parties involved. The types, scale, application contexts, and value realization pathways disclosed by enterprises provide essential insights for upstream and downstream partners. Upstream suppliers can optimize production schedules based on demand forecasts from core enterprises, mitigating fluctuations caused by inventory backlogs or supply shortages. Downstream enterprises can strategically plan orders and inventory using the disclosed production and logistics data. Financial institutions can more precisely evaluate supply chain risks by utilizing multi-link data asset information, offering targeted financial support for resilience development [11,17].

Secondly, data asset information disclosure enhances internal resource integration and collaborative allocation within the supply chain, thereby improving post-shock recovery efficiency. As a fundamental production factor in the digital economy, data assets are characterized by non-rivalry, shareability, and value multiplication, all of which facilitate value enhancement in supply chain collaboration [22]. Unlike conventional factors, data assets can be shared across multiple entities, amalgamating disparate resources and capabilities into collective benefits for the supply chain through information exchange [10]. Data asset information disclosure serves both as a means for enterprises to showcase their resource capabilities and as a crucial mechanism for enhancing supply chain resource integration. By revealing pertinent data, enterprises communicate their resource foundation and collaborative potential to partners, thereby attracting high-quality resources to the supply chain system [23]. Demand forecasts and customer analysis data provided by core enterprises can assist suppliers in accurate capacity planning and product enhancement. Meanwhile, process optimization and quality control information shared by suppliers can improve core enterprises’ assessments of supply chain reliability and foster long-term cooperative relationships [24]. This resource coordination mechanism significantly enhances the efficiency of supply–demand matching and the security of cash flow within the supply chain, thereby improving its capacity for rapid recovery after disruptions and ensuring a robust resource foundation for the stable and sustainable operation of the supply chain.

Ultimately, data asset information disclosure strengthens the adaptive capacity and sustainable development potential of the supply chain, thereby reinforcing its long-term viability. As a long-term dimension of supply chain resilience, adaptive capacity directly influences firms’ ability to undergo self-transformation and establish enduring competitive advantages following disruptions [2]. Data asset disclosure consistently enhances the supply chain’s capacity to adjust to external conditions by reinforcing critical aspects such as innovation output and human capital, thereby providing a fundamental impetus for long-term growth. The disclosed information creates a shared cognitive framework for supply chain participants, facilitating a cohesive strategy for risk management and innovation. It also enhances collaborative efforts in digital transformation and technological development, thereby offering directional guidance for improving supply chain adaptive capacity [25]. Conversely, information disclosure facilitates the dissemination of knowledge and managerial expertise throughout the supply chain, converting the competitive advantages of individual enterprises in data analysis, risk mitigation, and sustainable management into collective competencies of the entire supply chain. This process consistently enhances the dynamic adaptability of the supply chain, ultimately achieving a virtuous cycle of resilience enhancement and sustainable development. Based on the above theoretical analysis, we propose the following hypothesis:

H1. 

Data asset information disclosure has a significant positive impact on supply chain resilience.

2.2. Indirect Impact of Data Asset Information Disclosure on Supply Chain Resilience

Beyond its direct effects, data asset information disclosure can indirectly enhance supply chain resilience through two major transmission pathways: accelerating technological innovation and alleviating financing constraints. These pathways provide multidimensional support for the sustainable, stable, and efficient operation of supply chains. Supply chain resilience essentially reflects a firm’s ability to withstand, recover from, and develop after external shocks. Technological innovation is the core driver for firms to build long-term competitiveness and improve dynamic adaptability, while financing constraints represent a key bottleneck limiting firms’ risk resistance and emergency recovery capabilities. These two pathways correspond respectively to the long-term development and short-term risk dimensions of supply chain resilience. They are also highly consistent with the signaling value of data asset information disclosure—on the one hand, it releases signals of firms’ innovation potential to the capital market, attracting resources for technological innovation; on the other hand, it reduces information asymmetry and lowers financing costs, providing financial support for firms to cope with risks. Thus, these two pathways constitute the key transmission channels through which data assets enhance supply chain resilience. Based on the above analysis, this paper examines the mechanisms from these two dimensions: technological innovation and financing constraints.

2.2.1. Trajectory of Technological Innovation

The disclosure of data asset information systematically promotes technological innovation through two channels: resource provision and collaborative innovation. Technological innovation is an essential catalyst for firms seeking to overcome developmental constraints and achieve core competitiveness. However, its high investment requirements, elevated risks, and extended cycles make information asymmetry the primary obstacle to innovation [26]. Information asymmetry between investors and enterprises often prevents innovation initiatives from securing consistent financial backing. At the same time, informational barriers between upstream and downstream partners obstruct the collaborative integration of innovation resources. Data asset information disclosure serves as a reliable signaling mechanism in the digital economy, effectively resolving these difficulties. By disclosing the scale, ownership, application scenarios, and value creation pathways of their data assets, firms convey favorable signals to the capital market regarding their digital transformation and innovation potential. This attracts venture capital, private equity, and other innovation-oriented funds, while reducing the risk premium imposed by financial institutions on innovation projects. It thus ensures ongoing financial support for R&D activities [27]. Conversely, the disclosure of data asset information enables suppliers, customers, and other partners to understand the firm’s technological framework and innovation trajectory. This facilitates the adjustment of R&D strategies to achieve resource complementarity and risk sharing [22]. It assists firms in precisely identifying market demand, optimizing technological pathways, avoiding unwise investments, and ultimately accelerating the technological innovation process [16].

Technological innovation significantly enhances supply chain resilience through three mechanisms: risk resistance, dynamic adaptation, and rapid recovery. These mechanisms serve as the fundamental drivers for augmenting the system’s ability to withstand disturbances. Technological innovations such as artificial intelligence, big data, and the Internet of Things can enhance production processes, enable precise demand forecasting and real-time equipment monitoring, and mitigate risks including production disruptions and inventory imbalances, thereby bolstering system stability [25]. Dynamic adaptation, achieved through innovations such as flexible production and intelligent logistics, enhances the operational flexibility of the supply chain. This allows firms to swiftly adjust their production and distribution strategies in response to demand fluctuations and external changes, thereby improving adaptability [4]. Regarding shock recovery, big data analytics and intelligent decision-making technologies can swiftly identify the scope of disturbances, optimize resource allocation, accelerate operational restoration, and shorten the recovery cycle following disruptions [21]. Ongoing technological iteration can foster business model innovation within the supply chain and reduce reliance on a single market or technology, thereby fundamentally enhancing its long-term sustainable resilience. Based on the above analysis, this paper proposes the following hypothesis:

H2. 

Data asset information disclosure enhances supply chain resilience by accelerating technological innovation.

2.2.2. Trajectory of Financing Constraints

The disclosure of data asset information alleviates corporate financing constraints by reducing information asymmetry and conveying positive signals of high-quality development, thereby establishing a stable financial foundation for supply chain resilience. Financing constraints primarily arise from information asymmetry between financial institutions and firms, which complicates the assessment of asset value and future cash flows, thereby raising financing thresholds and costs. As a fundamental intangible asset in the digital economy, data assets—when disclosed by enterprises in terms of ownership, profit models, application scenarios, and related information—can effectively bridge the information gap in traditional financial reports. They clarify asset structure, development potential, and value space, provide a verifiable evaluation basis for financial institutions, and reduce uncertainty regarding future earnings [14]. At the same time, proactive and standardized information disclosure communicates positive signals of effective governance and transparency. This alleviates financial institutions’ concerns regarding moral hazard, resulting in more favorable credit conditions and extended financing durations [27]. Moreover, high-quality data asset information disclosure can attract institutional investors, supply chain finance platforms, and other funding sources. It can also expand financing channels, improve financing structures, and provide adequate financial support for corporate operations and supply chain resilience building.

The alleviation of financing constraints provides consistent financial backing for supply chain resilience and enhances the system’s risk resistance through three mechanisms: risk prevention, emergency recovery, and network stability. Enhancing supply chain resilience requires continuous capital investment, including safety inventories, digital transformation, and the establishment of emergency response mechanisms [1]. With reduced financing pressure, firms can allocate sufficient resources to establish safety inventories, develop diversified supplier networks, enhance logistics systems and risk warning platforms, and strengthen ex ante risk prevention capabilities. In the event of supply disruptions or demand fluctuations, firms can promptly deploy resources to explore alternative channels, adjust production arrangements, restore logistics connections, and improve response speed and recovery capacity following disruptions [28]. At the same time, alleviating financing constraints improves capital flow between upstream and downstream partners, reduces risk transmission caused by capital chain disruptions, and strengthens the overall stability of the supply chain network. This provides robust support for the sustainable functioning of the supply chain [10]. Based on the above analysis, this paper proposes the following hypothesis:

H3. 

Data asset information disclosure enhances supply chain resilience by alleviating financing constraints.

3. Research Design

3.1. Model Settings

This study empirically investigates the impact of data asset information disclosure on corporate supply chain resilience and identifies its underlying transmission mechanisms by constructing a two-way fixed effects baseline regression model and mediation effect models as follows:

$${SCR}_{it}={\alpha }_0+{\alpha }_1{DAID}_{it}+\sum {\alpha }_k{Controls}_{kit}+{\mu }_i+{\delta }_t+{\epsilon }_{it},$$

(1)

$$M_{it}={\beta }_0+{\beta }_1{DAID}_{it}+\sum {\beta }_k{Controls}_{kit}+{\mu }_i+{\delta }_t+{\epsilon }_{it},$$

(2)

$${SCR}_{it}={\gamma }_0+{\gamma }_1{DAID}_{it}+{\gamma }_2{M_{it}+\sum \gamma }_k{Controls}_{kit}+{\mu }_i+{\delta }_t+{\epsilon }_{it},$$

(3)

where i denotes the firm and t denotes the year. The dependent variable, ${SCR}_{it}$, measures the supply chain resilience of firm i in year t. The core independent variable, ${DAID}_{it}$, represents the level of data asset information disclosure of firm i in year t. $M_{it}$ serves as the mediating variable, corresponding to the two transmission pathways of technological innovation and financing constraints, respectively. ${Controls}_{kit}$ represents a set of control variables that may affect supply chain resilience. ${\mu }_i$ represents the firm-fixed effect, accounting for time-invariant firm heterogeneity.${\delta }_t$ represents the year-fixed effect, accounting for contemporaneous shocks such as macroeconomic trends and policy changes. ${\epsilon }_{it}$ represents the stochastic error term.

Model (1) is used to evaluate the overall impact of data asset information disclosure on supply chain resilience, corresponding to Hypothesis 1. Models (2) and (3) establish the framework for testing the mediation effects to validate the roles of technological innovation and the alleviation of financing constraints, corresponding to Hypotheses 2 and 3.

3.2. Variable Definition and Description

3.2.1. Dependent Variable: Supply Chain Resilience (SCR)

Supply chain resilience refers to a firm’s comprehensive ability to maintain supply chain stability, withstand risk disruptions, and rapidly restore operations in the face of external uncertainties. It also serves as a crucial foundation for achieving long-term sustainable supply chain development. Drawing on the theoretical framework of resilience and the practical characteristics of Chinese firms, and following Zhou (2026) and Yao (2025), this study constructs a supply chain resilience evaluation system based on three core dimensions: resistance capacity, recovery capacity, and development capacity [29,30]. These dimensions comprehensively capture the resilience characteristics throughout the entire process of risk shocks—covering the pre-shock, during-shock, post-shock, and long-term development stages. Specifically, resistance capacity corresponds to structural defense before shocks occur, recovery capacity corresponds to flexible adjustment during shocks and restoration after disruptions, and development capacity emphasizes that firms can not only return to their original state after shocks but also achieve leapfrog growth through innovation and learning. The precise mix of indicators and their weight distribution are outlined in

Table 1. Evaluation Index System and Weights of Supply Chain Resilience.

Dependent Variable	First-Level Indicator	Second-Level Indicator	Specific Data	Indicator Attribute	Weight
Supply Chain Resilience	Resistance Capacity	Gross Profit Margin	(Operating Income—Operating Cost)/Operating Income	Positive	0.0013
		Current Ratio	Current Assets/Current Liabilities	Positive	0.0243
		Total Patents Granted	Current Assets/Current Liabilities	Positive	0.1152
	Recovery Capacity	Operating Income Scale	Operating income (in 100 million yuan)	Positive	0.1053
		Net Cash Flow from Operations	Net cash flow generated from operating activities (in 100 million yuan)	Positive	0.0037
		Net Value of Fixed Assets	Fixed assets at cost—Accumulated depreciation—Impairment provision for fixed assets (in 100 million yuan)	Positive	0.0194
		Sales to Top 5 Customers	Total sales to the top five customers (in 100 million yuan)	Positive	0.2910
		Purchases from Top 5 Suppliers	Total purchases from the top five suppliers (in 100 million yuan)	Positive	0.1631
	Development Capacity	Net Profit Growth Rate	(Net profit in current year—Net profit in previous year)/Net profit in previous year	Positive	0.0030
		R&D Investment	R&D investment amount (in 100 million yuan)	Positive	0.2571
		Proportion of Employees with Bachelor’s Degree or Above	Number of employees with bachelor’s degree or above/Total number of employees	Positive	0.0166

Note: Weights are objectively assigned using the entropy weight method, reflecting the degree of variation across sample observations rather than theoretical importance. Some financial indicators (e.g., gross profit margin and net cash flow from operations) receive lower weights due to concentrated sample distributions, but they remain valuable for capturing supply chain resilience and are therefore retained.

.

Resistance Capacity. Resistance capacity measures a firm’s ability to maintain operations in the face of shocks, reflecting its inherent “robustness.” This study selects gross profit margin, current ratio, and total annual patents granted as secondary indicators. Gross profit margin reflects a firm’s cost control and market competitiveness; a higher margin implies stronger risk buffering capacity and serves as the financial foundation for resisting external shocks. The current ratio measures short-term solvency; sufficient current assets are key to managing liquidity risks and maintaining supply chain stability. Total annual patents granted reflects a firm’s technological innovation level and technological barriers; a greater number of patents indicates stronger resistance to technological changes and market disruptions.

Recovery Capacity. Recovery capacity measures a firm’s ability to adjust flexibly during shocks and recover quickly after disruptions, reflecting its “rebound speed.” This study selects operating revenue scale, net cash flow from operations, net fixed assets, sales to top five customers, and purchases from top five suppliers as secondary indicators. A larger operating revenue scale indicates stronger market resource integration capacity, enabling faster recovery of supply chain operations after shocks. Net cash flow from operations reflects the net cash flow generated by a firm’s operating activities; stable operating cash flow provides ongoing financial support for supply chain restoration. Net fixed assets represent the physical foundation of production operations; a larger scale of fixed assets provides greater capacity to redeploy production capacity and resume operations after shocks, thereby enhancing recovery capacity. Sales to top five customers and purchases from top five suppliers reflect the transaction scale between a firm and its key customers and key suppliers. Larger transaction volumes imply greater stability in upstream and downstream channels and resource supply, thereby strengthening the firm’s recovery capacity.

Development Capacity. Development capacity measures a firm’s ability to achieve learning transformation and sustained growth after shocks, reflecting its “evolutionary potential.” This study selects net profit growth rate, R&D investment, and the proportion of employees with a bachelor’s degree or above as secondary indicators. Net profit growth rate reflects a firm’s long-term profit growth potential; sustained profitability provides a stable source of funding for supply chain upgrading and resilience building. R&D investment reflects a firm’s capacity for technological upgrading and business model innovation; greater investment enables technology-driven improvements in long-term supply chain resilience. The proportion of employees with a bachelor’s degree or above reflects a firm’s human capital level; a higher proportion of highly educated employees indicates stronger operational management and innovation capabilities, thereby better supporting the sustainable development of the supply chain.

Weighting and Aggregation. Based on the above indicator system, this study employs the entropy weight-TOPSIS method to assign objective weights to each indicator. The entropy weight method automatically determines weights based on the degree of dispersion across sample observations; the greater the variation in an indicator, the lower its information entropy and the higher its weight, thereby avoiding potential biases associated with subjective weighting. The weight distribution of each indicator is presented in Table 1. From the weight distribution, indicators such as sales to top five customers, R&D investment, and purchases from top five suppliers receive relatively high weights, reflecting that these indicators exhibit considerable variation across sample firms, while also suggesting that the stability of upstream and downstream supply chain relationships and technological innovation capacity may be important factors influencing corporate supply chain resilience. On this basis, the TOPSIS method is used to calculate a firm-level comprehensive supply chain resilience score (SCR), where a higher score indicates stronger supply chain resilience.

Reliability and Validity Tests. To verify the scientific validity and effectiveness of the indicator system, this study conducted reliability and structural validity tests. The reliability test results show that the overall Cronbach’s α coefficient is 0.7217, indicating good internal consistency among the secondary indicators. This level of reliability meets the general standards for multidimensional construct research. For structural validity, this study employed principal component analysis (PCA) to reduce the dimensionality of the 11 secondary indicators. The results show that the first three principal components explain 69.29% of the total variance, exceeding the conventional threshold of 60%. This indicates that the indicator system effectively captures the core dimensions of supply chain resilience and demonstrates good structural validity. The factor loading matrix further reveals that each indicator loads highly onto its corresponding theoretical dimension, with clear dimensional distinctions that closely align with the theoretical framework constructed in this study.

3.2.2. Independent Variable: Data Asset Information Disclosure (DAID)

Drawing on the “seed word + Word2Vec similar word expansion” method of Yuan et al. (2022), this study uses text analysis techniques to mine the content of listed companies’ annual reports as the scope for measuring the level of data asset information disclosure [12]. Specifically, based on the definition of data assets in the “Data Asset Management Practice White Paper (Version 5.0)” issued by the China Academy of Information and Communications Technology (CAICT), “data assets” and “data resources” are taken as seed words, and Word2Vec is used to construct a similar word set to obtain keywords related to data assets. This study selects the top ten keywords with the highest similarity, which are: information resources, data mining, data sources, big data, data sharing, massive data, data platforms, data analysis systems, basic information, and knowledge bases. The total frequency of the above keywords appearing in the annual reports is counted. Since this frequency has a right-skewed distribution, it is logarithmically transformed to obtain the core explanatory variable DAID.

3.2.3. Control Variables

To mitigate estimation bias caused by omitted variables, this study follows Yuan et al. (2025) and selects control variables from two dimensions: firm characteristics and financial performance [31]. All selected control variables are clearly defined and distinct from both the core explanatory variable (DAID) and the dependent variable (SCR), ensuring the validity of the regression estimates.

Table 2. Definitions of Control Variables.

Variable Name	Symbol	Measurement
Firm Size	Size	Natural logarithm of total assets at the end of the year
Firm Age	Age	Natural logarithm of the number of years since the firm went public
Ownership Type	SOE	1 if the firm is state-owned, 0 otherwise
Leverage Ratio	LEV	Total liabilities/Total assets
Return on Assets	ROA	Net profit/Total assets
Operating Revenue Growth Rate	Grow	(Current year operating revenue—Previous year operating revenue)/Previous year operating revenue

presents the definitions of the control variables.

3.2.4. Sample Selection and Data Sources

This study uses Chinese A-share listed companies from 2012 to 2024 as the initial research sample and applies the following screening procedures. First, firms in the banking and insurance industries are excluded. Second, samples designated as ST, *ST, or those under delisting risk during the sample period are removed. Third, observations with missing data on key financial and corporate governance variables are eliminated. After these screening steps, the final sample consists of 27,928 firm-year valid observations. To mitigate the potential influence of outliers on the regression results, all continuous variables are winsorized at the 1st and 99th percentiles.

Annual report texts of listed companies are collected using Python scripts (Python 3.12), and text analysis algorithms are employed to obtain the word frequency statistics required to measure supply chain resilience and data asset information disclosure. Financial indicators, corporate governance information, and supply chain resilience data are primarily sourced from the China Stock Market & Accounting Research Database (CSMAR). Additional validation is conducted using the Wind Database, the RESSET Financial Research Database, and interim announcements of listed companies.

4. Results and Analysis

4.1. Descriptive Statistics and Multicollinearity Test

Descriptive statistical analysis is performed on the sample data to illustrate the distributional characteristics of the main variables.

Table 3. Descriptive Statistics of Variables.

Variable	Observations	Mean	Std. Dev.	Min	Median	Max
DAID	27,928	0.905	1.047	0.000	0.624	4.030
SCR	27,928	0.403	0.247	0.022	0.436	1.058
Size	27,928	22.319	1.332	19.886	22.078	26.599
Age	27,928	2.021	0.929	0.000	2.079	3.466
SOE	27,928	0.325	0.468	0.000	0.000	1.000
LEV	27,928	0.399	0.196	0.049	0.391	0.899
ROA	27,928	0.057	0.120	−1.098	0.069	0.346
Grow	27,928	0.118	0.308	−0.630	0.066	2.609

reports the descriptive statistics for the full sample of 27,928 firm-year observations. The dependent variable, supply chain resilience (SCR), has a mean of 0.403 with a standard deviation of 0.247, ranging from 0.022 to 1.058, indicating considerable heterogeneity in firms’ ability to withstand supply chain disruptions. The core explanatory variable, data asset information disclosure (DAID), has a mean of 0.905 and a standard deviation of 1.047, ranging from 0 to 4.030, demonstrating sufficient variability for empirical analysis. The control variables exhibit distributional patterns consistent with typical firm-level financial metrics, validating the rationale for their inclusion. Overall, the sample data are well-distributed and align with theoretical expectations, providing a solid foundation for subsequent regression analysis.

To ensure the reliability of the regression model, a multicollinearity test is conducted on all independent variables.

Table 4. Multicollinearity Test Results.

Variable	VIF	1/VIF
DAID	1.42	0.706
Size	1.87	0.535
Age	1.56	0.640
SOE	1.40	0.716
LEV	1.71	0.585
ROA	1.38	0.726
Grow	2.13	0.884
Mean VIF	1.638

reports the variance inflation factors (VIFs) for each variable. The results show that the VIF values range from 1.38 to 2.13, all well below the commonly used threshold of 10. The mean VIF is 1.638, and the corresponding tolerance values (1/VIF) are all greater than 0.4. These findings indicate that multicollinearity is not a concern, and the selected variables are sufficiently independent from one another, satisfying the prerequisite for regression analysis.

4.2. Baseline Regression

Table 5. Baseline Regression Results.

Variable	(1) SCR	(2) SCR
DAID	0.0293 ***	0.0267 ***
	(0.0022)	(0.0023)
Size		0.0170 ***
		(0.0040)
Age		0.0398 ***
		(0.0030)
SOE		0.0107
		(0.0088)
LEV		−0.0818 ***
		(0.004)
ROA		0.2397 ***
		(0.0228)
Grow		0.0022 ***
		(0.0029)
Constant	−0.3869	−0.0255
	(0.0032)	(0.0838)
Firm FE	Yes	Yes
Year FE	Yes	Yes
N	27,928	27,928
R2	0.5236	0.5637

Notes: Robust standard errors clustered at the firm level are reported in parentheses. *** indicate significance at the 1% level.

presents the baseline regression results of data asset information disclosure (DAID) on supply chain resilience (SCR). Column (1) includes only firm- and year-fixed effects, showing that the coefficient of DAID is positive and significant at the 1% level, providing initial evidence of the positive effect of data asset information disclosure on supply chain resilience. Column (2) further incorporates a set of control variables, including firm size, firm age, ownership type, leverage ratio, return on assets and operating revenue growth rate. The coefficient of DAID remains positive and significant at the 1% level, indicating that after accounting for individual heterogeneity, time trends, and other confounding factors, the enhancing effect of data asset information disclosure on supply chain resilience remains robust. Thus, Hypothesis H1 is empirically supported.

Regarding the control variables, the coefficients of Size, Age, LEV, ROA, and Grow are all statistically significant and exhibit signs consistent with theoretical expectations. Specifically, larger firm size and longer firm age imply richer resource endowments, which contribute to stronger supply chain resilience. A higher leverage ratio indicates greater financial risk, which negatively affects supply chain resilience. A higher return on assets reflects stronger profitability, providing sufficient internal funding for supply chain resilience building. A higher operating revenue growth rate indicates better growth prospects, enabling firms to invest more in supply chain resilience. The coefficient of SOE, although not statistically significant, exhibits the expected sign and is retained as a commonly recognized control variable in the existing literature to ensure the completeness of the model specification.

4.3. Robustness Checks

To ensure the reliability and generalizability of the main findings, this study conducts robustness checks from four dimensions to address concerns related to impression management risk, variable measurement bias, sample specificity, and model specification errors.

4.3.1. Excluding Impression Management Risk

Although the “seed word + Word2Vec similar word expansion” method adopted in this study is a well-established measurement framework in the field of data asset information disclosure research, dictionary-based text analysis may suffer from impression management risk—that is, firms’ data asset disclosure may merely represent strategic impression management rather than substantive information disclosure. To address this concern, this study further conducts a subsample regression and interaction effect model.

Specifically, the full sample is first divided into high-disclosure and low-disclosure groups based on the median value of DAID, and a high-disclosure dummy variable ($High\_DAID$) is constructed. To avoid multicollinearity, DAID is mean-centered to generate $DAID\_c$, and an interaction term ($DAID\_c\times High\_DAID$) is created. The regression model is specified as follows:

$${SCR}_{it}={\nu }_0+{\nu }_1{DAID\_c}_{it}+{\nu }_2{{(DAID\_c\times High\_DAID)}_{it}+\nu }_3{Controls}_{it}+{\mu }_i+{\delta }_t+{\epsilon }_{it}$$

(4)

where ${\nu }_1$ captures the marginal effect of DAID on supply chain resilience at the mean level of DAID, and ${\nu }_2$ captures the additional effect for the high-disclosure group relative to the mean level. The regression results are reported in

Table 6. Robustness Check Results (I).

Variable	(1) SCR	(2) SCR
$DAID\_c$	0.0217 ***
	(0.0027)
$DAID\_c\times High\_DAID$	0.0097 **
	(0.0041)
PDA		0.0704 ***
		(0.0109)
Constant	−0.0003	−0.0787
	(0.0840)	(0.0847)
Controls	Yes	Yes
Firm FE	Yes	Yes
Year FE	Yes	Yes
Observations	27,928	27,928
R2	0.5638	0.5602

Notes: Robust standard errors clustered at the firm level are reported in parentheses. *** and ** indicate significance at the 1% and 5% levels, respectively.

, Column (1). The coefficient of $DAID\_c$ is 0.0217 (p < 0.01), indicating that data asset information disclosure has a significant positive impact on supply chain resilience. The coefficient of the interaction term $DAID\_c\times High\_DAID$ is 0.0097 (p < 0.1), and the joint significance test yields an F-statistic of 5.50 (p = 0.019), suggesting that the promoting effect of DAID on supply chain resilience is significantly stronger in the high-disclosure group. These results rule out the possibility that impression management drives the main findings and confirm the validity of the DAID measure.

4.3.2. Alternative Measurement of the Explanatory Variable

This study uses the ratio of keyword frequency pertaining to corporate data assets within the total word frequency of annual reports (PDA) to replace the original DAID metric and re-estimates the baseline model. This indicator assesses the level of data asset information disclosure based on relative density, thereby mitigating the impact of variations in annual report length across different firms on the measurement outcomes. The regression results are presented in Table 6, Column (2). The coefficient of PDA on supply chain resilience is significantly positive at the 1% level, consistent with the baseline regression findings. This demonstrates that the positive correlation between data asset information disclosure and supply chain resilience persists regardless of whether absolute or relative word frequency is used.

4.3.3. Sample Screening and Removal of Confounding Factors

To account for potential disturbances caused by industry characteristics and specific external shocks, this study implements two types of sample adjustments. First, it excludes samples from high-tech industries to mitigate the influence of systematic discrepancies in data asset disclosure motivations among high-tech firms. Second, it omits samples from 2021 to 2022, a period marked by public health crises, to eliminate the impact of unique external shocks on corporate operational behavior. The regression results following these two adjustments are presented in

Table 7. Robustness Check Results (II).

Variable	(1) SCR	(2) SCR	(3) SCR
DAID	0.0106 ***	0.0292 ***	0.0269 ***
	(0.0033)	(0.0024)	(0.0022)
Constant	−0.0532	0.0559	−0.2432 **
	(0.1688)	(0.0814)	(0.1194)
Controls	Yes	Yes	Yes
Firm FE	Yes	Yes	Yes
Year FE	Yes	Yes	Yes
Province FE	No	No	Yes
Industry FE	No	No	Yes
Observations	10,420	22,009	27,928
R2	0.4807	0.6093	0.5689

Notes: Robust standard errors clustered at the firm level are reported in parentheses. *** and ** indicate significance at the 1% and 5% levels, respectively.

, Columns (1) and (2), respectively. The sign and significance of the coefficient for the core explanatory variable remain consistent with the baseline results, indicating that the positive effect of data asset information disclosure on supply chain resilience is independent of specific industries or particular shocks.

4.3.4. Inclusion of Higher-Dimensional Fixed Effects

Building on the baseline model, this study further incorporates province and industry fixed effects to control for unobserved heterogeneity at the regional and industry levels, while maintaining clustered robust standard errors at the firm and year levels. As shown in Table 7, Column (3), the direction and significance of the coefficient for the core explanatory variable remain largely unchanged, suggesting that the baseline model specification is appropriate and that the main conclusions are robust.

4.4. Endogeneity Tests

The baseline regression may encounter endogeneity problems, including omitted factors, reverse causality, and sample self-selection. Enterprises exhibiting greater supply chain resilience may be more inclined to reveal data asset information, resulting in reverse causality bias. Conversely, unobservable characteristics like managerial competence and business culture may concurrently influence data asset disclosure decisions and supply chain resilience, leading to omitted variable bias. Moreover, systematic disparities exist between firms that reveal data assets and those that refrain from doing so, potentially resulting in sample self-selection bias. To address the aforementioned endogeneity issues, this study employs several methodologies for testing, hence reinforcing the causal framework of data asset information disclosure that enhances the development of sustainable supply chain resilience.

4.4.1. One-Period Lagged Regression of the Primary Explanatory Variable

To mitigate endogeneity bias arising from reverse causality, this study lags the primary explanatory variable by one period (x_lag) and re-estimates the baseline model. The lagged information disclosure behavior is less influenced by the reverse effect of current supply chain resilience, thereby clarifying the direction of causality. The regression results are presented in

Table 8. Endogeneity Test Results.

Variable	(1) SCR	(2) DAID	(3) SCR	(4) SCR
x_lag	0.0041 *
	(0.0023)
IV		0.4818 ***
		(0.0685)
DAID			0.1779 ***	0.0240 ***
			(0.0364)	(0.0023)
Constant	−0.1024			−0.0207
	(0.0987)			(0.0866)
Controls	Yes	Yes	Yes	Yes
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
Observations	27,928	27,928	27,928	22,634
R2	0.5804			0.5224

Notes: Robust standard errors clustered at the firm level are reported in parentheses. *** and * indicate significance at the 1% and 10% levels, respectively.

, Column (1). The coefficient of the lagged core explanatory variable is positive and significant at the 10% level, consistent with the baseline regression findings. This suggests that the positive effect of data asset information disclosure on supply chain resilience is persistent over time, and reverse causality does not substantially alter the main conclusion.

4.4.2. Instrumental Variable Approach

This study uses the software revenue of the region where the firm is located as an instrumental variable and adopts the two-stage least squares (2SLS) method. Regarding relevance, regional software revenue reflects the level of local digital economic development. In regions with better digital infrastructure and more developed software industries, firms undergo faster digital transformation, accumulate richer data assets, and have stronger incentives and capabilities to disclose data asset information. Regarding exogeneity, regional software revenue is a macro-level variable that individual firms cannot influence. Moreover, this variable affects supply chain resilience primarily through promoting digital transformation and data asset accumulation, rather than directly affecting firms’ supply chain relationships.

The validity tests for the instrumental variable show that the first-stage F-statistic is 49.40, well above the Stock–Yogo critical value of 16.38 at the 10% level, rejecting the weak instrument hypothesis. The Kleibergen–Paap rk LM statistic yields a p-value of less than 0.001, rejecting the null hypothesis of under-identification. The first-stage regression results are reported in Table 8, Column (2), where the coefficient of the IV is 0.4818 (p < 0.01), indicating a significant positive correlation between regional software revenue and DAID. The second-stage regression results are presented in Column (3), where the coefficient of DAID is 0.1779 (p < 0.01), consistent with the direction and significance of the baseline results, suggesting that the positive effect of data asset information disclosure on supply chain resilience remains robust after addressing endogeneity concerns.

4.4.3. Propensity Score Matching Technique

To mitigate sample self-selection bias, this study employs the propensity score matching (PSM) method. Firms with data asset information disclosure (x > 0) are designated as the treatment group (treat = 1), while non-disclosing firms serve as the control group (treat = 0). After estimating the propensity score based on control variables, a 1:1 nearest neighbor matching without replacement (caliper = 0.05) is implemented. The final matched sample consists of 22,634 observations. The balance test results show that the standardized bias of each variable decreases significantly after matching. The mean bias declines from 9.7% to 5.4%, and the overall bias B value decreases from 29.7% to 18.3%, satisfying the balance test requirements and indicating that the differences in characteristics between the treatment and control groups are effectively eliminated. The common support test results indicate that 98% of the treatment group samples overlap with the control group, and no significant samples fall outside the common support domain, demonstrating an effective matching procedure. Regression analysis based on the matched sample, reported in Table 8, Column (4), shows that the coefficient of DAID is 0.0240 (p < 0.01), consistent with the baseline findings. This indicates that after mitigating sample self-selection bias, the positive effect of data asset information disclosure on supply chain resilience remains robust.

4.5. Mechanism Tests

Based on the theoretical analysis above, this study conducts mediation effect tests from two pathways—technological innovation and alleviation of financing constraints—to identify the underlying transmission mechanisms through which data asset information disclosure affects supply chain resilience. The results are reported in

Table 9. Mechanism Test Results.

Variable	(1) TI	(2) SCR	(3)Fc	(4) SCR	(5) SCR
DAID	0.0765 **	0.0258 ***	−0.0082 ***	0.0265 ***	0.0288 ***
	(0.0105)	(0.0016)	(0.0006)	(0.0012)	(0.0025)
TI		0.0115 ***			0.0035 **
		(0.0010)			(0.0014)
Fc				−0.0509 ***	−0.1001 ***
				(0.0089)	(0.0358)
Constant	−9.6436 ***	0.1847 ***	−3.6268 ***	0.1848 ***	−0.6036 ***
	(0.2844)	(0.0412)	(0.0043)	(0.0326)	(0.1568)
Controls	Yes	Yes	Yes	Yes	Yes
Firm FE	Yes	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes	Yes
Observations	27,928	27,928	27,928	27,928	27,928
R2	0.1932	0.5744	0.8931	0.5465	0.5830
Sobel test	12.58 ***		−4.15 ***
Mediation effect/Total effect	3.30%		1.57%
Bootstrap test	95% confidence interval is [0.0049, 0.0069]		95% confidence interval is [−0.00068, −0.00024]

Notes: Robust standard errors clustered at the firm level are reported in parentheses. *** and ** indicate significance at the 1% and 5% levels, respectively.

.

4.5.1. Technological Innovation Pathway

Technological innovation is a core driver for firms to build long-term competitiveness and enhance dynamic adaptability. Following Huang et al. (2026), this study uses the number of patent applications (Inv) as an indicator of technological innovation [15]. The regression results are presented in Table 9, Columns (1) and (2). The coefficient of DAID on technological innovation is significantly positive, indicating that data asset information disclosure effectively alleviates R&D funding constraints, identifies innovation directions, and promotes supply chain collaborative innovation, thereby enhancing technological innovation capability. In Column (2), after including technological innovation in the baseline model, the coefficient of Inv on supply chain resilience is significantly positive, and the coefficient of DAID decreases compared with the baseline regression, indicating that technological innovation supports resilience building by strengthening risk resistance, dynamic adaptation, and rapid recovery capabilities. This study further conducts Sobel and Bootstrap (500 repetitions) tests. The Sobel test yields a Z-value of 12.58 (p < 0.01), and the Bootstrap 95% confidence interval excludes zero, with a mediation ratio of 3.30%. These findings confirm the significance of the mediation effect of technological innovation. Thus, data asset information disclosure enhances supply chain resilience by accelerating technological innovation, supporting Hypothesis H2.

4.5.2. Alleviation of Financing Constraints Pathway

Financing constraints represent a key bottleneck limiting firms’ risk resistance and emergency recovery capabilities. Following Shi et al., this study uses the SA index to measure financing constraints (Fc), where a larger index indicates tighter financing constraints [16]. The results are presented in Table 9, Columns (3) and (4). In Column (3), the coefficient of DAID on financing constraints is significantly negative, indicating that data asset information disclosure significantly alleviates corporate financing constraints. The underlying logic is that disclosure reduces information asymmetry and conveys signals of firms’ development potential, thereby broadening financing channels, lowering financing costs, and releasing capital for supply chain operations. In Column (4), after including Fcs in the model, the coefficient of Fcs on supply chain resilience is significantly negative, and the coefficient of DAID decreases compared with the baseline regression, consistent with partial mediation. This suggests that adequate capital support facilitates supply chain digital upgrading, emergency reserve building, and collaborative cooperation, effectively enhancing the system’s resilience. The Sobel test yields a Z-value of −4.15 (p < 0.01), and the Bootstrap 95% confidence interval excludes zero, with a mediation ratio of 1.57%. These findings confirm the significance of the alleviation of financing constraints pathway. Therefore, data asset information disclosure enhances supply chain resilience by alleviating financing constraints, supporting Hypothesis H3.

4.5.3. Joint Test of Mediators

To examine whether the two mediators interfere with each other, this study includes both technological innovation (Inv) and financing constraints (Fc) simultaneously in the regression model. The results are presented in Table 9, Column (5). The coefficient of Inv is significantly positive, the coefficient of Fcs is significantly negative, and the coefficient of DAID is significantly positive. Compared with the separate tests, the signs and significance of the mediators remain largely unchanged, indicating that the two pathways are relatively independent with minimal interference, further validating the robustness of the mechanism analysis.

4.6. Heterogeneity Tests

The baseline regression results indicate that data asset information disclosure has a significant positive effect on corporate supply chain resilience. However, the magnitude of this effect may vary depending on firm characteristics, supply chain position, and external environmental conditions. To comprehensively identify the boundary conditions of the impact of data asset information disclosure on supply chain resilience, this study conducts group tests from five dimensions: ownership type, supply chain position, supply chain concentration, regional marketization level, and industry competition intensity. The results are presented in

Table 10. Heterogeneity Test Results (I).

Variable	(1) SOE	(2) Non-SOE	(3) Core Position	(4) Non-Core Position	(5) High Concentration	(6) Low Concentration
DAID	0.0167 ***	0.0299 ***	0.0339 *	0.0233 ***	0.0418 ***	0.0061 ***
	(0.0043)	(0.0026)	(0.0181)	(0.0112)	(0.0033)	(0.0027)
Constant	−0.0259	−0.0150	0.2422	0.5811	0.0339	−0.3452
	(0.2059)	(0.0885)	(0.5135)	(0.9116)	(0.1171)	(0.1134)
Controls	Yes	Yes	Yes	Yes	Yes	Yes
Firm FE	Yes	Yes	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Observations	9077	18,851	3520	7704	15,972	11,956
R2	0.5126	0.6040	0.1245	0.1275	0.6417	0.1914
Chow test F-statistic	27.95 ***		7.80 ***		27.33 ***

Notes: Robust standard errors clustered at the firm level are reported in parentheses. *** and * indicate significance at the 1% and 10% levels, respectively.

and

Table 11. Heterogeneity Test Results (II).

Variable	(1) High Marketization	(2) Low Marketization	(3) High-Competition Industries	(4) Low-Competition Industries
DAID	0.0274 ***	0.0224 ***	0.0108 ***	0.0409 ***
	(0.0024)	(0.0067)	(0.0029)	(0.0032)
Constant	−0.0276	−0.3375	−0.0753	0.0132
	(0.0832)	(0.3609)	(0.1227)	(0.1084)
Controls	Yes	Yes	Yes	Yes
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
Observations	24,654	3274	13,964	13,964
R2	0.5714	0.5193	0.4513	0.6565
Chow test F-statistic	47.52 ***		40.81 ***

Notes: Robust standard errors clustered at the firm level are reported in parentheses. *** indicate significance at the 1% level.

.

4.6.1. Heterogeneity in Ownership Type

State-owned enterprises (SOEs) and non-state-owned enterprises (non-SOEs) differ significantly in resource acquisition capacity, policy support, and incentives for information disclosure. This study divides the sample into SOEs and non-SOEs. The group regression results presented in Table 10, Columns (1) and (2), show that the coefficients of DAID are significantly positive at the 1% level in both groups, and the coefficient for non-SOEs is higher than that for SOEs. The Chow test yields an F-value of 27.95 (p < 0.01), indicating a significant difference between the two groups. The reason is that non-SOEs face more severe information asymmetry and financing constraints. As a high-quality signal, data asset information disclosure can more effectively alleviate their resource acquisition difficulties, thereby generating a stronger marginal effect on supply chain resilience.

4.6.2. Heterogeneity in Supply Chain Position

Differences in a firm’s position within the supply chain affect its information transmission efficiency and resource coordination capacity. Using information on top five customers and top five suppliers disclosed in annual reports, this study identifies firms’ core positions in the supply chain. Specifically, the number of customers and suppliers for each firm is counted annually, and the total number of supply chain connections (the sum of customers and suppliers) is calculated. A higher total number of connections indicates more transaction relationships with upstream and downstream partners, implying a more central position in the supply chain network. The median value of total connections is calculated for each year, and firms with total connections equal to or greater than the annual median are defined as core enterprises, while the rest are defined as non-core enterprises. Due to incomplete disclosure of customer and supplier names by some firms, the effective sample sizes for this dimension are 3520 (core position) and 7704 (non-core position). The group regression results presented in Table 10, Columns (3) and (4), show that the coefficients of DAID are significantly positive in both groups, and the coefficient for the core position group is higher than that for the non-core position group, with a Chow test F-value of 7.80 (p < 0.01). This finding indicates that firms in core supply chain positions have stronger information integration capabilities and upstream–downstream synergy advantages, enabling them to more fully utilize data asset information disclosure to enhance supply chain resilience.

4.6.3. Heterogeneity in Supply Chain Concentration

Supply chain concentration reflects the degree of a firm’s dependence on its major customers and suppliers, which may affect the value transformation efficiency of information disclosure. This study calculates customer concentration (the ratio of sales to the top five customers to total sales) and supplier concentration (the ratio of purchases from the top five suppliers to total purchases), and constructs a supply chain concentration indicator by averaging the two. The sample is divided into high-concentration and low-concentration groups based on the median value of supply chain concentration. The group regression results presented in Table 10, Columns (5) and (6), show that the coefficients of DAID are significantly positive at the 1% level in both groups, and the coefficient for the high-concentration group is significantly higher than that for the low-concentration group, with a Chow test F-value of 27.33 (p < 0.01). The reason is that firms with high supply chain concentration have closer relationships with their core partners, and data asset information disclosure can more effectively convey trust signals and strengthen collaborative relationships, thereby amplifying its positive effect on supply chain resilience.

4.6.4. Heterogeneity in Regional Marketization Level

The level of regional marketization reflects the development of factor markets and the quality of the institutional environment, which may affect the value realization efficiency of data asset information disclosure. This study adopts the China Provincial Marketization Index compiled by Wang et al. (2021) and divides the sample into high-marketization and low-marketization groups based on the annual median value [32]. The group regression results presented in Table 11, Columns (1) and (2), show that the coefficients of DAID are significantly positive at the 1% level in both groups, and the coefficient for the high-marketization group is higher than that for the low-marketization group, with a significant difference confirmed by the Chow test. Regions with high marketization have more developed factor markets, stricter disclosure regulations, and more efficient supply chain finance systems, which can fully amplify the enabling effect of data asset information disclosure. In contrast, low-marketization regions suffer from inadequate institutional support, which weakens the marginal contribution of disclosure.

4.6.5. Heterogeneity in Industry Competition Intensity

The intensity of industry competition affects the demand for supply chain collaboration and the value transformation efficiency of information disclosure. Following Yuan et al. (2021), this study classifies industries into competitive and non-competitive categories based on the industry classification of listed companies, constructing a dummy variable (competitive industry = 1, otherwise = 0) [33]. The group regression results presented in Table 11, Columns (3) and (4), show that the coefficient of DAID for the low-competition group is significantly higher than that for the high-competition group, with a significant difference confirmed by the Chow test. Firms in highly competitive industries are already highly market-sensitive and have strong supply chain collaboration, with relatively mild information asymmetry, leaving limited room for marginal improvement through disclosure. In contrast, firms in low-competition industries tend to have lower operational efficiency and weaker collaboration, and data asset disclosure can more significantly activate their resilience enhancement potential by breaking down information barriers and optimizing resource allocation.

In summary, the positive effect of data asset information disclosure on supply chain resilience is more pronounced in non-state-owned enterprises, firms in core supply chain positions, firms with high supply chain concentration, regions with high marketization levels, and industries with low competition intensity. These findings provide empirical evidence for different types of firms to formulate differentiated information disclosure strategies.

5. Conclusions and Implications

Against the backdrop of global industrial and supply chain restructuring and escalating external risks, enhancing supply chain resilience has become a critical issue for high-quality corporate development and national security strategy. It also serves as a core pillar for promoting sustainable supply chain development and ensuring the long-term stable operation of industrial and supply chains. As an important pathway for optimizing supply chain collaboration and unlocking the value of data elements in the digital economy, the mechanisms and boundary conditions of data asset information disclosure’s impact on supply chain resilience urgently require systematic investigation. This paper takes Chinese A-share listed companies from 2012 to 2024 as the research sample and empirically examines the impact, transmission mechanisms, and heterogeneous characteristics of data asset information disclosure on supply chain resilience. The findings are as follows:

First, data asset information disclosure significantly enhances corporate supply chain resilience. This conclusion remains robust after a series of robustness tests, including alternative variable measurements, exclusion of special samples and shock years, and the introduction of multidimensional fixed effects. Moreover, after mitigating endogeneity concerns through lagged regression, instrumental variable approaches, and propensity score matching, the positive causal relationship is further verified, providing solid empirical support for sustainable supply chain resilience building. Second, mechanism tests indicate that data asset information disclosure enhances supply chain resilience primarily through two pathways. The first is promoting technological innovation to strengthen risk resistance capabilities, thereby consolidating the technological foundation for sustainable supply chain development. The second is alleviating financing constraints to provide financial support for stable supply chain operations, breaking through the funding bottlenecks of sustainable supply chain development. These two pathways, corresponding to capacity building and resource security respectively, jointly constitute the mechanism through which data asset information disclosure empowers supply chain resilience. Third, heterogeneity analysis reveals that the positive effect of data asset information disclosure on supply chain resilience is more pronounced in non-state-owned enterprises, firms in core supply chain positions, firms with high supply chain concentration, regions with high marketization levels, and industries with low competition intensity. These significant contextual differences provide precise guidance for different types of enterprises to promote sustainable supply chain resilience building through information disclosure.

Based on the above findings, this study proposes policy implications from three dimensions: corporate practice, institutional improvement, and market participation.

At the corporate level, data asset information disclosure should be transformed from a compliance requirement into a strategic tool, achieving a shift from formal disclosure to value creation. First, firms should integrate data asset information disclosure into their supply chain management strategies, systematically establishing a “Data Assets and Digital Capabilities” section in the Management Discussion and Analysis (MD&A) of annual reports, clearly disclosing the scale, ownership, application scenarios, and value creation pathways of data assets in key supply chain links such as procurement optimization, production scheduling, and inventory management. Second, firms should build a virtuous cycle of “disclosure-driven innovation and innovation-reinforced resilience,” attracting innovation capital and alleviating R&D financing pressures through high-quality data asset disclosure, while continuously increasing R&D investment to avoid the crowding out of innovation budgets by short-term performance pressures. Third, firms can use data asset information disclosure to broaden financing channels. It is particularly recommended that small- and medium-sized enterprises and non-state-owned enterprises proactively disclose the value and application prospects of their data assets to financial institutions and supply chain platforms, using data assets as a credit enhancement tool, and channel the released funds into safety inventory building, diversified supplier development, and emergency response mechanism improvement. Finally, firms should formulate differentiated disclosure strategies based on their own characteristics: non-state-owned enterprises should focus on leveraging the branding and financing functions of data asset disclosure; core supply chain enterprises should use disclosure to drive upstream and downstream digital collaboration, enhancing overall chain resilience; enterprises in low-competition industries should use disclosure as a differentiated competitive tool, activating resilience enhancement potential by breaking down information barriers; enterprises in highly market-oriented regions or cities with data trading platforms can use platform advantages to amplify the effect of disclosure on alleviating financing constraints.

At the government and regulatory level, institutional improvements are needed to guide the high-quality development of data asset information disclosure. On the one hand, based on the existing “Interim Provisions on the Accounting Treatment of Enterprise Data Resources,” detailed operational guidelines for data asset information disclosure should be developed, clarifying disclosure boundaries, format requirements, and key indicators, thereby reducing disclosure costs and improving information comparability and decision-making usefulness. On the other hand, a data asset information disclosure quality evaluation mechanism can be established, with stock exchanges or industry associations regularly issuing evaluation reports and selecting model enterprises, promoting best practices of substantive disclosure and guiding enterprises from “slogan-style mentions” to “substantive disclosure.” Meanwhile, regulatory agencies can use text analysis techniques to strengthen the identification of “impression management” behaviors, issuing warnings against empty disclosure and templated expressions to ensure the authenticity and effectiveness of information disclosure.

At the investor and supply chain partner level, data asset information disclosure should be fully utilized to improve value assessment and collaborative trust mechanisms. Investors can incorporate data asset information disclosure into their corporate value assessment systems, focusing on the richness and substantive content of disclosure as an important reference for judging firms’ long-term competitiveness and risk resistance capabilities. Upstream and downstream supply chain enterprises can use disclosure to establish a collaborative trust ecosystem: core enterprises actively disclose their data asset applications to convey operational transparency signals; small- and medium-sized supporting enterprises disclose their digital transformation progress to gain the trust and order support of core enterprises, jointly building an information-driven, multi-party collaborative supply chain resilience development pattern.

6. Discussion: Practical Barriers to Data Asset Information Disclosure and Corresponding Strategies

Although this study confirms the positive impact of data asset information disclosure on supply chain resilience, in practice, firms still face multiple barriers when disclosing data asset information. Identifying and overcoming these barriers is of great significance for fully realizing the enabling effect of data asset information disclosure.

Technical barriers to data asset recognition and measurement. Data assets have unique characteristics such as non-exhaustibility, non-rivalry, and value volatility. Their recognition boundaries, measurement bases, and valuation methods have not yet formed unified standards in accounting practice. Many firms lack the experience and technical capability to recognize data resources as assets, resulting in their inability to make standardized disclosures in financial statements even when they possess abundant data resources. To overcome this barrier, firms should strengthen their data governance capabilities, establish internal management processes for data asset inventory, classification, and valuation, and gradually develop a technical pathway for data assetization. Regulatory authorities could accelerate the issuance of operational guidelines for data asset recognition and measurement, providing industry-specific case examples to lower the technical threshold for firms.

Cost–benefit considerations of data asset information disclosure. High-quality data asset disclosure requires substantial investment of manpower and resources for data organization, value assessment, and text preparation. However, the benefits of disclosure—such as improved financing access and market recognition—often involve a time lag. For small- and medium-sized enterprises and firms with weaker disclosure capabilities, the short-term costs of disclosure may outweigh the expected benefits, suppressing their willingness to disclose. To alleviate this issue, governments could provide incentives such as fiscal subsidies or tax benefits to firms that engage in standardized disclosure, thereby reducing the disclosure costs for SMEs. At the same time, financial institutions could incorporate data asset disclosure quality into their credit approval and interest rate pricing models, making the economic benefits of disclosure more direct and perceptible.

The dilemma of balancing data security and trade secrets. Data assets often involve firms’ core business information, customer privacy, and technical secrets. Firms are concerned that detailed disclosure of data asset information may leak trade secrets and weaken competitive advantages. This security concern is a significant factor hindering in-depth disclosure. To balance disclosure and confidentiality, regulatory authorities should clarify the boundaries of data asset information disclosure, distinguishing between “mandatory disclosure” of core information and “confidential” sensitive information, and establish a tiered disclosure system. Firms could adopt methods such as desensitization and aggregated disclosure to convey the value signals of data assets while protecting core secrets.

The lack of unified disclosure standards and industry benchmarks. Currently, data asset information disclosure is still at a voluntary stage, lacking uniform format requirements and content standards. The disclosure content varies greatly across firms, reducing the comparability of information. At the same time, the absence of benchmark firms within industries leaves other firms without a reference model for disclosure. To address this, industry associations could take the lead in developing guidelines for data asset information disclosure, publishing case examples, and gradually promoting the formation of industry practices. Regulatory authorities could consider gradually transitioning from voluntary to mandatory disclosure when conditions mature, thereby standardizing disclosure requirements.

In summary, promoting data asset information disclosure requires collaborative efforts from firms, regulatory authorities, and market participants. Only by overcoming multiple barriers—technical, cost-related, security-related, and standardization-related—can the enabling potential of data asset information disclosure for supply chain resilience be truly unleashed.