Next Article in Journal
Concrete or Abstract? The Impact of Green Advertising Appeals and Information Framing on Consumer Responses
Previous Article in Journal
Customer-Centric Decision-Making with XAI and Counterfactual Explanations for Churn Mitigation
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Enterprise Digital Transformation Drivers: Market or Government? A Case Study from China

School of Economics and Statistics, Guangzhou University, Guangzhou 510006, China
*
Author to whom correspondence should be addressed.
J. Theor. Appl. Electron. Commer. Res. 2025, 20(2), 131; https://doi.org/10.3390/jtaer20020131
Submission received: 25 March 2025 / Revised: 1 May 2025 / Accepted: 27 May 2025 / Published: 3 June 2025
(This article belongs to the Section Digital Business Organization)

Abstract

:
The relative dominance of government and market forces in enterprise digital transformation remains underexplored. This paper aims to provide new insights into this topic. Using data from China’s A-share-listed companies (2007–2023), we test the short- and long-term impacts of government subsidies and market strength on enterprise digital transformation, quantify their relative contributions, explore substitution effects, investigate synergistic interactions, and examine heterogeneous impacts across different enterprise ownership types and life cycle stages, using the fixed effects regression model, exponential smoothing, interaction term, coupling coordination model, and group regression. The results indicate that both government subsidies and market forces drive digital transformation in enterprises, with government subsidies having a slightly stronger effect. There is a substitutive relationship between the two, and their synergy significantly promotes digital transformation. However, their impact varies across different types of enterprises. Stronger market forces do not always lead to greater transformation; in fact, for non-state-owned enterprises, market strength can hinder digital transformation. Similarly, government subsidies do not consistently promote digital transformation. Their effect is less pronounced for growth and maturity-stage enterprises, while declining enterprises are more motivated to pursue digital transformation to benefit from subsidies.

1. Introduction

In the early stages of digital transformation, businesses often face high costs and limited technical capabilities [1]. The government plays a crucial role in supporting this transition by utilizing policy tools and investing in infrastructure. These efforts help share risks, reduce costs, and guide the digitalization process [2,3,4]. For SMEs, the government provides fiscal subsidies and special loans to lower digitalization expenses. Key infrastructure developments, such as the industrial internet and 5G, also require strong government leadership, including setting goals for 5G base stations and promoting co-building initiatives. Furthermore, the government helps direct companies’ focus on strategic areas through planning and demonstration projects [5,6,7]. As government support lowers transformation barriers, market forces drive deeper changes through digital adoption and competition, with e-commerce as a key catalyst. With consumers increasingly turning to digital platforms, businesses must adapt to an internet-driven market by integrating online and offline operations, streamlining processes, enhancing customer experience, and reducing costs [8,9,10]. Simultaneously, rising competition, especially in manufacturing, retail, and services, compels companies to leverage digital tools to improve supply chain efficiency, product quality, and responsiveness, securing a competitive advantage [11,12].
Current literature on enterprise digital transformation is mostly subject-oriented, with a focus on the driving forces of government and market forces. Well known for promoting digital transformation, the government assists enterprises with policy frameworks and strategic direction, financial assistance and resource allocation, and innovation environments and technology promotion.
First, governments facilitate digital transformation via the creation of robust policy and regulatory frameworks. Tangwaragorn et al. [13] illustrated the point by revealing how strategic policy, legal environment, and quality of regulation cumulatively spearhead enterprise digital transformation. Likewise, Chuc et al. [14] illustrated how digital economy infrastructures and taxation reforms contribute, emphasizing the importance of policy and strategy in driving digital transformation.
Furthermore, the allocation of resources and financial support is essential, as numerous studies indicate that subsidies, tax reductions, and pilot policies effectively promote digitalization. Using firm-level data from China, Wang et al. [2] illustrated that subsidies and tax reductions significantly promote digitalization. Liu et al. [15] and Zhang et al. [16] further determined that national pilot policies, such as the innovative city and “specialization and new” policies, drive digital transformation by inducing improved internal control, higher R&D, and relaxed funding curbs. In resource-constrained settings, Senyo et al. [5] suggested that governments can also support digital transformation by using lean innovation and bundling resources, highlighting the important role of financial support in reducing digital transformation costs, particularly for SMEs.
Finally, governments play an important role in fostering innovation ecosystems and promoting technological advancement, which are key driving forces for digital adoption. Chuc et al. [14] and Mai et al. [17] reported that government policy in scientific and technological advancement positively impacts SME innovation and digitalization across borders, whereas Lutfi et al. [18] verified that government interventions are critical in technology uptake among Jordanian SMEs. Such cross-context analyses cumulatively validate that state interventions—running from policy creation, fiscal encouragement, and innovation hubs—represent a universal catalyst for enterprise digital transformation.
On the contrary, existing literature emphasizes the intricate role of market forces in enterprise digital transformation, primarily investigating mechanisms from the following three dimensions: economic conditions, competitive pressure, and customer and environmental demand. Economic conditions are viewed as the core, providing the required resources and market demand. Tangwaragorn et al. [13], Peng et al. [19], and Guo et al. [20] emphasized the role of the economic environment in rendering digital transformation favorable, while Nikopoulou et al. [21] showed that firms with more robust financial abilities are more likely to invest in digital technologies; the authors also recognized resource availability as a determinant of success.
Intense competition in the market drives companies to push competitiveness by embracing digital transformation. Competition pressure drives enterprises to adopt digital transformation in a complex way; Fan and Xu [11] identified a U-shaped relationship between digital transformation and product market competition, noting that the intention to adopt digital strategies first drops but increases as competition intensifies.
Customer needs and environmental pressures further drive digital adoption. Haffke et al. [22] noted that growing demands for personalization and responsiveness drive enterprises to provide these digitally. Likewise, Lokuge et al. [23] and Luo et al. [24] discovered that external environmental pressures such as compliance costs and regulatory time maximally trigger digital transformation, with customer expectations and business environments playing a role.
Previous research has provided valuable insights into the roles of government and market in enterprise digital transformation, yet three critical limitations persist in elucidating the driving mechanisms. First, existing studies primarily focus on the individual impacts of government and market forces on enterprise digitalization, leaving the relative power of each force unclear. For example, regarding the government’s role, some scholars have explored how government attention to digitalization guides transformation [3,7], while others have examined the influence of digital economy infrastructure and fiscal policy [14,25]. Concerning the market’s role, researchers have examined factors such as external environmental pressures [23], business environments [24], product market competition [11], and demands for personalization and responsiveness [22]. Second, much of the literature treats the government and market roles as separate entities, neglecting their potential dynamic interaction. Many studies focus on the isolated impact of either government or market factors in their models, as observed in the works of Peng et al. [19], Li and Yue [7], and others. Third, existing studies often rely on static cross-sectional analyses, overlooking the cumulative temporal effects of these driving forces. Most empirical models are constructed using variables from the current period, as illustrated in the studies of Peng et al. [3], Li and Yue [7], and Luo et al. [24], which fail to account for the evolving nature of digital transformation over time. In summary, the following gaps need to be addressed when exploring the roles of government and market forces in enterprise digitalization: the unclear relative power of these forces, the need to consider the cumulative impacts of these driving forces, and the lack of research on their interaction.
This article goes beyond theoretical contributions by providing a practical foundation for effectively adjusting digital transformation strategies. In today’s competitive environment, digital transformation is not just a means to gain a competitive edge—it is crucial for survival and long-term growth. However, it is shaped by both government policies and market dynamics, rather than being driven by a single factor. These forces continuously interact, creating a complex system that influences enterprise strategies. Policymakers who fail to understand this relationship risk implementing weak or delayed digital transformation initiatives. Without this insight, they may make misguided decisions, such as offering inadequate government support, raising barriers to digital transformation, or formulating excessive strategies that do not align with the current development landscape. Such decisions can lead to wasted resources and missed opportunities.
This study addresses both practical and research needs. By drawing on institutional theory, resource-based view, market failure theory, and stakeholder theory, it proposes hypotheses and constructs econometric models. It examines the impacts of government subsidies and market strength on enterprise digital transformation, identifies key drivers based on partial regression sum of squares, and incorporates an exponential smoothing model for dynamic analysis of the lagged and cumulative effects of government subsidies and market strength. Additionally, it develops a government-market dynamic interaction model to explain the shifting roles of synergistic dominance and phased substitution, while exploring the heterogeneity based on enterprise ownership and business life cycle.
The structure of this paper is as follows: Section 2 presents the research design; Section 3 discusses the empirical results; Section 4 presents further analysis; and Section 5 concludes with the main findings and policy implications.

2. Research Design

2.1. Hypothesis Analysis

The advancement of digital transformation in enterprises is heavily influenced by strong government policy guidance. The government promotes this transformation through a dual approach, that is, demand-side stimulation and institutional environment restructuring, creating a coupled mechanism that combines demand-driven momentum with institutional support.
On the demand side, the government uses targeted fiscal policies to activate internal momentum for digital transformation [2,4,14]. Government subsidies reduce marginal costs, making the transformation more feasible and alleviating financial pressure. This encourages more enterprises to invest in digital transformation. At the same time, the government, as a key market player, sends a positive signal through targeted subsidies and loan programs, indicating strong market potential and policy support for digital products and services. This encourages businesses to increase their investment in digital transformation and gradually fosters broad societal recognition and acceptance of digital change. This shift in social perception translates into sustained market demand for digital products and services, motivating companies to accelerate their digital transformation to meet demand and enhance competitiveness, creating a positive cycle between demand-side stimulation and corporate digital transformation [3,6,7]. On the institutional supply side, the government establishes a foundational framework to optimize the market environment and mitigate transformation risks [13,17]. A clear legal framework ensures efficient data use and enables enterprises to engage in digital transformation within a stable, predictable environment. In conclusion, demand-side stimulation guides enterprise digital transformation, while institutional restructuring provides a stable, fair market environment. Together, these efforts reduce risks and transaction costs, helping enterprises successfully navigate and achieve their digital transformation goals.
Therefore, this paper proposes the hypothesis:
Hypothesis 1.
Government subsidies promote enterprise digital transformation.
Market power theory suggests that a company’s market influence significantly shapes its behavior and strategic decisions. Firms with greater market power enjoy higher pricing control, stronger bargaining leverage, and more stable market shares, which provide them with a distinct advantage in the process of digital transformation [26].
From a motivational standpoint, these powerful companies are driven to pursue digital transformation as a means of strengthening and expanding their market position. They understand that, in today’s digital economy, transformation is crucial for sustainable growth. Continuous digital adaptation is essential to staying competitive. For instance, large traditional companies, faced with the rise of digital-native competitors, are increasing their digital investments to preserve their industry leadership and explore new business models [26,27].
When it comes to resources, companies with significant market power are better equipped to navigate the uncertainties of digital transformation. Their robust financial standing, vast resources, and market prominence allow them to make substantial investments in R&D and talent acquisition [19]. For example, major tech companies, with substantial profits and cash flow, are able to invest in advanced technologies—even when short-term returns are uncertain. They also have the ability to attract top talent, build specialized R&D teams, and form partnerships with universities and research institutions, thereby accelerating digital transformation while minimizing risks associated with resource limitations.
Therefore, this paper proposes the following hypothesis:
Hypothesis 2.
A company’s market strength drives digital transformation.
Government plays a key role by leveraging its unique functions and resources, particularly in addressing market failures and public interests. Digital transformation faces significant market failures, where government subsidies become crucial. From an externality perspective, research and development investments in digital transformation create positive externalities [28]. While enterprises invest heavily in developing digital management systems or smart manufacturing technologies, these innovations can be quickly copied by competitors, leaving private returns below social returns. Government subsidies, whether through financial support or tax incentives, help reduce the costs of innovation, encouraging further investment in digital transformation.
The infrastructure needed for digital transformation—such as cloud computing platforms and data centers—possesses public good characteristics. These infrastructures entail high construction and maintenance costs and are non-exclusive, meaning individual companies find it difficult to bear these costs on their own [29]. Government subsidies help reduce these costs, prevent redundant investments, and improve overall resource allocation efficiency in society.
As digital transformation deepens, market forces gradually come into play. Capital markets, driven by profit motives, favor companies that have successfully undergone digital transformation. These companies achieve technological innovation and market expansion, offering higher returns to investors, making it easier to secure financing, and accelerating growth. In contrast, companies that fail or progress slowly struggle to secure capital and risk cash flow problems [26,27]. Consumers also favor companies that provide more efficient, higher-quality digital products or services. Successful digital transformation allows companies to earn trust and loyalty by enhancing quality and meeting consumer needs, ultimately increasing market share [30]. Those companies that fail to meet consumer demands are phased out, optimizing overall resource allocation.
The interaction between government subsidies and market forces in digital transformation highlights the inherent differences between the government and the market in terms of resource allocation, risk-sharing, innovation incentives, and market dynamics. Initially, government subsidies correct market failures, supply public goods, and share the risks associated with transformation, providing a foundation for digital progress. Over time, market forces drive further transformation through economies of scale, innovation monopolies, data resource monopolies, and market-driven mechanisms. As such, this paper proposes the following hypothesis:
Hypothesis 3.
Government subsidies and market strength interactively drive corporate digital transformation.

2.2. Models

This study uses fixed-effects regression, lag models, partial squared regression, and coupling coordination models. Specifically, we apply fixed-effects regression to assess the impact of government subsidies and market strength on corporate digital transformation, considering their cumulative and lagged effects through lag models. Partial least squares regression helps identify key drivers of enterprise digital transformation. We also evaluate substitution effects using a control condition approach with fixed-effects regression. The coupling coordination model measures the combined impact of subsidies and market strength, which is integrated into the regression to examine their joint effect. Lastly, heterogeneity analysis is conducted based on company characteristics using fixed-effects regression. Methods are detailed below.

2.2.1. Baseline Regression Model

To explore the influence of government subsidies and market strength on digital transformation, a fixed-effects regression model is used. This approach helps control for unobserved factors that may vary across industries and years, allowing us to better isolate the impact of key variables—government subsidies and market strength—on digital transformation. The model is specified as follows:
D i g i t a l i t = β 0 + β 1 G o v i t + β 2 M a r i t + θ C o n t r o l i t + λ i + μ t + ε i t
In this model, i represents individual enterprises and t denotes year. The dependent variable, D i g i t a l i t measures the degree of digital transformation of enterprise i in year t . Explanatory variables G o v i t and M a r i t capture government subsidies and market strength, respectively, while C o n t r o l i t includes control variables, with λ i and μ t denoting time and industry fixed effects. The key focus is on coefficients β 1 and β 2 : positive values indicate a facilitating effect on digital transformation, while negative values suggest an inhibiting impact.

2.2.2. Lag Regression Model

The influence of government subsidies and market strength on enterprise digital transformation experiences a delay due to various causative factors. Policy implementation and dissemination inherently lead to an initial delay [31]. Secondly, enterprises possess limited absorptive capacity, which hinders their ability to rapidly assimilate and utilize new information or resources [19]. Additionally, market forces frequently exhibit delayed feedback cycles, with the full effects of these forces becoming apparent only over time [31]. The gradual process of technology diffusion and acceptance, along with the time required for companies to adopt and adapt to new technologies, contributes to the observed lag [32]. The cumulative effects are progressive rather than immediate, highlighting the necessity of considering long-term implications when evaluating the influence of market forces and government subsidies on digital transformation.
The distributed lag model is a prevalent approach for addressing lag effects. The model is designed to incorporate multiple lagged variables of the explanatory variables to capture dynamic effects. This method’s strength lies in its ability to explicitly capture lag effects; however, it also presents notable disadvantages. Since multiple lagged terms are introduced, it is easy to encounter multicollinearity issues, which can make parameter estimation difficult and lead to a loss of degrees of freedom. Given the limitations of the distributed lag model, this study uses a model with exponentially smoothed variables. This model employs exponential smoothing to construct a dynamic weighting mechanism. On the one hand, by using decaying weights, it effectively balances the short-term and long-term effects. In the gradual process of digital transformation for enterprises, recent changes often have a more direct and significant impact on the current level of transformation, but the accumulated long-term effects cannot be overlooked. Exponential smoothing helps coordinate these two aspects. On the other hand, compared to static lag models, it can capture the gradual diffusion process of policy effects. Additionally, by smoothing the data, it reduces multicollinearity between variables, making the model more stable and reliable. The model is specified as follows:
D i g i t a l i t = β 0 + β 1 G o v s m o o t h e d i t + β 2 M a r s m o o t h e d i t + θ C o n t r o l s i t + λ i + μ t + ε i t
In this model, D i g i t a l i t represents the dependent variable capturing the degree of digital transformation for enterprise i at time t . G o v _ s m o o t h e d and M a r _ s m o o t h e d are exponentially smoothed indicators for government subsidies and market strength, respectively, reflecting their cumulative and lagged effects. The control vector C o n t r o l s i t includes variables such as enterprise size, age, R&D investment, and profitability, which influence digital transformation. λ i and μ t are fixed effects accounting for industry-specific and temporal factors, while ε i t denotes the error term. This specification ensures the model captures the dynamic and long-term impacts of government and market drivers.
The exponential smoothing formula is specified as:
D r i v e r s m o o t h e d i t = α D r i v e r i t + 1 α D r i v e r i t 1
where D r i v e r s m o o t h e d i t represents the exponentially smoothed government and market driving forces for enterprise i in year t ; D r i v e r i t denotes the actual driving forces for enterprise i in year t ; α is the smoothing factor, a value between 0 and 1.
The smoothing factor determines how much weight is given to the most recent data. A high α (closer to 1) places more emphasis on recent observations, making the smoothed series more reactive to short-term changes. A low α (closer to 0) reduces the impact of recent fluctuations, prioritizing historical trends and reducing noise. In practice, α values between 0.01 and 0.3 are commonly selected to balance adaptability and stability [33]. This range ensures the smoothed series captures gradual shifts in driving forces while mitigating idiosyncratic volatility. Here, we adopt α = 0.1 .

2.2.3. Contribution Comparison

To systematically evaluate the relative importance of governmental and market forces in driving enterprise digital transformation, this study employs partial regression sum of squares (PRSSs). PRSSs quantify the proportion of the total variation in digital transformation that can be explained by either government or market forces, after controlling for other factors [34].
The analysis proceeds as follows:
    • Step 1: Estimate the full specification model, which includes both government subsidies and market strength as explanatory variables (Equations (2) and (4) are two identical equations):
      D i g i t a l i t = β 0 + β 1 G o v s m o o t h e d i + β 2 M a r s m o o t h e d i + θ X i + λ i + μ i + ε i
    • Step 2: Estimate two restricted baseline models to isolate the individual impacts of government subsidies and market strength:
Government-exclusive specification:
D i g i t a l i t = β 0 + β 1 G o v s m o o t h e d i + θ X i + λ i + μ i + ε i
Market-exclusive specification:
D i g i t a l i t = β 0 + β 2 M a r s m o o t h e d i + θ X i + λ i + μ i + ε i
    • Step 3: Estimate the individual explanatory contribution of each factor, which is estimated by the PRSS, which is estimated through variance decomposition. Specifically, the PRSS for government subsidies and market strength is estimated as follows:
      P R S S G o v = R f u l l 2 R w i t h o u t _ G o v 2
P R S S M a r = R f u l l 2 R w i t h o u t _ M a r 2
where R f u l l 2 denotes the R-squared for the full specification model (Equation (4)), and R w i t h o u t _ G o v 2 and R w i t h o u t _ M a r 2 denote the R-squared values of the government-exclusive and market-exclusive models, respectively.
    • Step 4: Identification of the primary driver:
Compare the PRSS values of government subsidies and market strength to identify the core driving force of enterprise digital transformation. If the market strength has a higher PRSS value than the government subsidies, it indicates that market strength is a more significant driver of digital transformation than government subsidies. This comparison allows us to determine which factor plays a more dominant role in shaping digital transformation.

2.2.4. Coupling Coordination Model

To assess the joint interaction between government subsidies and market strength, the coupling coordination model (CCM) is employed. The CCM is commonly used to evaluate the degree of coordination or synergy between different subsystems and their collective influence on overall performance or development [35,36,37]. The CCM implementation follows these steps:
    • Step 1: Compute the coupling degree to measure interaction intensity. The coupling degree (CD) is given by the following formula:
      C D = s m o o t h _ G o v × s m o o t h _ M a r [ ( s m o o t h _ G o v + s m o o t h _ M a r ) / 2 ] 2 2
    • Step 2: Compute the comprehensive evaluation to assess overall development. In this step, the weights of government subsidies and market subsidies are assumed to be equal. The comprehensive evaluation (T) is calculated as follows:
      T = 0.5 × s m o o t h _ G o v + 0.5 × s m o o t h _ M a r
    • Step 3: Compute the coupling coordination degree. The coupling coordination degree (CCD) is determined by multiplying the coupling degree by the following comprehensive evaluation:
      C C D = C D × T
This methodology provides a structured approach to evaluating the interaction and coordination between government subsidies and market strength, reflecting their combined impact on overall development.

2.2.5. Robustness Tests Methodology

To ensure the validity and reliability of the study’s findings, we performed three robustness checks by systematically altering key model specifications. This approach helps mitigate potential biases arising from model misspecification, omitted variables, or assumptions related to parameter estimation. Below is a detailed explanation of each robustness test.
  • Switch to random effects regression models. The key difference between the random effects and fixed effects regression models lies in how they handle unobserved heterogeneity (i.e., differences between enterprises that are not captured by the observed variables) and how they treat the enterprise-specific effects. The random effects model assumes that the entity-specific effects are uncorrelated with the independent variables in the model. The random effects model includes these entity-specific effects as part of the error term.
  • Substitute control variables with analogous variables. We replaced certain control variables with comparable alternatives. Specifically, profitability was replaced with the operating income growth rate and the cost-to-profit ratio, while enterprise size was substituted with total assets. This substitution helps test whether the results are robust to different ways of measuring the same underlying concepts.
  • Change the parameter estimation method to weighted least squares (WLS). The ordinary least squares (OLS) method assumes constant variance in the errors. When this assumption is violated, WLS provides a more appropriate method. It addresses heteroscedasticity by assigning weights inversely proportional to the variance of residuals. Observations with smaller error variance receive larger weights, as they contain more reliable information. The methodology for WLS is as follows: Estimate initial residuals from the OLS model. Compute the weights as w i = 1 σ ^ i 2 , where σ ^ i 2 is the estimated variance of residuals for observation i . Re-estimate the model using WLS with these weights.

2.3. Variables and Data

This section details the variables and their measurement methods employed in this study. The variables are categorized into dependent, independent, and control variables, with detailed descriptions provided in Table 1. The empirical data employed in this analysis were sourced exclusively from the China Stock Market & Accounting Research (CSMAR) database.
Enterprise digital transformation is the dependent variable. Digital transformation is defined as the strategic process in which businesses leverage digital technologies to reshape their processes, organizational structures, products, services, and business models [38]. Common measurement methods include expert evaluation, text analysis, and the intangible asset ratio method [12,39]. Expert evaluation relies on specialists to assess multiple factors like strategy, technology, and organizational change, but results can be biased and lack comparability. Text analysis uses natural language processing to extract digital transformation-related keywords from texts, efficiently handling large volumes but only capturing textual attention to transformation, not actual outcomes. The intangible asset ratio method relies on details from a company’s financial statements, identifying intangible assets related to digital transformation and calculating their proportion to assess transformation extent. The reason for using the proportion of digital intangible assets is that changes in their scale and proportion reflect a company’s investment in digital transformation [40]. Digital intangible assets—like software, digital platforms, and intellectual property—are clear signs of a company’s commitment to adopting and integrating digital technologies [41]. Additionally, the results of digital transformation often show up as digital intangible assets. This method provides strong data availability and comparability since digital intangible assets are generally well-documented and measurable, offering a consistent and reliable metric across enterprises. It also helps reduce potential biases that could arise from relying on subjective measures of transformation. For these reasons, this study uses the digital intangible asset ratio method to measure digital transformation.
The independent variables include government subsidies and market strength. Government subsidies are measured using government subsidies for enterprise digitalization. These subsidies help alleviate resource constraints, support R&D, and promote digital technology adoption. Data on subsidies, sourced from the CSMAR database, are extracted using keywords such as “artificial intelligence”, “blockchain”, “big data”, and “cloud computing”. The variable is measured by the ratio of subsidies related to digitalization to a company’s total assets. Market strength refers to the objective forces driven by external market competition, changes in customer demand, and other market mechanisms. The core logic behind this is that changes in market structure, through competitive pressure and upgraded demand, force companies to actively or passively implement digital transformation to survive and grow. The Lerner Index of a company reflects its position in the market, making it reasonable to use the Lerner Index to measure the market drivers behind digital transformation. The Lerner Index gauges a company’s market position by the degree of deviation between its pricing power and marginal cost. A higher value indicates a stronger market position, while a lower value suggests a weaker position.
Control variables include enterprise characteristics, financial factors, and the external microeconomic environment. Enterprise size and age are critical factors: larger enterprises typically have more resources and are more inclined toward digital transformation, while older enterprises may exhibit organizational inertia, hindering transformation efforts. Financial control variables, such as current ratio, debt-to-asset ratio, profitability, and cash flow, assess enterprises’ capacity to invest in digital transformation. Loss-making enterprises may exhibit heightened motivation to transform to improve efficiency and profitability. Lastly, external economic factors, such as market volatility, influence enterprises’ strategic planning and investment in digital transformation, as economic pressures may constrain or accelerate these efforts.
This study employs data from China’s A-share-listed companies spanning 2003 to 2023, obtained from the CSMAR Economic and Financial Research Database. The data were processed using the following filters: financial institutions were excluded, delisted companies during the study period were removed, enterprises with significant missing financial data were eliminated, and enterprises with only one or two years of data were excluded. After screening, the final sample includes 28,324 observations from 3422 enterprises.

3. Econometric Analysis of Digital Transformation Drivers

3.1. Regression Analysis of Digital Transformation Drivers

Table 2 presents the baseline regression results, examining the effects of government and market forces on enterprise digital transformation. Column (1) assesses the impact of government subsidies (Gov), while column (2) examines market strength (Mar). Column (3) incorporates both variables to analyze their relative contributions. Columns (4)–(6) introduce their smoothed counterparts (Gov_smoothed and Mar_smoothed) to capture potential long-term effects.
Table 2 examines the impacts of government subsidies and market strength on the digital transformation of enterprises. To be specific, the regression analysis reveals that the parameter of government subsidies is significantly positive at the 5% significance level, indicating a significant positive impact of government subsidies on digital transformation within enterprises, thus supporting Hypothesis 1. Similarly, the regression parameter of market strength is significantly positive at the 1% level, suggesting that the enterprises’ market strength is also crucial in motivating enterprises to adopt digital transformation, supporting Hypothesis 2. This finding indicates that the driver of enterprise digital transformation is not isolated.
The positive influence of these factors is not limited to the present but extends over time. Specifically, the time-aggregated regression coefficients for government subsidies and market strength are statistically significant and positive at the 1% level, with values higher than those for the current period. This indicates that both government subsidies and market strength have cumulative effects in promoting enterprise digital transformation.
This result is consistent with existing literature that establishes the critical roles of both the government and the market in determining the diffusion of digital technology. Previous studies [7,8,14] have pointed out that government initiatives—such as the development of digital infrastructure, subsidies for digital technology adoption, and regulatory policies—create a favorable environment for businesses to become digital. On the other hand, competitive forces and consumer demand have also been acknowledged as fundamental drivers of digital innovation within enterprises [19]. The findings in Table 2 reinforce these findings, highlighting the significance of these drivers in fostering digital transformation.
Moreover, these results reflect real-world trends, as governments worldwide increasingly recognize the importance of digital technologies and have implemented many initiatives aimed at promoting business digitization. Examples include national programs like “Made in China 2025” and the EU’s “Digital Single Market,” which are designed to accelerate digital adoption across industries. Concurrently, market pressures—such as the rise of e-commerce, the growing importance of data analytics, and the rapid pace of technological innovation—are compelling companies to undergo digital transformation to remain competitive and meet evolving customer demands.
To ensure the reliability of our findings, we conducted robustness tests using three alternative approaches.
As Table 3 clearly shows, the conclusion that government subsidies as well as market strength drive enterprise digital transformation is robust. The two variables are positive and significant at the 1% level. The coefficient of government subsidies is notably greater than that of the market. This result remains stable across different scenarios, whether altering the method of parameter estimation or changing the control variables.

3.2. Analysis of Core Drivers of Digital Transformation

To quantify the relative contributions of government and market forces to enterprise digital transformation, we employ the partial regression sum of squares (PRSS) method. Table 4 presents the estimation results, where Columns (1) and (2) examine the individual effects of government and market forces, respectively, while Column (3) incorporates both variables to assess their combined influence.
Table 4 suggests that government subsidies play a significant role in driving enterprise digital transformation when compared with market strength. The partial regression sum for government subsidies is calculated as the difference between the R2 of the full regression model (which includes government subsidies) and the R2 of the model without government subsidies. Similarly, the partial regression sum for market strength can be calculated by subtracting the R2 of the model with no market strength from the R2 of the entire model, and that will result in 0.066. Since the partial regression sum for government subsidies (0.068) is slightly higher than that for market strength (0.066), it indicates that government subsidies have a marginally more significant impact on driving enterprise digital transformation, though the difference is small.
This can be explained through resource dependence theory and the institutional environment. Firstly, according to resource dependence theory, an enterprise’s strategies are shaped by its reliance on external resources. Government subsidies, particularly in funding and technology, provide critical support during digital transformation. Enterprises often require these resources to remain competitive and adapt to technological advancements. With government subsidies, enterprises can overcome financial and technical challenges, enhancing their ability to innovate and transform. Secondly, companies are also influenced by external institutional frameworks. When the government offers subsidy policies (such as those promoting digital technology or smart upgrades), these policies create pressure on enterprises to align their strategies with government goals. For example, subsidies for smart manufacturing may encourage enterprises to update production lines, integrate IoT devices, or build data management platforms. In this context, subsidies act not only as financial support but also as institutional guidelines, nudging enterprises toward transformation goals, such as meeting technical standards or adhering to implementation timelines. Maintaining market power is important, but with government support, enterprises can more effectively navigate the challenges of digitalization, enhancing their adaptability and innovation. Therefore, while both market position and government subsidies are important, the latter can play a slightly more significant role in driving digital transformation.

3.3. Analysis of the Substitution Effect Between Government and Market Forces

The influence of government subsidies and market strength is not static; instead, these two drivers may alternate in their prominence depending on the stage of digital transformation. To further explore this interaction, we examine whether an alternation mechanism exists between government subsidies and market strength.
We begin by introducing an interaction term into the regression model to test for the presence of a substitution effect. Using the principle of control variates, we then analyze how the substitution effect manifests by examining changes in the coefficients of government subsidies across the groups categorized by market strength. Finally, we perform statistical tests to validate the differences across these groups using regressions that include interaction terms.
The interaction term is formed by multiplying the centralized values of government subsidies and market strength. In the regression model, the coefficient of this interaction term indicates how the effect of one variable (e.g., government subsidies) on the outcome (e.g., digital transformation) changes as the other variable (e.g., market strength) varies. A negative coefficient suggests that the marginal effect of government subsidies on digital transformation diminishes as market strength increases, signaling a substitution effect. In other words, as market forces strengthen, the additional impact of government subsidies decreases. Conversely, a positive coefficient implies a complementary effect, where the influence of government subsidies increases as market forces become more powerful. The results are presented in Table 5.
Table 5 demonstrates the existence of substitution effects between government subsidies and market strength, thereby verifying Hypothesis 3. The coefficient of the interaction term is negative and statistically significant at the 5% significance level. This result suggests that as market strength increases, the marginal impact of government subsidies on digital transformation decreases.
To confirm and clarify the substitution effects, we categorized the companies into three groups based on their market strength, dividing them into thirds according to quantiles. This approach follows the principle of control variates, which emphasizes holding other factors constant to isolate the effect of a specific variable. By grouping companies based on market strength, we control for variations in market forces across firms, allowing us to focus on how government subsidies interact with market strength within each group. The results are presented in Table 6.
The comparative contribution of government subsidies and market strength to digital transformation reveals a clear substitution effect, demonstrating the robustness of the findings. In enterprises with low market strength, the coefficient of Gov_smoothed is positive and statistically significant at the 1% level, while the coefficient of Mar_smoothed is not statistically significant. However, in enterprises with medium and high market strength, the coefficient of Gov_smoothed becomes statistically insignificant, while the coefficient of Mar_smoothed is positive and statistically significant. This suggests that for enterprises with strong market power, digital transformation is driven more by their own market strength than by government subsidies. In contrast, for enterprises with limited market strength, government subsidies play a more significant role in promoting digital transformation.
We conduct additional statistical tests to assess the differences in the effects of government subsidies across market strength groups. To formally examine these group-specific differences, we employ an interaction term approach. For example, to test the difference in coefficients between the low- and mid-market strength groups, we set M a r D u m m y = 0   for the low-market strength group (the reference group) and M a r D u m m y = 1   for the mid-market strength group. The interaction term in this case would be Gov_smoothed × MarDummy. We then regress digital on Gov_smoothed, MarDummy, and the interaction term Gov_smoothed × MarDummy, while controlling for other relevant factors. The rationale behind this approach is that the interaction term helps quantify how the marginal effect of government subsidies varies across groups defined by market strength. If the interaction term is statistically significant, it indicates that the marginal effect of government subsidies differs across market strength groups, thus confirming differences between these groups. We conduct pairwise comparisons as follows: low vs. mid, low vs. high, and mid vs. high. Note that the interaction data should be centered to avoid multicollinearity and ensure accurate interpretation of the results. The results are presented in Table 7.
Table 7 confirms the differences among the three groups. The interaction coefficients in columns (1) and (2) are negative and statistically significant, indicating a significant difference in the effect of subsidies between the low- and mid-market strength groups, as well as between the low- and high-market strength groups. The interaction coefficient in column (3) is not statistically significant, suggesting no significant difference between the mid- and high-market strength groups, which also aligns with the results in Table 6.
The capability life cycle theory offers a framework for understanding the substitution effect in organizational development. In the early stages of digital transformation, companies with weak market power enter the capability exploration phase. Here, they face resource scarcity and lack the technical foundation necessary for independent transformation, making government subsidies crucial to reduce risks. As time passes, companies transition into the capability accumulation phase or growth stage. During this time, they begin developing digital capabilities, such as basic IT infrastructure and data collection methods, though they may not yet be profitable. In this phase, subsidies play a vital role by accelerating capability advancement through targeted support, like R&D tax credits, and helping build market competitiveness by attracting users via digital products. Upon reaching the maturity stage, or capability self-enhancement phase, companies achieve endogenous digital capabilities. They can make data-driven decisions and develop a stable profit model. Market power becomes essential here, functioning through two main mechanisms: reinvestment of profits for technology upgrades and the creation of ecosystem synergies that lower transformation costs.
The substitution effect occurs as market power gradually replaces subsidies, facilitating ongoing digital transformation. To transition from subsidy dependency to market independence, companies must reach specific technical (core digital system deployment) and economic (break-even) thresholds during the subsidy period. Meeting these thresholds allows market forces to take over, driving mature digital transformation. Failure to do so may result in a subsidy dependency trap.

4. Further Analysis

4.1. Analysis of the Joint Role of Government and Market Drivers

In practice, government policy guidance and market-driven mechanisms jointly influence the digital transformation process of enterprises. While they often complement each other, they can also create friction. To better understand the dynamics between these two forces, the coupling coordination model—a method used to study the interaction and coordination between systems—is applied in this study. This approach allows for a systematic exploration of how the degree of coordination between government subsidies and market strength impacts enterprise digital transformation.
First, we calculated the coordination between government subsidies and market strength and then incorporated it into the regression model to explore its impacts through the coefficient. The results are presented in Table 8.
As shown in Table 8, the coordination between government subsidies and market forces plays a positive role in driving enterprise digital transformation. The coefficient of coordination is positive and significant at the 5% level, suggesting that as the coordination between government subsidies and market strength improves, the level of enterprise digital transformation increases as well. This finding aligns with the common viewpoints in the literature that emphasize the synergistic effects of external support and internal capabilities in enhancing organizational performance.
From the perspective of synergy theory, the collaboration between government subsidies and market forces can be seen as a dynamic complementary mechanism between public policy tools and a company’s endogenous capabilities. The core idea is that, through the integration of resources and complementary capabilities, the limitations of individual actors or resources are surpassed, resulting in a digital transformation effect where the whole exceeds the sum of its parts. Government subsidies, such as public policy tools, primarily provide short-term resource support, including financial injections, tax reductions, and policy assistance. These resources help alleviate the financial pressures and technological bottlenecks faced by businesses during the early stages of digital transformation. Market forces, on the other hand, rely on their long-term accumulated capabilities, such as continuous technological iteration, extensive user networks, and mature market operation mechanisms, to provide ongoing momentum for digital transformation. This complementary synergy enables companies to not only gain short-term resource support but also leverage long-term market capabilities for sustainable digital development.

4.2. Analysis of Heterogeneity

4.2.1. Analysis of Heterogeneity by Ownership Type

State-owned and private enterprises have different business goals and incentive structures. State-owned enterprises tend to focus more on social responsibility and national objectives, while private companies are more focused on profit and market share growth. As a result, the impact of market competition and government drivers on their digital transformation efforts also differs. Table 9 presents the results of the heterogeneity analysis, exploring how the effects of government and market forces differ across enterprise ownership types. Columns (1) and (2) compare state-owned and non-state-owned enterprises.
Government subsidies and market strength affect digital transformation in state-owned enterprises (SOEs) and non-state-owned enterprises (NSOEs) in a heterogeneous way. Estimates from Column (1) for state-owned enterprises indicate that both government subsidies and market strength significantly promote digital transformation, with coefficients of 16.37 and 0.187, respectively. Column (2) for non-state-owned enterprises shows a different pattern. While government subsidies still have a significant positive impact (4.383), market strength shows a negative and statistically significant relationship (−0.0829) with digital transformation. This suggests that, for NSOEs, digital transformation is primarily driven by government subsidies, while market strength may act as a limiting factor.
The results indicate that government subsidies significantly aid the digital transformation of enterprises, regardless of their ownership structure. The mechanisms driving this effect have been discussed previously and will not be reiterated here. This section focuses on how market strength impacts SOEs and NSOEs differently. According to Property Rights Theory, the difference lies in strategic focus: SOEs balance policy goals with economic outcomes, viewing digital transformation as part of national strategies, while NSOEs typically prioritize short-term financial performance and profit maximization.
For NSOEs, their market strength may hinder digital transformation in several ways. Enterprises with a strong market position often rely on established business models, creating deep path dependencies. This reliance on stable networks and processes makes them more cautious about embracing the costly and disruptive changes needed for digital transformation. Moreover, the long-term benefits of digital transformation conflict with the short-term performance metrics. With limited resources, management tends to prioritize investments in traditional business areas that generate immediate profits, leaving digital transformation efforts underfunded. Additionally, digital transformation is a high-risk, long-term investment, and NSOEs—focused on maintaining their competitive edge—are often risk-averse. This leads to delays or even abandonment of transformation efforts, despite recognizing the long-term value of digital change.
These findings challenge the common assumption that enterprises, particularly those with strong market positions, will always embrace transformation to maintain dominance. In reality, not all high-market-position enterprises actively pursue digital transformation, especially not all high-market-position NSOEs.

4.2.2. Analysis of Heterogeneity by Life Cycle

The business life cycle stage (growth, maturity, and decline) influences a company’s resource acquisition, strategic focus, and market pressures, which in turn affect the driving forces behind digital transformation. In this section, we examine how the effects of government subsidies and market strength impact business across different enterprise life cycle stages. To classify enterprises by their life cycle stage, we use a comprehensive index that incorporates the following key variables: sales revenue growth rate, retention rate of earnings, capital expenditure rate, and company age, with equal weights assigned to the scores of each variable. The score for each variable is determined by its rank relative to other enterprises in the industry. These scores are then averaged, with equal weight given to each variable, to calculate the overall composite score. To ensure balanced sample sizes and minimize potential biases, we divided the companies into three groups based on their combined scores. Companies in the top third are classified in the growth phase, those in the bottom third are placed in the decline phase, and those in the middle third are categorized as mature enterprises. Table 10 presents the results of our heterogeneity analysis, with columns (1)–(3) showing the analysis for growth-stage, mature-stage, and decline-stage companies, respectively.
As shown in Table 10, the impact of government subsidies and market strength on digital transformation varies significantly across enterprises at different stages. Government subsidies have a significant positive effect on digital transformation in the decline stage, but their influence becomes statistically insignificant in both the growth and maturity stages. In contrast, market strength consistently shows a significant positive impact across all stages.
This conclusion aligns with the theory of the business life cycle and is also consistent with the law of diminishing marginal returns. When a company is in the decline stage, resources are typically limited, and the risks are higher. Digital transformation, which allows the company to meet government subsidy requirements, can help secure the external resources it needs. However, for companies in the growth and maturity stages, their internal resources (such as profits and financing capacity) are generally more abundant. As a result, the marginal effect of government subsidies diminishes and becomes less significant. This phenomenon suggests that government subsidies should be adjusted based on the company’s stage of development to improve policy efficiency.

5. Conclusions and Discussions

Our empirical analysis of the core drivers of enterprise digital transformation is structured in two parts. First, we test whether government subsidies and market strength drive enterprise digital transformation using a fixed-effects regression model and smoothing exponential, examining both short-term and long-term impacts. Next, we explore the core drivers through a partial regression sum of squares. We also investigate the substitution effects between government subsidies and market strength through group regressions and interaction terms. The results show that both government subsidies and market strength promote enterprise digital transformation, with government subsidies playing a slightly more dominant role than market strength. Additionally, we observe substitution effects, where the dominance of government subsidies gradually shifts to market strength as enterprises’ market strength increases. Building on these findings, we analyzed the impact of coordination between government and market forces in enterprise digital transformation and found their joint role in promoting digitalization. We examined the varying effects of government subsidies and market strength across different enterprise types. The results reveal that government subsidies support digital transformation regardless of ownership, while market strength drives transformation in state-owned enterprises but may limit it in non-state-owned ones. Subsidies significantly impact transformation in the decline stage but lose significance in the growth and maturity stages, while market strength consistently has a positive impact across all stages.
Findings from this study offer valuable insights into the digital transformation of enterprises. To maximize the effectiveness of policy subsidies, it is essential to implement differentiated subsidy strategies and dynamic exit mechanisms. For enterprises in the decline phase or those with weak market forces, high-intensity subsidies should be provided to lower the costs of digital transformation. For growth-stage enterprises with increasingly strong internal capabilities, the intensity should gradually decrease to prevent wasteful allocation of resources. However, if evaluation criteria are unclear during implementation, it may lead to unfair resource allocation. Some enterprises may exploit this by intentionally delaying their capacity development or exaggerating their decline to continue receiving subsidies. This undermines the policy’s incentivizing effect and can result in resource waste. Furthermore, poorly designed exit mechanisms might cause enterprises to suddenly lose support during their growth phase, hindering their long-term development. To address these issues, a multidimensional, dynamic evaluation system should be established. This system would regularly assess the life cycle stage of enterprises, track their progress in digital transformation, and evaluate the actual impact and utilization of subsidies. These ongoing assessments will serve as a foundation for adjusting subsidy levels as needed. Moreover, policy design must pay special attention to the unique challenges faced by non-state-owned enterprises, particularly those with strong market positions, as they may resist or slow down their digital transformation efforts for various reasons. To better understand these barriers, in-depth research, including enterprise interviews and case studies, should be conducted to uncover the common reasons behind this reluctance. Based on these findings, policies should be refined to better guide corporate behavior, ensuring a balanced and positive interaction between market forces and policy incentives.
This study makes several key theoretical contributions. First, it integrates market forces, government intervention, and enterprise digital transformation into a framework, moving beyond a narrow focus solely on either the market or the government. This framework also considers the cumulative effects of both government and market influences. Secondly, it identifies and illustrates the substitution effects between government subsidies and market strength, which expands our understanding of the relationship between government intervention and market mechanisms. Thirdly, it enriches the theoretical understanding of the contextual dependencies of government and market strength in digital transformation. A common perspective suggests that greater market strength enables enterprises to access more resources, thereby driving their digital transformation. While market strength does contribute to digital transformation, its impact depends on the ownership structure of the enterprise. The strength of the market may actually work against the transformation of non-state-owned enterprises. Traditionally, government subsidies are viewed as a positive incentive for digital transformation. However, the findings of this study show that the impact of subsidies is stage-dependent; that is, subsidies may not significantly promote transformation in mature or growth-stage enterprises, but they have a clear positive effect in enterprises that are in decline.
The limitations of this study are as follows: First, the generalizability of the conclusions is restricted. This study uses A-share-listed companies in China as case studies, and significant differences in economic development, institutional environments, and cultural contexts between countries may result in different drivers of digital transformation, making it difficult to directly apply the conclusions to other countries or regions. Second, the explanation of how non-state-owned enterprises inhibit digital transformation is theoretical, and the study does not empirically explore the specific mechanisms through which they suppress it. Additionally, a simplified linear model was used in the analysis of variable relationships, but the digital transformation process may involve complex nonlinear effects that were not sufficiently considered in this study. Future research could focus on the following areas: On the one hand, cross-country comparative studies could be conducted to comprehensively analyze the interaction mechanisms between government subsidies and market forces under different institutional environments and development contexts, revealing the commonalities and differences in digital transformation policies across countries. On the other hand, a deeper exploration of the specific mechanisms through which market forces in non-state-owned enterprises suppress digital transformation is needed. Factors such as path dependence, organizational inertia, information asymmetry, technological lock-in, and governance structures warrant further investigation in terms of their inhibiting paths and mechanisms. Moreover, it is necessary to examine the nonlinear effects of government subsidies and market forces and develop more complex analytical models to capture the nonlinear relationships between variables. Finally, introducing a dynamic perspective is crucial, as digital transformation is a dynamic and evolving process. The impact of government subsidies and market forces on enterprises may change at different stages, and future research should examine these dynamic effects to provide more timely policy recommendations.

Author Contributions

Conceptualization, T.L.; methodology, L.N.; software, L.N.; formal analysis, Y.X. and L.N.; data curation, Y.X.; writing—original draft preparation, Y.X. and L.N.; writing—review and editing, T.L.; supervision, T.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Restrictions apply to the availability of the data that support the findings. The data were obtained from the China Stock Market & Accounting Research Database (CSMAR), a commercial provider available at https://data.csmar.com/, accessed on 1 December 2024. Under the terms of the license agreement, the authors are not permitted to share the raw data. Researchers must obtain the data directly from CSMAR.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
CCMcoupling coordination model
SMEssmall- and medium-sized enterprises
PRSSpartial regression sum of squares
CSMARChina Stock Market & Accounting Research database
WLSweighted least squares
SOEsstate-owned enterprises
NSOEsnon-state-owned enterprises

References

  1. Reis, J.; Melão, N. Digital Transformation: A Meta-Review and Guidelines for Future Research. Heliyon 2023, 9, e12834. [Google Scholar] [CrossRef] [PubMed]
  2. Wang, S.; Li, X.; Li, Z.; Ye, Y. The Effects of Government Support on Enterprises’ Digital Transformation: Evidence from China. Manag. Decis. Econ. 2023, 44, 2520–2539. [Google Scholar] [CrossRef]
  3. Peng, Z.; Huang, Y.; Liu, L.; Xu, W.; Qian, X. How government digital attention alleviates enterprise financing constraints: An enterprise digitalization perspective. Financ. Res. Lett. 2024, 67, 105883. [Google Scholar] [CrossRef]
  4. Song, S.; Jiao, J.; Yang, R. The Impacts of Government R&D Subsidies on Green Innovation: Evidence from Chinese Energy-Intensive Firms. J. Clean. Prod. 2019, 233, 819–829. [Google Scholar]
  5. Senyo, P.K.; Karanasios, S.; Komla Agbloyor, E.; Choudrie, J. Government-Led Digital Transformation in Fintech Ecosystems. J. Strateg. Inf. Syst. 2024, 33, 101849. [Google Scholar] [CrossRef]
  6. Jin, X.; Pan, X. Government Attention, Market Competition and Firm Digital Transformation. Sustainability 2023, 15, 9057. [Google Scholar] [CrossRef]
  7. Li, X.; Yue, S. Does the government digital attention improve China’s digital economy output efficiency: Accelerator or inhibitor. Econ. Anal. Policy 2025, 85, 607–625. [Google Scholar] [CrossRef]
  8. Verhoef, P.C.; Kannan, P.K.; Inman, J.J. From multi-channel retailing to omni-channel retailing: Introduction to the special issue on multi-channel retailing. J. Retail. 2015, 91, 174–181. [Google Scholar] [CrossRef]
  9. Teece, D.J.; Pisano, G.; Shuen, A. Dynamic Capabilities and Strategic Management. Strateg. Manag. J. 1997, 18, 509–533. [Google Scholar] [CrossRef]
  10. Kamble, S.S.; Gunasekaran, A.; Dhone, N.C. Industry 4.0 and the digitalisation of supply chains in industrial manufacturing: A systematic review. Ind. Mark. Manag. 2023, 61, 1186–1221. [Google Scholar]
  11. Fan, J.; Xu, M. The Power of Competition: Unveiling China’s Digital Transformation through Product Market Dynamics. Econ. Lett. 2024, 234, 111479. [Google Scholar] [CrossRef]
  12. Yan, A.; Ma, H.; Zhu, D.; Xie, J. Digital Transformation and Corporate Resilience: Evidence from China during the COVID-19 Pandemic. Quant. Financ. Econ. 2024, 8, 779–814. [Google Scholar] [CrossRef]
  13. Tangwaragorn, P.; Charoenruk, N.; Viriyasitavat, W.; Tangmanee, C.; Kanawattanachai, P.; Hoonsopon, D.; Pungpapong, V.; Pattanapanyasat, R.-P.; Boonpatcharanon, S.; Rhuwadhana, P. Analyzing Key Drivers of Digital Transformation: A Review and Framework. J. Ind. Inf. Integr. 2024, 42, 100680. [Google Scholar] [CrossRef]
  14. Chuc, N.D.; Anh, D.T. Digital Transformation in Vietnam: Policies, Results and Recommendations. Southeast Asian Econ. 2023, 40, 127–144. [Google Scholar] [CrossRef]
  15. Liu, B.; Li, Y.; Liu, J.; Hou, Y. Does Urban Innovation Policy Accelerate the Digital Transformation of Enterprises? Evidence Based on the Innovative City Pilot Policy. China Econ. Rev. 2024, 85, 102167. [Google Scholar] [CrossRef]
  16. Zhang, H.; Liu, X.; Quan, T. The Impact and Mechanism of the ‘Specialized and New’ Policy on Enterprise Digital Transformation—Evidence from China. Technol. Anal. Strateg. Manag. 2024, 1–14. [Google Scholar] [CrossRef]
  17. Mai, B.T.; Nguyen, P.V.; Ton, U.N.; Ahmed, Z.U. Government Policy, It Capabilities, Digital Transformation, and Innovativeness in Post-COVID Context: Case of Vietnamese Smes. Int. J. Organ. Anal. 2023, 32, 333–356. [Google Scholar] [CrossRef]
  18. Lutfi, A.; Alsyouf, A.; Almaiah, M.A.; Alrawad, M.; Abdo, A.A.; Al-Khasawneh, A.L.; Ibrahim, N.; Saad, M. Factors Influencing the Adoption of Big Data Analytics in the Digital Transformation Era: Case Study of Jordanian Smes. Sustainability 2022, 14, 1802. [Google Scholar] [CrossRef]
  19. Peng, H.; Bumailikaimu, S.; Feng, T. The Power of Market: Venture Capital and Enterprise Digital Transformation. North Am. J. Econ. Financ. 2024, 74, 102218. [Google Scholar] [CrossRef]
  20. Guo, B.; Feng, Y.; Lin, J. Digital Inclusive Finance and Digital Transformation of Enterprises. Financ. Res. Lett. 2023, 57, 104270. [Google Scholar] [CrossRef]
  21. Nikopoulou, M.; Kourouthanassis, P.; Chasapi, G.; Pateli, A.; Mylonas, N. Determinants of Digital Transformation in the Hospitality Industry: Technological, Organizational, and Environmental Drivers. Sustainability 2023, 15, 2736. [Google Scholar] [CrossRef]
  22. Haffke, I.; Kalgovas, B.; Benlian, A. The Transformative Role of Bimodal IT in an Era of Digital Business. In Proceedings of the Annual Hawaii International Conference on System Sciences, Waikoloa Village, HI, USA, 4–7 January 2017. [Google Scholar]
  23. Lokuge, S.; Sedera, D.; Cooper, V.; Burstein, F. Digital Transformation: Environmental Friend or Foe? Panel Discussion at the Australasian Conference on Information Systems. Commun. Assoc. Inf. Syst. 2021, 48, 616–634. [Google Scholar]
  24. Luo, Y.; Cui, H.; Zhong, H.; Wei, C. Business Environment and Enterprise Digital Transformation. Financ. Res. Lett. 2023, 57, 104250. [Google Scholar] [CrossRef]
  25. Cheng, Y.; Xu, Z. Fiscal Policy Promotes Corporate Green Credit: Experience from the Construction of Energy Conservation and Emission Reduction Demonstration Cities in China. Green Financ. 2024, 6, 1–23. [Google Scholar] [CrossRef]
  26. Sheng, D.; Guyot, O. Market Power, Internal and External Monitoring, and Firm Distress in the Chinese Market. Data Sci. Financ. Econ. 2024, 4, 285–308. [Google Scholar] [CrossRef]
  27. Fitzgerald, M.; Kruschwitz, N.; Bonnet, D.; Welch, M. Embracing Digital Technology: A New Strategic Imperative. MIT Sloan Manag. Rev. 2013, 55, 1–12. [Google Scholar]
  28. Sama, L.M.; Stefanidis, A.; Casselman, R.M. Rethinking corporate governance in the digital economy: The role of stewardship. Bus. Horiz. 2022, 65, 535–546. [Google Scholar] [CrossRef]
  29. Chen, Y.; Wang, Z.; Ortiz, J. A Sustainable Digital Ecosystem: Digital Servitization Transformation and Digital Infrastructure Support. Sustainability. 2023, 15, 1530. [Google Scholar] [CrossRef]
  30. Liu, R.; Zheng, L.; Chen, Z.; Cheng, M.; Ren, Y. Digitalization through supply chains: Evidence from the customer concentration of Chinese listed companies. Econ. Model. 2024, 134, 106688. [Google Scholar] [CrossRef]
  31. Hamilton, J.D. Daily monetary policy shocks and new home sales. J. Monet. Econ. 2008, 55, 1171–1190. [Google Scholar] [CrossRef]
  32. Zhu, K.; Kraemer, K.L.; Xu, S. The process of innovation assimilation by firms in different countries: A technology diffusion perspective on e-business. Manag. Sci. 2006, 52, 1557–1576. [Google Scholar] [CrossRef]
  33. Montgomery, D.C.; Johnson, L.C. Simple exponential smoothing. In Forecasting and Time Series Analysis, 1st ed.; Montgomery, D.C., Johnson, L.C., Eds.; McGraw-Hill: New York, NY, USA, 1976; p. 67. [Google Scholar]
  34. Bring, J. How to Standardize Regression Coefficients. Am. Stat. 1994, 48, 209–213. [Google Scholar] [CrossRef]
  35. Yang, Y.; Wang, R.; Tan, J. Coupling Coordination and Prediction Research of Tourism Industry Development and Ecological Environment in China. Discret. Dyn. Nat. Soc. 2021, 2021, 6647781. [Google Scholar] [CrossRef]
  36. Li, X.; Lu, Z.; Hou, Y.; Zhao, G.; Zhang, L. The coupling coordination degree between urbanization and air environment in the Beijing(Jing)- Tianjin(Jin)-Hebei(Ji) urban agglomeration. Ecol. Indic. 2022, 137, 108787. [Google Scholar] [CrossRef]
  37. Liu, M.; Lu, M.; Li, Z. Coupling coordination analysis on digital economy-tourism development-ecological environment. J. Clean. Prod. 2024, 470, 143320. [Google Scholar] [CrossRef]
  38. Li, Z.; Lai, Q.; He, J. Does digital technology enhance the global value chain position? Borsa Istanb. Rev. 2024, 24, 856–868. [Google Scholar] [CrossRef]
  39. Fu, X.; Xu, Y. The impact of digital technology on enterprise green innovation: Quality or quantity? Green Financ. 2024, 6, 484–517. [Google Scholar] [CrossRef]
  40. Li, Z.; Guo, F.; Du, Z. Learning from Peers: How Peer Effects Reshape the Digital Value Chain in China. J. Theor. Appl. Electron. Commer. Res. 2025, 20, 41. [Google Scholar] [CrossRef]
  41. Li, Z.; Xu, Y.; Du, Z. Valuing financial data: The case of analyst forecasts. Financ. Res. Lett. 2025, 75, 106847. [Google Scholar] [CrossRef]
Table 1. Overview of variables and measurement.
Table 1. Overview of variables and measurement.
TypeVariableProxyMeasurement
Interpreted variableDigital transformationProportion of digital intangible assets in total intangible assets (Digital)Digital intangible assets/Total intangible assets
Explanatory variablesMarket strengthLerner index (Mar)(Price − Marginal Cost)/Price
Government subsidiesGovernment subsidies on digital transformation (Gov)Digital transformation subsidies/Total assets
Control variablesEnterprise factorsSize (Size)Natural logarithm of the total number of employees
Enterprise age (Age)Fiscal year minus incorporation year plus 1
Financial factorsCurrent ratio (Cur)Current assets/Current liabilities
Debt-to-asset ratio (DE)Total liabilities/Total assets
Profitability (Pro)Net profit/Net assets
Cash flow (CF)Operating cash flow/Total assets
Whether it is a loss-making business (Los)Dummy variable: 1 if net income <0 in the current fiscal year, 0 otherwise
External economic environmentMarket volatility risk (Mvs) Annualized   volatility :   Standard   deviation   of   daily   stock   returns   ×   252
Table 2. Results of baseline regression.
Table 2. Results of baseline regression.
Variables(1) Digital(2) Digital(3) Digital(4) Digital(5) Digital(6) Digital
Gov2.793 ** 2.869 **
(1.163) (1.163)
Mar 0.052 ***0.053 ***
(0.0128)(0.0128)
Gov_smoothed 6.796 *** 6.984 ***
(1.755) (1.755)
Mar_smoothed 0.066 ***0.067 ***
(0.014)(0.014)
Constant0.247 ***0.235 ***0.233 ***0.245 ***0.230 ***0.226 ***
(0.010)(0.011)(0.011)(0.010)(0.011)(0.011)
Control variableYesYesYesYesYesYes
Time fixedYesYesYesYesYesYes
Industry fixedYesYesYesYesYesYes
N28,32428,32428,32428,32428,32428,324
R20.2910.2910.2910.2910.2910.291
Notes: Significance level is denoted by *** and ** for 1% and 5%. Standard errors are in parentheses.
Table 3. Results of the robustness test.
Table 3. Results of the robustness test.
(1) Change Regression Model(2) Substitute Control Variables(3) Change Parameter Estimation Method
VariablesDigitalDigitalDigital
Gov_smoothed12.140 ***6.979 ***6.689 ***
(3.081)(1.755)(0.165)
Mar_smoothed0.213 ***0.067 ***0.067 ***
(0.028)(0.014)(0.001)
Constant0.164 ***0.226 ***0.225 ***
(0.013)(0.011)(0.001)
Control variableYesYesYes
Time fixed YesYes
Industry fixed YesYes
N28,32428,32428,324
R20.02830.2910.985
Notes: Significance level is denoted by *** for 1%. Standard errors are in parentheses.
Table 4. Results of partial regression sum of squares.
Table 4. Results of partial regression sum of squares.
Variables(1) Digital(2) Digital(3) Digital
Gov_smoothed27.13 *** 27.72 ***
(1.967) (1.961)
Mar_smoothed 0.197 ***0.201 ***
(0.014)(0.014)
Constant0.247 ***0.209 ***0.194 ***
(0.011)(0.011)(0.011)
Control variableYesYesYes
Time fixedYesYesYes
Industry fixedYesYesYes
N28,32428,32428,324
R20.03580.03560.0424
Notes: Significance level is denoted by *** for 1%. Standard errors are in parentheses.
Table 5. Identification of substitution effects.
Table 5. Identification of substitution effects.
VariablesDigital
Gov_smoothed (centralized)7.396 ***
(2.247)
Mar_smoothed (centralized)0.0665 ***
(0.0189)
Interaction−56.67 **
(23.56)
Constant0.237 ***
(0.0122)
Control variableYes
Time fixedYes
Industry fixedYes
N28,324
R20.292
Notes: Significance level is denoted by *** and ** for 1% and 5%. Standard errors are in parentheses.
Table 6. Results of substitution effects.
Table 6. Results of substitution effects.
(1) Low(2) Mid(3) High
VariablesDigitalDigitalDigital
Gov_smoothed12.67 ***4.2314.059
(2.795)(3.077)(3.301)
Mar_smoothed0.04350.109 **0.0707 ***
(0.049)(0.044)(0.026)
Constant0.181 ***0.181 ***0.631 ***
(0.019)(0.019)(0.023)
Control variableYesYesYes
Time fixedYesYesYes
Industry fixedYesYesYes
N944094409439
R20.2920.3140.332
Notes: Significance level is denoted by *** and ** for 1% and 5%. Standard errors are in parentheses.
Table 7. Results to validate the difference among groups.
Table 7. Results to validate the difference among groups.
(1) Low vs. Mid(2) Low vs. High(3) Mid vs. High
VariablesDigitalDigitalDigital
Gov_smoothed11.930 ***14.780 ***4.613
(2.975)(3.002)(4.288)
MarDummy−0.008 ***0.013 ***0.011 ***
(0.003)(0.004)(0.004)
Gov_smoothed × MarDummy−8.507 *−12.190 **−1.097
(5.154)(5.124)(5.888)
Constant0.206 ***0.259 ***0.253 ***
(0.014)(0.016)(0.016)
Control variableYesYesYes
Time fixedYesYesYes
Industry fixedYesYesYes
N18,88318,88318,882
R20.2930.2900.300
Notes: Significance level is denoted by ***, **, and * for 1%, 5%, and 10%. Standard errors are in parentheses.
Table 8. Test of government and market synergy on digital transformation.
Table 8. Test of government and market synergy on digital transformation.
VariablesDigital
Coordination1.077 ***
(0.229)
Constant0.245 ***
(0.00996)
Control variableYes
Time fixedYes
Industry fixedYes
N28,324
R20.291
Notes: Significance level is denoted by *** for 1%. Standard errors are in parentheses.
Table 9. Heterogeneity by enterprise ownership.
Table 9. Heterogeneity by enterprise ownership.
(1) State-Owned(2) Non-State-Owned
VariablesDigitalDigital
Gov_smoothed16.37 ***4.383 **
(3.631)(1.949)
Mar_smoothed0.187 ***−0.0829 ***
(0.0214)(0.0197)
Constant0.200 ***0.242 ***
(0.0175)(0.0146)
Control variableYesYes
Time fixedYesYes
Industry fixedYesYes
N12,63915,685
R20.2800.345
Notes: Significance level is denoted by *** and ** for 1% and 5%. Standard errors are in parentheses.
Table 10. Heterogeneity by enterprise life cycle.
Table 10. Heterogeneity by enterprise life cycle.
(1) Growth(2) Maturity(3) Decline
VariablesDigitalDigitalDigital
Gov_smoothed−1.8783.55734.59 ***
(2.518)(3.513)(4.129)
Mar_smoothed0.0394 *0.128 ***0.150 ***
(0.0239)(0.0259)(0.0286)
Constant0.185 ***0.184 ***0.289 ***
(0.0186)(0.0203)(0.0246)
Control variableYesYesYes
Time fixedYesYesYes
Industry fixedYesYesYes
N924688798330
R20.2760.3020.344
Notes: Significance level is denoted by *** and * for 1% and 10%. Standard errors are in parentheses.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Li, T.; Ni, L.; Xu, Y. Enterprise Digital Transformation Drivers: Market or Government? A Case Study from China. J. Theor. Appl. Electron. Commer. Res. 2025, 20, 131. https://doi.org/10.3390/jtaer20020131

AMA Style

Li T, Ni L, Xu Y. Enterprise Digital Transformation Drivers: Market or Government? A Case Study from China. Journal of Theoretical and Applied Electronic Commerce Research. 2025; 20(2):131. https://doi.org/10.3390/jtaer20020131

Chicago/Turabian Style

Li, Tinghui, Linteng Ni, and Yanting Xu. 2025. "Enterprise Digital Transformation Drivers: Market or Government? A Case Study from China" Journal of Theoretical and Applied Electronic Commerce Research 20, no. 2: 131. https://doi.org/10.3390/jtaer20020131

APA Style

Li, T., Ni, L., & Xu, Y. (2025). Enterprise Digital Transformation Drivers: Market or Government? A Case Study from China. Journal of Theoretical and Applied Electronic Commerce Research, 20(2), 131. https://doi.org/10.3390/jtaer20020131

Article Metrics

Back to TopTop