Who Is Manipulating Corporate Wallets Amid the Ever-Changing Circumstances? Digital Clues, Information Truths and Risk Mysteries

Abstract

Digital transformation (DT) has emerged as a key strategic lever for enhancing firm resilience and competitiveness, yet its influence on non-productive investment behaviors, such as corporate financial investment, remains underexplored. Existing studies have largely focused on DT’s role in innovation and operational efficiency, leaving a significant gap in understanding how DT reshapes firms’ financial asset allocation. Drawing on a unique panel dataset of A-share main board-listed firms in China from 2011 to 2023, this study provides novel empirical evidence that DT significantly restrains financial investment, with pronounced heterogeneity across ownership types. More importantly, this paper uncovers a multi-layered mechanism: DT enhances the corporate information environment, which subsequently reduces financial investment. In addition, the analysis reveals a moderated mediation mechanism wherein economic uncertainty dampens the information-enhancing effect of DT. Unlike previous research that treats corporate risk-taking as a parallel mediator, this study identifies a sequential mediation pathway, where improved information environments suppress financial investment indirectly by influencing firms’ risk-taking behavior. These findings offer new theoretical insights into the financial implications of DT and contribute to the broader understanding of enterprise behavior in the context of digitalization and economic volatility.

1. Introduction

In the face of mounting international uncertainty and elevated global recessionary risks, trade frictions—most notably the series of tariff measures enacted by the Trump administration—have driven capital into safe-haven assets. Concurrently, corporate holdings of financial assets have surged, with non-financial firms’ global portfolios exceeding USD 10 trillion by the end of 2023—marking a 15 percent increase from 2021 levels. Although this accumulation largely reflects a prolonged period of historically low interest rates and heightened risk aversion, it has also given rise to greater liquidity risk and the crowding out of productive real investments. In response, Chinese firms have accelerated their digital transformation (DT) efforts to enhance operational efficiency, lower financing costs under uncertainty, and rebalance their investment portfolios more effectively [1]. Given that investment constitutes the core driver of corporate survival, expansion, and sustainable development, it is imperative to examine investment behavior within the DT context [2].

A review of the existing literature reveals that most research on the impact of DT on corporate investment has primarily concentrated on areas such as innovation investment, green investment, investment efficiency, and R&D [3,4,5], whereas its influence on corporate financial asset investment remains relatively underexplored and lacks clear theoretical grounding. Existing empirical evidence is both limited and contradictory: while some studies suggest that DT facilitates financial investment [6,7], others argue it inhibits such behavior [8,9]. These inconsistencies point to a lack of conceptual clarity and reveal an important research gap. Moreover, recent studies suggest that DT not only directly affects corporate investment decisions but also reshapes the firm’s information environment by enhancing transparency, accessibility, and external attention [10,11]. A high-quality information environment reduces information asymmetry and improves managerial decision-making, thereby influencing the allocation of financial resources [12,13]. Although the prior literature acknowledges the critical role of the information environment in corporate decision-making [14], little is known about how DT-induced improvements in the information environment may mediate its impact on corporate financial investment. These research gaps are particularly evident in the context of China’s main board-listed firms, which have received limited empirical attention in the existing literature [6,15]. This paper aims to address these gaps by asking the following: (1) Does DT suppress corporate financial investment? (2) If so, what are the underlying mechanisms—especially with respect to enhancements in the information environment—through which this suppression occurs?

To answer these questions, this study utilizes an original panel dataset of A-share main board-listed firms on the Shanghai and Shenzhen Stock Exchanges from 2011 to 2023. The empirical analysis reveals that DT significantly restrains financial investment activities. Mediation analysis further shows that DT improves the corporate information environment, which in turn exerts a negative influence on financial investment behavior. Notably, a further study identifies a moderated mediation structure: economic uncertainty weakens the influence of DT on the information environment but does not disrupt the established pathway from the improved information environment to investment choices. This suggests that while external shocks may hinder the realization of informational benefits, their downstream effects on financial investment decisions remain robust once informational improvements are realized. Furthermore, sequential mediation analysis reveals that corporate risk-taking acts as a downstream mediator rather than an independent parallel pathway. This layered mechanism uncovers a more nuanced process: DT enhances the information environment, which subsequently affects risk attitudes, and ultimately alters financial investment behavior. This insight diverges from existing models that treat mediators in isolation and highlights the dynamic, interlinked channels through which DT shapes corporate strategy. Lastly, the heterogeneity analysis shows that the DT–financial investment relationship varies significantly across ownership types, further enriching our contextual understanding of DT’s effects.

This study offers several contributions: First, it expands the DT–investment literature by shifting the analytical focus from commonly studied areas such as innovation, green, and R&D investments to financial investment, an area marked by theoretical uncertainty and empirical inconsistency [6,9,16]. This repositioning provides a novel lens through which to interpret DT’s broader impact on corporate asset allocation strategies, especially in the context of the neglected Chinese main board-listed companies.

Second, the study constructs an innovative composite index of the corporate information environment, integrating four dimensions: analyst coverage, research report frequency, media coverage breadth, and media sentiment balance. This multidimensional approach not only operationalizes the information environment more comprehensively than previous proxies but also facilitates a more precise investigation of its mediating role.

Third, the identification of economic uncertainty as a moderator in the first stage of the mediation pathway contributes to an emerging strand of research that examines the interaction between macro-level externalities and firm-level digital strategies. By clarifying how economic volatility shapes the benefits of DT, this study provides fresh empirical evidence to inform strategic planning in uncertain environments.

Fourth, by proposing and empirically validating a sequential mediation mechanism, this paper introduces a more dynamic model that explains how DT impacts corporate financial behavior through both informational and behavioral channels. Unlike traditional single or parallel mediation frameworks, this sequential model captures the cascading effects of digitalization on firm decision-making.

Collectively, these contributions enhance the theoretical understanding of how DT reshapes corporate financial strategies and offer methodologically robust insights that bridge gaps in both the DT and corporate finance literature.

2. Literature Review and Hypothesis Development

2.1. Literature Review

DT has emerged as a key driver reshaping corporate investment behavior across multiple dimensions. Existing studies primarily focus on the effects of DT on innovation-related investment, green investment, investment efficiency, and outward foreign direct investment (OFDI), offering a multifaceted understanding of how DT influences enterprise-level decision-making.

2.1.1. DT and Innovation Investment

A considerable body of research shows that DT positively impacts firms’ innovation investment. For instance, Li et al. (2022) [17] argue that DT encourages firms to enhance R&D activities and strengthen core technologies to align with market orientation. Using a sample of Chinese manufacturing firms listed on the A-share market, Wen et al. (2022) [18] found that DT significantly increases innovation investment, even after addressing endogeneity concerns. Yu et al. (2024) [19] also confirmed this relationship using data from 2012 to 2021. Moreover, DT is found to have differentiated effects: while it promotes exploitative innovation investment, it suppresses exploratory innovation investment [20]. Zhang et al. (2023) [21] reveal a positive association between DT and innovation investment volatility, reflecting the dynamic adjustment firms undertake in the face of digital disruption.

2.1.2. DT and Green Investment

DT has also been linked to enhanced corporate green investment. Ding et al. (2022) [22], based on data from highly polluting industries, find that digital finance significantly boosts green investment by expanding coverage and digital access. Similarly, Cao et al. (2022) [3] and Lin et al. (2024) [4] show that DT promotes green investment directly and indirectly by easing financing constraints. Other studies confirm that DT facilitates corporate investment in environmentally sustainable initiatives, including renewable energy projects [5,23]. Furthermore, Zhao and Yuan (2024) [24], Zhang et al. (2024) [25] and Tsybuliak et al. (2025) [26] demonstrate that DT reinforces the herd effect in green investments, implying strong peer influence in sustainability-oriented capital allocation.

2.1.3. DT and Investment Efficiency

Another stream of research explores how DT improves the efficiency of corporate capital deployment. Empirical evidence suggests that DT enhances labor investment efficiency by reducing both over- and under-investment [15,27]. Guo et al. (2025) [28] attribute these improvements to a reduction in information asymmetry facilitated by DT. Zhai et al. (2023) [29] further reveal that DT effectively restrains over-investment behavior among Chinese listed firms.

2.1.4. DT and Outward Foreign Direct Investment (OFDI)

Several studies investigate the impact of DT on firms’ internationalization strategies. Based on panel data from 2009 to 2022, multiple studies confirm that DT significantly promotes outward foreign direct investment [30,31,32,33]. These findings suggest that digital capabilities enhance firms’ ability to overcome cross-border information frictions and improve international market responsiveness.

2.1.5. DT and Sector-Specific Investment

Beyond the above areas, DT also influences sector-specific investment behavior. In the context of the digital economy, businesses can leverage consumer-side data to optimize production and supply chain strategies [34], accelerating supply chain investment [35]. DT also enables high-tech firms to expand into trans-regional markets [36]. In the service sector, particularly tourism, hospitality, and transportation, Hu et al. (2024) [37] find that DT significantly boosts investment activities, signaling its role in revitalizing traditional service industries.

2.1.6. DT and Financial Investment

While substantial attention has been given to the impact of DT on innovation, green initiatives, R&D, and investment efficiency, its effect on corporate financial asset allocation remains insufficiently addressed and lacks solid theoretical underpinnings. The limited body of empirical research offers conflicting results—some scholars report a positive association between DT and financial investment [6,7], while others find a negative relationship [8,9]. This divergence underscores a lack of theoretical coherence and highlights a notable gap in the literature.

2.1.7. DT and Information Environment

Furthermore, prior research has demonstrated that DT can significantly improve the quality of a firm’s information environment by enhancing transparency, increasing data accessibility, and attracting greater attention from external information intermediaries such as analysts and the media [10,11]. An improved information environment, in turn, helps to reduce information asymmetry, thereby enabling more accurate forecasting and more efficient allocation of financial resources [12,13]. These findings suggest that the information environment may serve as a key mechanism through which digital transformation influences corporate financial investment. Nevertheless, empirical studies that directly examine this mediating effect remain scarce. This study aims to fill this gap by investigating the role of the information environment as a mediator in the DT–investment relationship.

2.2. Hypothesis Development

According to the precautionary savings theory, firms may allocate more capital to financial assets in response to external uncertainties, as these assets are liquid and can help buffer operational risks [38,39]. However, DT changes firms’ financial and strategic behavior. By improving internal data systems and enabling better forecasting, DT reduces uncertainty and enhances operational efficiency, which in turn reduces the reliance on liquid financial assets for precautionary purposes [40,41].

Moreover, DT can expand access to external financing by enhancing information transparency and creditworthiness [42,43], thereby lowering financing constraints. When financing becomes more accessible and cost-effective, firms are less incentivized to hold financial assets merely for liquidity purposes. Instead, they may reallocate resources toward innovation or production-related investments [44,45].

In line with this reasoning, prior studies have shown that technology-driven shifts reduce speculative or non-core investments [8]. Therefore, it is reasonable to expect that digital transformation negatively influences financial investment.

H1. 

Corporate DT has a negative impact on financial investment.

The theory of information asymmetry suggests that unequal access to information among market participants can distort resource allocation and hinder optimal decision-making [46]. In corporate finance, such asymmetry often results from the information gap between corporate insiders and external stakeholders such as investors, analysts, and the media. This gap tends to increase financing costs, lead to suboptimal investment decisions, and intensify agency conflicts, ultimately reducing firm value [47].

A high-quality information environment, which features transparency, accessibility, and timely disclosure, can help mitigate these problems by improving the ability of external stakeholders to monitor and evaluate firm behavior [12,14]. Prior studies have shown that an improved information environment is associated with more efficient capital allocation and reduced reliance on short-term financial investments [48]. However, the relationship between the information environment and financial investment is not consistently established. Some studies argue that greater transparency reduces firms’ need to hold financial assets for precautionary purposes [12,48,49]. In contrast, other research suggests that excessive transparency may increase external pressure and scrutiny, prompting managers to increase holdings of liquid financial assets as a risk management strategy or a signal of financial stability [50,51].

Similarly, the existing literature presents mixed evidence on the relationship between DT and financial investment. Some scholars find that DT lowers the precautionary motive to hold financial assets by improving access to financing and reducing external uncertainty, which may lead to a decline in financial investment [8,9]. Conversely, other studies suggest that DT increases the flexibility of financial resource deployment. This may lead firms to allocate more resources to financial assets, particularly in response to strategic uncertainty or changes in market conditions [6,7].

These inconsistencies imply that the impact of DT on financial investment may vary depending on the mediating role of other organizational factors. One such factor is the firm’s information environment, which DT can improve through real-time data exchange, automated disclosure processes, and enhanced data accessibility [52]. Therefore, the information environment may serve as a key mechanism through which DT influences corporate financial investment. Based on this reasoning, the following hypothesis is proposed:

H2. 

The information environment mediates the relationship between DT and corporate financial investment.

3. Research Method

3.1. Sample Selection

This study selects A-share companies listed on the main boards of the Shanghai and Shenzhen Stock Exchanges from 2011 to 2023 as the research sample. The data are obtained from the China Stock Market and Accounting Research Database (CSMAR). To ensure data reliability and consistency, firms in the financial and insurance industries are excluded, as well as those labeled as ST or PT. Additional screening removes companies with discontinued operations, non-standard or negative audit reports, ongoing litigation cases, or records of regulatory sanctions or disciplinary actions. Observations lacking essential variables are also omitted. After applying these criteria, the final dataset consists of 6896 firm-year observations.

3.2. Model Specification

Following Heese et al. (2021) [53], Kong et al. (2020) [54], and Li et al. (2025) [55], this paper constructs models to empirically test the first two hypotheses.

$${\mathrm{F}\mathrm{I}}_{\mathrm{i}\mathrm{t}}={\alpha }_0+{\alpha }_1{\mathrm{D}\mathrm{T}}_{\mathrm{i}\mathrm{t}}+\sum \alpha {\mathrm{X}}_{\mathrm{i}\mathrm{t}}+\sum \lambda +\sum \mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+{\epsilon }_{\mathrm{i}\mathrm{t}}$$

(1)

$${\mathrm{I}\mathrm{E}}_{\mathrm{i}\mathrm{t}}={\beta }_0+{\beta }_1{\mathrm{D}\mathrm{T}}_{\mathrm{i}\mathrm{t}}+\sum \beta X_{it}+\sum \lambda +\sum \mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+{\epsilon }_{\mathrm{i}\mathrm{t}}$$

(2)

$${\mathrm{F}\mathrm{I}}_{\mathrm{i}\mathrm{t}}={\gamma }_0+{\gamma }_1{\mathrm{D}\mathrm{T}}_{\mathrm{i}\mathrm{t}}+{\gamma }_2{\mathrm{I}\mathrm{E}}_{\mathrm{i}\mathrm{t}}+\sum c{X}_{it}+\sum \lambda +\sum \mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+{\epsilon }_{\mathrm{i}\mathrm{t}}$$

(3)

The first model relies on several key assumptions. First, it assumes strict exogeneity, i.e., the explanatory variables are uncorrelated with the error term across all time periods. Second, it presumes that unobserved heterogeneity is time-invariant and fully absorbed by the fixed effects. Third, it requires sufficient within-firm variation in the key independent variable (DT) to identify its effect.

In the above models, FI denotes the dependent variable, measuring corporate financial investment. IE serves as the mediating variable, reflecting the quality of the information environment. X is a vector of control variables. ∑λ and ∑Year represent firm and year fixed effects, respectively. ε_it is the random error term.

3.3. Variable Measurement

Following the methodological framework of Chen and Xu (2022) [56], this study begins by identifying the core technologies underlying DT. To assess the degree of digitalization at the firm level, it employs a logarithmic transformation of keyword frequency data as a proxy measurement.

Based on the methodology proposed by Duchin et al. (2017) [57], this paper measures financial investment using the proportion of financial assets relative to total assets. The scope of financial assets encompasses trading securities, derivative instruments, net loans provided, available-for-sale securities, held-to-maturity investments, investment properties, and other related financial items.

Building on prior studies [48,58], this study constructs a composite index that comprehensively captures the quality of a firm’s information environment. Specifically, the index is measured based on the following four dimensions:

Analyst coverage, measured by the number of financial analysts covering the firm, reflects the degree of attention from professional intermediaries and the strength of external monitoring.

Number of analyst reports captures the richness of professional information and reflects the depth of market analysis regarding the firm’s value and risk.

Breadth of media coverage, measured by the volume of both online and print media reports, indicates the scope of public information dissemination and the extent of media attention.

Balance of media sentiment, calculated as (number of positive reports−number of negative reports) divided by the total number of reports, separately for online and print media, is used to assess the neutrality or balance of tone in the media.

To ensure comparability across dimensions, the four core indicators are standardized and then averaged to construct the final composite index of the information environment. A higher index value indicates more abundant information sources, broader market attention, and more balanced external evaluations, thus reflecting a higher-quality information environment with lower levels of information asymmetry.

Regarding control variables, this study accounts for firm-specific factors known to influence investment decisions, as referenced in the prior literature [17,59]. Specifically, the control variables include the leverage ratio, operating efficiency, return on assets (ROA), ownership concentration (top 10 shareholders), market capitalization, managerial ownership, and total assets (

Table A1. Variable definitions.

Label	Definition
Digital Transformation	The level of enterprise digitalization is measured using the logarithmic transformation of word frequency statistics.
Financial Investment	Calculated as financial assets scaled by total assets
Leverage Ratio	Calculated as the company’s asset–liability ratio.
Operating Efficiency	Calculated from the company’s inventory turnover rate.
Return on Assets	Calculated from the company’s profit margin.
Top 10 Shareholders	Calculated based on the shareholding size of the top ten shareholders.
Management Ownership	Calculated based on the size of management shareholdings.
Market Capitalization	Calculated from the size of circulating shares in the market
Total Assets	Calculated from the company’s asset size level
Information Environment	Standardizes the four core indicators and computes their average to form the final composite index
Economic Uncertainty	Uses the Baker index as a measure of economic uncertainty
Risk-taking	The volatility of ROA in the observation period is used to measure the risk-taking level of enterprises.

).

3.4. Summary Statistics

Table 1. Summary statistics.

	Mean	SD	Min	P50	Max
Financial Investment	0.2125	0.1347	0.0012	0.1817	0.9323
DT	0.4300	0.0990	0.2182	0.4284	0.8004
Information Environment	0.0208	0.7777	−1.6766	−0.0453	3.0189
EU	1.4658	0.1785	1.2503	1.4361	1.8523
Risk-Taking	0.0405	0.0862	−0.8469	0.0438	0.5046
Leverage Ratio	0.4510	0.1985	0.0098	0.4496	3.9191
Return on Assets	0.0306	0.5690	−33.8802	0.0581	8.1766
Operating Efficiency	0.6258	0.4665	0.0010	0.5353	7.3887
Top 10 Shareholders	55.7640	15.0802	8.9700	55.8400	100.9700
Management Ownership	1.7027	0.4571	1.0000	2.0000	2.0000
Market Capitalization	10.5683	0.8933	4.7185	10.5401	14.1784
Total Assets	22.6359	1.4271	17.8787	22.4640	28.6969
Observations	6896	6896	6896	6896	6896

presents summary statistics for the main variables employed in the analysis, including central tendency indicators such as the mean and median, as well as measures of dispersion like the standard deviation and minimum and maximum values. These statistics help illustrate the distributional features of the dataset and offer initial insights into the variability and characteristics of the sample, laying the groundwork for the following regression analysis.

4. Results

4.1. Impact of DT on Financial Investment

As shown in

Table 2. The impact of DT on financial investment: fixed-effects analysis.

	(1)	(2)	(3)	(4)
	Two-Way FE	Three-Way FE(Province)	Three-Way FE(Industry)	Excluding 2015
DT	−0.000137 **	−0.000142 **	−0.000140 **	−0.000122 **
	(−2.34)	(−2.42)	(−2.41)	(−2.03)
Leverage Ratio	−0.0147 *	−0.0143 *	−0.00966	−0.0136 *
	(−1.84)	(−1.78)	(−1.21)	(−1.66)
Return on Assets	0.00156	0.00157	0.00178	0.00189
	(1.16)	(1.16)	(1.34)	(1.44)
Operating Efficiency	−0.00873 **	−0.00867 **	−0.00863 **	−0.00956 **
	(−2.39)	(−2.37)	(−2.28)	(−2.55)
Top 10 Shareholders	−0.000568 ***	−0.000571 ***	−0.000569 ***	−0.000511 ***
	(−4.02)	(−4.00)	(−4.01)	(−3.42)
Management Ownership	0.0000845	−0.000245	−0.000374	0.00270
	(0.03)	(−0.09)	(−0.15)	(1.02)
Market Capitalization	−0.00377 *	−0.00367	−0.00442 *	−0.00315
	(−1.65)	(−1.60)	(−1.93)	(−1.32)
Total Assets	−0.00729 ***	−0.00734 ***	−0.00709 ***	−0.00552 **
	(−2.98)	(−2.99)	(−2.86)	(−2.14)
Constant	0.293 ***	0.293 ***	0.293 ***	0.239 ***
	(5.46)	(5.46)	(5.36)	(4.21)
Adj. R-squared	0.656	0.656	0.664	0.684
Observations	6093	6093	6093	5671

t statistics in parentheses. *, **, and *** denote significance at the 10%, 5%, and 1% levels. * p < 0.1, ** p < 0.05, *** p < 0.01.

, in the baseline model with firm and year fixed effects (Column 1), the coefficient of DT is −0.000137 and statistically significant at the 5% level. This result provides empirical support for Hypothesis 1, suggesting that DT is associated with a reduction in corporate financial investment. One possible explanation for this finding lies in the resource reallocation effect of DT. As firms implement digital technologies, they may reorient resources away from short-term financial investments and toward long-term strategic and operational initiatives, such as innovation and digital infrastructure. This aligns with Zhang et al. (2023) [9] and Jin & Xie (2024) [8], who find that DT mitigates over-financialization by curbing excessive holdings of financial assets, implying a shift back to core business activities.

Moreover, DT may improve internal efficiency and productivity, reducing the reliance on financial asset returns as a buffer against uncertainty. Digital tools such as AI, cloud computing, and big data analytics can enhance operational transparency and decision-making, thus decreasing the perceived need for defensive financial investments [60,61]. Additionally, institutional improvements triggered by DT—such as better governance structures and enhanced stakeholder monitoring—may further disincentivize speculative or opportunistic financial investment behaviors [62,63]. Overall, this result indicates that DT not only alters firms’ technological capabilities but also reshapes their investment logic through deeper structural and institutional mechanisms.

4.2. Robustness and Endogeneity Tests

To enhance robustness, this study incorporates firm-year, province, and industry fixed effects in the regression analysis. As shown in Table 2, DT consistently exhibits a significant negative effect on financial investment at the 5% level across all models, confirming the stability of the results. Moreover, excluding the impact of the 2015 stock market crash in China, the results remain robust.

Table 3. Endogeneity tests for financial investment.

	(1)	(2)	(3)	(4)
	PSM	GMM	Random Indep	Random Dep
DT	−0.000 *	−0.001 ***		−0.000
	(−1.94)	(−3.60)		(−1.35)
Leverage Ratio	−0.093 ***	−0.020 ***	−0.015 *	−0.024
	(−13.31)	(−3.13)	(−1.90)	(−1.59)
Return on Assets	0.000	−0.000	0.002	−0.001
	(0.62)	(−0.44)	(1.12)	(−0.26)
Operating Efficiency	−0.010 ***	−0.002	−0.009 **	−0.005
	(−4.68)	(−0.79)	(−2.39)	(−0.66)
Top 10 Shareholders	−0.000	0.000	−0.001 ***	0.000
	(−0.27)	(0.67)	(−3.98)	(0.49)
Management Ownership	−0.002	0.000	−0.000	0.001
	(−0.87)	(0.11)	(−0.03)	(0.20)
Market Capitalization	0.001	−0.002 **	−0.004 *	0.001
	(0.41)	(−1.98)	(−1.74)	(0.21)
Total Assets	0.002	0.000	−0.008 ***	−0.002
	(1.36)	(0.32)	(−3.17)	(−0.46)
L. Financial Investment		0.729 ***
		(19.53)
Randomized DT			−0.000
			(−1.51)
Constant	0.050 **	0.030 **	0.304 ***	0.091
	(2.40)	(2.39)	(5.71)	(0.90)
Observations	5173.000		6093.000	6093.000
Adj. R-squared	0.049		0.656	−0.012

Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels.

reports endogeneity tests using PSM, System GMM, and randomization methods. The PSM results show a significant negative effect, supporting H1 and indicating that DT may curb financial speculation. GMM’s overidentification test (Sargan p = 0.21) and instrument strength (Cragg-Donald F = 132.6) both meet the statistical requirements, confirming model specification rationality. Randomization tests yield near-zero, insignificant coefficients, suggesting the observed relationship is unlikely due to chance, effectively ruling out spurious correlations.

4.3. Mediation Analysis

Table 4. Mediation analysis: information environment on financial investment.

	(1)	(2)	(3)
	FI	Information Environment	FI
DT	−0.0001 **	0.001 ***	0.000
	(−2.34)	(2.67)	(0.47)
Leverage Ratio	−0.015 *	−0.196 ***	−0.023 **
	(−1.84)	(−2.74)	(−2.14)
Return on Assets	0.002	0.000	0.000 **
	(1.16)	(0.09)	(2.00)
Operating Efficiency	−0.009 **	0.158 ***	−0.003
	(−2.39)	(4.62)	(−0.51)
Top 10 Shareholders	−0.001 ***	0.005 ***	−0.001 ***
	(−4.02)	(3.43)	(−3.41)
Management Ownership	0.000	−0.010	−0.011 ***
	(0.03)	(−0.42)	(−3.01)
Market Cap	−0.004 *	0.028	−0.003
	(−1.65)	(1.30)	(−1.06)
Total Assets	−0.007 ***	0.355 ***	−0.011 ***
	(−2.98)	(14.94)	(−3.01)
Information Environment			−0.007 **
			(−2.31)
Constant	0.293 ***	−8.579 ***	0.400 ***
	(5.46)	(−16.31)	(4.88)
Adj. R-squared	0.656	0.821	0.655
F-test	7.027	41.681	7.355

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

shows that DT positively impacts the corporate information environment, which in turn negatively affects financial investment, suggesting a mediating role. While the total effect of DT on financial investment is significantly negative, the direct effect is positive but insignificant, indicating a “suppression mediation” pattern. DT may slightly increase financial investment directly, but its improvement of the information environment generates a stronger indirect effect that suppresses investment. Overall, DT indirectly suppresses corporate financial investment by enhancing the quality of the information environment, potentially guiding firms to adjust their investment structure. This may be because a better information environment reduces the uncertainty of real investment or increases market attention on a firm’s real business performance [9]. Our findings also support the information asymmetry theory, which posits that reduced information asymmetry lowers managerial discretion and curbs overinvestment [64,65]. In this context, DT, by improving disclosure quality and reducing uncertainty, may prompt managers to reconsider marginal or speculative investments [10,66,67].

5. Further Study

After analyzing how the information environment mediates the impact of DT on financial investment, this study further investigates how macro-level environmental factors influence this mediation mechanism. Corporate decision-making is embedded into a broader macro environment, in which economic uncertainty (EU), as a critical external factor, may shape both the manner and extent of the influence exerted by DT and the information environment. Economic uncertainty not only directly impacts corporate decision-making preferences but may also alter the relationship between strategic initiatives and outcomes.

Within the framework of information asymmetry theory, heightened economic uncertainty increases the difficulty of information collection and processing. This may either weaken or strengthen the pathway through which DT affects investment behavior via the information environment. Therefore, incorporating economic uncertainty into the analytical framework—as a moderating variable in the mediation model—helps deepen our understanding of the boundary conditions under which DT influences investment behavior.

Accordingly, this study further examines how the EU (specifically, the external macro-level uncertainty faced by firms) moderates the mediating pathway through which DT influences corporate investment behavior via the information environment. Constructing a moderated mediation model is not only methodologically sound but also essential for understanding how DT affects corporate investment behavior through the information environment under varying economic circumstances.

5.1. Moderated Mediation

Table 5. Economic uncertainty moderates information environment path.

	(1)	(2)	(3)
	Step 1	Step 2	Step 3
DT	−0.0001 **	0.010 ***	0.000
	(−2.34)	(5.47)	(0.78)
Leverage Ratio	−0.015 *	−0.157 **	−0.023 **
	(−1.84)	(−2.21)	(−2.15)
Return on Assets	0.002	0.000	0.000 **
	(1.16)	(0.13)	(2.00)
Operating Efficiency	−0.009 **	0.165 ***	−0.003
	(−2.39)	(4.86)	(−0.51)
Top 10 Shareholders	−0.001 ***	0.005 ***	−0.001 ***
	(−4.02)	(3.72)	(−3.41)
Management Ownership	0.000	−0.006	−0.011 ***
	(0.03)	(−0.24)	(−3.07)
Market Cap	−0.004 *	0.033	−0.003
	(−1.65)	(1.53)	(−1.07)
Total Assets	−0.007 ***	0.342 ***	−0.011 ***
	(−2.98)	(14.42)	(−3.02)
EU		0.000	0.000
		(.)	(.)
DT # EU		−0.008 **
		(−2.04)
Information Environment			−0.007 **
			(−2.28)
Information Environment # EU			−0.001
			(−0.11)
Constant	0.293 ***	−8.823 ***	0.402 ***
	(5.46)	(−17.02)	(4.89)
Observations	6093.000	3003.000	3003.000
Adj. R-squared	0.656	0.823	0.655
F-test	7.027	41.043	6.105

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

’s Column (2) analyzes the impact of DT on the mediating variable and its interaction with economic uncertainty. Column (3) displays the results of the full model, which includes all main effects and moderating variables. Overall, the results reveal a complex moderated mediation mechanism: economic uncertainty plays a moderating role mainly in the first stage, weakening the positive effect of DT on the information environment but not altering the fundamental mechanism through which the information environment influences financial investment decisions. This finding implies that in highly uncertain times, the information advantages produced by DT may be limited. However, once the information environment is established, its influence on investment decisions remains relatively stable and less susceptible to external economic fluctuations.

5.2. Mediation Analysis: Risk-Taking on Investments

In analyzing the mechanism through which DT affects corporate investment behavior, introducing corporate risk-taking as an additional mediating variable holds important theoretical and practical significance. Corporate investment decisions are essentially a process of balancing risk and return [68]. DT not only enhances a firm’s information processing capabilities but may also reshape its risk preference structure, thereby influencing investment behavior [69]. However, risk-taking is not a completely independent mediating variable; its formation and evolution are highly dependent on improvements in the information environment [70]. DT first improves the information environment by increasing transparency and processing capacity and then indirectly influences risk-taking by strengthening risk perception and management capabilities, which ultimately affects investment decisions. This sequential mediation mechanism better reflects the reality of corporate decision-making compared to a parallel mediation structure, and it also explains why the independent effect of risk-taking may not be significant in empirical studies. Therefore, due to the close interaction between the information environment and risk-taking, both should be considered together in mediation analysis to more comprehensively reveal the pathways through which DT influences corporate investment behavior.

In light of this, the present study begins with an analysis of the mediating effect of corporate risk-taking. The empirical results concerning financial investment are presented in the following table:

From the empirical results (

Table 6. Mediation analysis: risk-taking on financial investment.

	(1)	(2)	(3)
	Total Effect	Path a	Direct Effect
DT	−0.001 **	0.000	0.000
	(−2.34)	(0.72)	(0.53)
Leverage Ratio	−0.015 *	−0.232 ***	−0.025 **
	(−1.84)	(−14.84)	(−2.00)
Return on Assets	0.002	0.000	−0.000
	(1.16)	(1.19)	(−0.55)
Operating Efficiency	−0.009 **	0.062 ***	−0.008
	(−2.39)	(9.19)	(−1.51)
Top 10 Shareholders	−0.001 ***	0.000	−0.001 ***
	(−4.02)	(0.41)	(−3.75)
Management Ownership	0.000	0.006	−0.011 ***
	(0.03)	(1.27)	(−3.14)
Market Cap	−0.004 *	−0.015 ***	−0.002
	(−1.65)	(−3.76)	(−0.55)
Total Assets	−0.007 ***	0.025 ***	−0.015 ***
	(−2.98)	(5.49)	(−4.38)
Risk-Taking			0.047 ***
			(2.93)
Constant	0.293 ***	−0.308 ***	0.486 ***
	(5.46)	(−3.12)	(6.34)
Observations	6093.000	3060.000	3060.000
Adj. R-squared	0.656	0.454	0.658
F-test	7.027	40.693	8.831

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

), DT shows a significant negative effect on financial investment in the total effect model, while its impact on risk-taking is not significant. However, risk-taking itself has a positive effect on financial investment. This pattern suggests that risk-taking may play a complex mediating role in the relationship between DT and financial investment.

Therefore, this study needs to further consider the mediating role of risk-taking as a derived mediating effect. In terms of the transmission mechanism, corporate risk-taking often stems from changes in the external environment. Thus, this paper introduces a stepwise mediation analysis, incorporating the information environment as a foundational mediator and then expanding to include the mediation of the mediator—risk-taking. The detailed analysis results are presented in the following table:

The results of this sequential mediation analysis suggest that corporate risk-taking primarily functions as a downstream mediator of the information environment in the process by which DT influences investment behavior, rather than as an independent parallel mediator (

Table 7. Sequential mediation analysis on financial investment.

	(1)	(2)	(3)	(4)
	Total Effect	Path a1	Path a2, d21	Direct Effect b1, b2
DT	−0.0001 **	0.001 ***	0.000	0.000
	(−2.34)	(2.67)	(0.24)	(0.55)
Leverage Ratio	−0.015 *	−0.196 ***	−0.090 ***	−0.025 *
	(−1.84)	(−2.74)	(−7.79)	(−1.91)
Return on Assets	0.002	0.000	0.192 ***	−0.002
	(1.16)	(0.09)	(47.38)	(−0.28)
Operating Efficiency	−0.009 **	0.158 ***	0.035 ***	−0.006
	(−2.39)	(4.62)	(7.16)	(−1.16)
Top 10 Shareholders	−0.001 ***	0.005 ***	0.000	−0.001 ***
	(−4.02)	(3.43)	(0.53)	(−3.72)
Management Ownership	0.000	−0.010	0.004	−0.011 ***
	(0.03)	(−0.42)	(1.24)	(−3.03)
Market Cap	−0.004 *	0.028	−0.008 ***	−0.002
	(−1.65)	(1.30)	(−2.91)	(−0.52)
Total Assets	−0.007 ***	0.355 ***	−0.004	−0.012 ***
	(−2.98)	(14.94)	(−1.09)	(−3.24)
Information Environment			0.021 ***	−0.008 **
			(7.45)	(−2.48)
Risk-Taking				0.056 **
				(2.40)
Constant	0.293 ***	−8.579 ***	0.215 ***	0.411 ***
	(5.46)	(−16.31)	(2.88)	(4.96)
Observations	6093.000	3003.000	2941.000	2941.000
Adj. R-squared	0.656	0.821	0.732	0.645
F-test	7.027	41.681	336.244	8.058

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

). This “sequential mediation” mechanism provides a more complete and nuanced explanation of how DT, by altering the firm’s information environment and risk decisions, ultimately affects investment behavior. It offers a more comprehensive theoretical framework for understanding corporate investment behavior in the context of the digital economy.

5.3. Analysis of Heterogeneity

Table 8. Heterogeneous effects by ownership type.

	(1)	(2)	(3)	(4)
	State-Owned	Foreign	Private	Others
Digital Transformation	−0.0000652	0.0000497	−0.000636 **	−0.000186
	(−0.91)	(0.52)	(−2.52)	(−0.46)
Leverage Ratio	−0.0180	−0.0146	0.0131	−0.0281
	(−0.89)	(−0.99)	(0.19)	(−0.37)
Return on Assets	0.00395	0.00190	0.0430 *	0.00271
	(0.32)	(1.58)	(1.82)	(0.14)
Operating Efficiency	−0.00531	−0.0146 *	−0.0397	0.00492
	(−0.75)	(−1.94)	(−1.04)	(0.56)
Top 10 Shareholders	−0.000768 **	−0.000434	0.000149	0.0000653
	(−2.40)	(−1.43)	(0.16)	(0.05)
Management Ownership	0.00270	−0.00383	0.0317	0.00930
	(0.46)	(−0.81)	(1.45)	(0.79)
Market Capitalization	−0.00541	−0.00258	−0.00659	−0.00683
	(−1.18)	(−0.60)	(−0.33)	(−0.52)
Total Assets	−0.00103	−0.0131 **	0.0285	0.00718
	(−0.19)	(−2.55)	(1.16)	(0.26)
Constant	0.172	0.409 ***	−0.532	−0.0324
	(1.37)	(3.29)	(−1.16)	(−0.06)
Adj. R-squared	0.714	0.608	0.764	0.947
Observations	2082	3635	165	113

t statistics in parentheses. Notes: two-way fixed effects (firm and year) included. * p < 0.1, ** p < 0.05, *** p < 0.01.

shows that the impact of DT on financial investment exhibits significant heterogeneity across ownership types. Among all types of enterprises, DT generally shows an inhibitory effect: private enterprises have a coefficient of −0.000636 (5% significance), state-owned enterprises show −0.000652 (not significant), and other types of enterprises show −0.000186 (not significant). Notably, only foreign-invested enterprises show a positive coefficient of 0.000497 (though not significant). This directional difference suggests that enterprises with higher degrees of marketization and internationalization may be better at converting digital technology into investment opportunities. In contrast, the negative coefficient of state-owned enterprises reflects how institutional constraints may limit the capital allocation efficiency of DT.

6. Conclusions

This study finds that DT significantly restrains corporate financial investment, with the effect varying significantly across different ownership types. The information environment plays a mediating role in this process. Economic uncertainty weakens the positive impact of DT on the information environment but does not alter the fundamental mechanism through which the information environment affects financial investment decisions. Additionally, the sequential mediation analysis indicates that corporate risk-taking primarily functions as a downstream mediator in the process through which DT influences financial investment, rather than serving as an independent parallel mediator.

This study offers several practical and policy-relevant implications grounded in the empirical findings:

First, firms undertaking DT should be aware that improvements in the information environment—though typically considered beneficial—may inadvertently suppress financial investment activities. Our findings demonstrate that DT significantly enhances the corporate information environment, which in turn exerts a negative effect on financial investment. This effect is especially relevant in contexts of high economic uncertainty, where the positive influence of DT on the information environment is weakened. Consequently, firms should avoid over-relying on informational advantages and instead adopt flexible investment strategies tailored to their industry characteristics and prevailing market conditions. This helps mitigate the potential unintended consequence of under-investment in financial assets during the DT process.

Second, the study identifies economic uncertainty as a significant moderator that dampens the impact of DT on the information environment. Given the rising global volatility in trade and macroeconomic indicators, firms must monitor changes in external uncertainty and assess how such shifts may constrain the informational benefits of DT and, by extension, investment behavior. Policymakers, in turn, should work toward fostering a more stable and transparent macroeconomic environment. Enhancing regulatory consistency and promoting information disclosure standards can strengthen firms’ ability to realize the full benefits of DT even under uncertain conditions, thereby improving their investment confidence.

Third, the sequential mediation analysis reveals that DT enhances the information environment, which subsequently shapes firms’ risk-taking behavior, ultimately influencing financial investment decisions. This implies that firms should not only focus on technology adoption but also invest in building robust information systems and upgrading data analytics capabilities. Improving risk identification and management processes can enable firms to better navigate the indirect impacts of DT on investment behavior, ensuring alignment between strategic transformation and financial decision-making.

Finally, the heterogeneity analysis shows that the relationship between DT and financial investment varies significantly across ownership types. Policymakers should therefore design more targeted and differentiated support policies that consider ownership-specific dynamics. In addition, encouraging investment in digital infrastructure—particularly among state-owned and small-to-medium enterprises—can improve the overall quality of the information environment, strengthen risk governance structures, and enhance the long-term effectiveness of DT initiatives in promoting sustainable corporate development.

Limitations and Future Research Directions

However, this study may have several limitations. While fixed-effects models and GMM estimators partially address unobserved heterogeneity and dynamic endogeneity, they do not fully resolve potential biases stemming from measurement error in DT proxies or unobserved time-varying firm-level shocks. Moreover, the PSM approach relies solely on observable covariates, which may fail to capture selection on unobservables.

Future studies could extend this research by leveraging quasi-experimental designs, such as difference-in-differences (DID) approaches with exogenous policy shocks (e.g., digital infrastructure rollouts or digital subsidy programs), to better establish causality. Moreover, future studies may enhance precision and depth by incorporating alternative data sources—such as digital trace data or high-frequency operational datasets—that capture more granular and real-time aspects of firms’ digital engagement. Additionally, the analysis of underlying mechanisms lacks depth. Future research should further explore the underlying mechanisms through which DT affects financial investment in order to deepen and enrich our understanding of this relationship.