The Impact of Digital Transformation on Firm Performance - An Empirical Study Based on Business Administration Perspective
Data publikacji: 19 mar 2025
Otrzymano: 01 lis 2024
Przyjęty: 03 lut 2025
DOI: https://doi.org/10.2478/amns-2025-0518
Słowa kluczowe
© 2025 Ying Wang et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
In recent years, the development of digital technology continues to promote the digital transformation of enterprises, in this process, all industries in different degrees of traditional skills and digital technology for the integration of the development of artificial intelligence, the Internet of Things, cloud computing and other digital technologies, the rapid development of this has had a profound impact. Enterprises use digital technology as a means to form a closed loop from collection to feedback, so that the data barrier between the enterprise and the various levels of the industry can be further opened up, promoting the improvement of the operational efficiency of the overall business, and then establishing a new digital system [1–4]. It can be found that digitization is an important driving force for the development of the current digital economy, which aims at integrating various information in the external environment, thus enhancing its ability to process data and thus improving enterprise performance [5–7].
Enterprise performance refers to the performance and benefits obtained by an enterprise, as well as its managers, using management methods and techniques to apply the enterprise's resources to the production and operation of products or services and investment in a specific business cycle [8–10]. It can reflect whether the enterprise can effectively utilize the resources it possesses in its business activities for the purpose of creating value [11–12]. In order to develop healthily and sustainably, enterprises need to actively adjust their management decisions according to market demand, put digital change in the company's development strategy, and fully utilize the research results of big data, cloud computing and artificial intelligence to help the company achieve better development.
Enterprise performance is an important measure of success, and business administration factors play a key role in shaping enterprise performance. Organizational structure is the framework within the firm that determines the distribution of authority and responsibility, information flow and decision-making hierarchy, and its impact on firm performance includes efficiency and innovation aspects. Literature [13] points out that enterprises need to introduce a form of organizational structure adapted to the digital era to maintain the effective operation of the enterprise, digital transformation can make the organization better adapted to the needs of the enterprise, and the reduction of organizational structure layers driven by digitalization will also promote employee cooperation to accelerate the efficiency of enterprise operation, which is a major breakthrough in the traditional bureaucratic organizational model. Literature [14] demonstrates through empirical experiments that enterprise digital transformation strategy promotes organizational efficiency, especially in the economic, human resources and internationalization dimensions, which translates into organizational effectiveness and enterprise performance. Literature [15] explored the impact of digital transformation as well as loose organizational forms on firm performance; while digital technology is essential for firms, loose operations and human resources will improve organizational effectiveness to a greater extent, and decision makers can drive business performance by exploiting the synergies between the two. Literature [16] uses a multi-layer framework of organizational activities to analyze the digital transformation process of enterprises and finds that the transformation of enterprises with fuzzy platform ecosystems with business boundaries is more successful, suggesting that decentralized versus centralized thinking in the organizational hierarchy of an enterprise as well as the establishment of organizational boundaries will significantly affect the business strategy and operational results of an enterprise. In summary, the rationality of the organizational structure is crucial to the maximization of enterprise efficiency, and the realization of a flat and loose organizational structure by effectively reducing the number of management levels can not only improve the speed of information flow, but also better motivate employees, so that the enterprise efficiency can be maximized.
In addition to this, organizational structure also has a significant impact on employee innovation. Literature [17] investigated the mediating role of digital transformation strategy and organizational innovation on enterprise performance, the use of digital technology will promote the development of digital transformation strategy, but also actively promote enterprise organizational innovation, which will lead to the improvement of enterprise performance. Literature [18] describes the competitive advantage brought about by the development of innovative employees in the digital transformation of enterprises, and the use of digital technology identifies the vehicle for the development of employees in innovative enterprises, which promotes the rapid diffusion of technology within the organization and continuous improvement in the process of use, which contributes to the formation and enhancement of the competitive advantage of innovative enterprises. Literature [19] emphasizes that digital strategic orientation is a new dimension in shaping the performance of innovative organizations, and that employees improve their own digital knowledge capabilities and innovation capabilities during the process of enterprise digital transformation, and that employees' capabilities in turn drive the overall performance of the organization as well as the strategic orientation of the enterprise to further improve. It can be found that the enterprise digital transformation strategy promotes the formation of an open organizational structure that can encourage cross-departmental cooperation and knowledge sharing among employees, clearing away information silos and barriers to innovation among teams, thus making it easier to innovate.
This paper puts forward the research hypotheses of this paper through the theoretical analysis of the direct impact of digital transformation on enterprise performance and the impact of digital transformation on enterprise performance through the mediating role of enterprise management capabilities. Tobin Q is selected as the explanatory variable to measure enterprise performance. The keywords are retrieved from the shared digital vocabulary network of enterprises through text mining, and the filtered keywords are mapped to the high-dimensional space by using the BERT model, which is combined with the K-menas clustering algorithm and the TF-IDF algorithm to complete the measurement of the explanatory variable, “Degree of Digital Integration (DIG)”. The mediator variable, “management capability”, is measured from the two paths of operation efficiency and management efficiency. A benchmark regression model is constructed to study the specific impact of digital transformation on enterprise performance. Meanwhile, in order to highlight the mediating effect of “management ability”, a mediation effect test model is constructed to investigate its role in the process of digital transformation on enterprise performance. Based on the results of empirical analysis such as benchmark regression, robustness test, heterogeneity test and mediation effect test, the research hypotheses of this paper on the impact of digital transformation on enterprise performance are verified.
Digital transformation, as a unique and innovative approach in today's economic era, has revolutionized all aspects of enterprise production, operation, management, and service methods. Enterprise digital transformation can improve enterprise performance [20], which can be explored and analyzed from two aspects of cost reduction and efficiency improvement.
First, digital transformation is conducive to enterprises to reduce transaction costs and communication costs. The application and development of digital technology in the enterprise promotes the mutual integration of the various departments of the overall value chain of the enterprise, integrates material resources and human resources, reduces the transaction costs of the enterprise, and enhances the value-added capacity of the enterprise's products and services, so as to improve the enterprise performance. In the era of digital economy, data has become the third major factor of production in addition to material and human resources. Enterprises are able to use digital technology to capture all kinds of information related to target customers or potential customers through the Internet, so as to achieve accurate marketing to customers, but also to enhance the sense of user experience, and to a certain extent, reduce the cost of business-to-person communication. At the same time, due to the use of computers, intelligent equipment, etc., so that the information asymmetry between the enterprises can be gradually eliminated, the barriers between the industry is also getting smaller and smaller, the competition and cooperation between enterprises have been optimized to enhance the model. So digital technology also makes the communication cost between enterprises can be reduced.
Secondly, digital transformation is conducive to enterprise efficiency. Some enterprises carry out digital transformation is more inclined to internal enterprises, the use of digital technology to build information systems, is conducive to the optimization of internal business processes, to achieve the purpose of efficiency improvement. Big data, AI and other types of intelligent elements continue to pour into enterprises, bringing opportunities for transformation and change, as well as improving internal management and input-output efficiency for enterprises. Digital transformation can use its own characteristics to efficiently and quickly deploy a variety of production resources, so that they can be utilized with maximum efficiency, and from the overall interests of the enterprise, help the enterprise's sustainable development, improve the competitiveness of the enterprise, and enhance the performance of the enterprise. Therefore, this paper proposes the following hypotheses:
H1: Digital transformation shows a positive correlation with enterprise performance
From the perspective of business administration, in an increasingly competitive market environment, it is crucial for enterprises to stand out if their operation and management capabilities can cope with the market environment. From the perspective of cost reduction and efficiency enhancement, an enterprise's operation and management capability mainly involves the control of the enterprise's management expense ratio and the improvement of operational efficiency, and digital transformation can help enterprises make improvements through the use of digital technology.
First, digital transformation can improve its operating ability by reducing management costs, thus affecting enterprise performance. Based on the theory of organizational change and information systems in management, digital transformation can help enterprises optimize their organizational structure, break the “data silos” between departments, realize the flow of data in the whole process, and with the data analysis tools, it can provide real-time data and analysis for the management, which greatly improves the efficiency of corporate decision-making. Meanwhile, according to the principal-agent theory, enterprises connect the four key links of data and information collection, processing, analysis and application through automation and intelligent solutions, which can minimize the cost of information disclosure, reduce the asymmetry of information, improve the efficiency of interdepartmental communication, and truly realize the standardization, transparency and high efficiency of the inter-departmental management process, which helps to solve the problem of intra-enterprise principal-agent and improve the internal management structure, etc., thus enhancing internal management efficiency and reducing management costs.
Secondly, digital transformation can further promote the improvement of enterprise performance by improving the operational efficiency of enterprises. Based on resource allocation theory and supply chain management theory, digital transformation has a unique advantage in optimizing resource allocation and improving production and research and development efficiency. Product production and research and development is the core competitiveness of enterprises, the application of digital technology through intelligent production equipment, real-time monitoring of big data can be found in a timely manner in the production link of the product loopholes, the enterprise through the optimization of the product production process, improve the efficiency of the product production, so as to enhance the competitiveness of the product. At the same time, the use of digital technology can also provide real-time inventory management and demand forecasting, according to changes in the market environment, timely early warning enterprises to reduce inventory levels and reduce the cost of capital employed. Reducing excess inventory and scrap reduces capital tie-ups, helps assets flow more efficiently, and improves total asset turnover and enterprise operational efficiency. Based on the above analysis, this paper proposes the following hypothesis:
H2: Operational management capability has a mediating role in the impact of digital transformation on enterprise performance.
A-share listed companies from 2011-2023 are selected as the sample. The relevant financial indicators at the enterprise level are from CSMAR database. In order to ensure the validity of the data and the authenticity of the empirical results, the data are processed as follows: (1) Companies with abnormal indicators are excluded. (2) Companies with data years less than 5 years are excluded. (3) Remove companies with missing main variables and ST and *ST enterprises. After manually removing extreme data from some companies, a total of 3,640 samples were obtained.
Explained variable: Tobin's Q (Tobin Q). In this paper, Tobin Q (Tobin Q) is used to describe firm performance [21]. It is calculated as the ratio of the market value of the enterprise to the replacement cost of assets.
Explanatory variable: degree of digital transformation (DIG). This paper adopts the method of text mining to construct a digital vocabulary network shared by enterprises using specific keywords, and generates a digital transformation degree indicator by determining the degree of embedding of enterprises in the digital vocabulary network [22]. The specific steps are as follows:
Digital vocabulary network construction.
In this paper, we searched the official websites of the People's Government of China and the Ministry of Industry and Information Technology, manually screened the policy documents related to the digital economy at the national level and previous literature, and obtained a total of 145 vocabularies related to the digital transformation of enterprises after judgment and retention. Subsequently, based on the semantic and syntactic similarity of the vocabularies, a BERT model is used to map the screened digital key vocabularies into a high-dimensional vector space to form a structured digital vocabulary network. Finally, the vocabulary in the high-dimensional vector space is clustered using the K-means++ algorithm to classify semantically similar words into the same cluster. The optimal number of clusters is determined by contour coefficients in order to accurately divide the vocabulary. The final results are shown in Table 1, where 145 digitized vocabularies are divided into 12 clusters in this paper.
Enterprise text content
In order to accurately measure the progress of enterprise digitization and exclude the influence of anticipatory and negative content on the construction of indicators, this paper extracts the enterprise analysis text through positive affirmation and negative negation, the specific steps are as follows: using Python text analysis method, the sentences containing one or more fields of the 145 keywords in the Management Discussion and Analysis section of the annual report are retained as the initial text objects to be analyzed. Next, natural language processing techniques are used to determine whether the enterprise has really carried out digital transformation. In the first stage, in order to exclude the expected content in the initial text and the citation of the digital policy document, the determination method is that if the initial text contains one or more of the 145 keywords, and the same sentence does not contain characters such as “will”, etc., the positive affirmation has been passed. The second stage is to exclude sentences that contain negative words in the initial text to indicate that the enterprise has not yet carried out digital transformation, and the method of determining is that the sentences contain “does not exist”, “no”, “did not happen”, “did not take”, “did not carry out” and “not carried out”, and these words are followed by digital transformation keywords, or sentences containing both “may” and “digital transformation” and the distance between the two is no more than 5 characters, which is determined to be negative exclusion. Only sentences that pass positive affirmations and negative exclusions serve as the final text constructed by the indicator.
Constructing features
There are 12 clusters in the digital vocabulary network of this paper, the degree of digital transformation is measured by calculating the sum of the TF-IDF values of the keywords within each cluster involved in the document.The TF-IDF algorithm is a statistical method used for textual analysis that consists of the word frequency (TF) and the inverse document frequency (IDF).The TF measures the frequency of occurrence of a word in the text and the mathematical expression is:
Mediating variable: operation and management capability. This paper adopts operation capability as the mediating variable of enterprise digital transformation affecting enterprise performance, and analyzes it from two paths: operation efficiency and management efficiency. The total asset turnover ratio is used to measure operational efficiency (ATO), and the management expense ratio is used to measure enterprise management efficiency (Admin). The management quality and operational efficiency of the enterprise are directly related to the development of the enterprise, and the impact of digitalization on operational management capabilities can be more intuitively reflected by studying the impact of digitalization on the path of enterprise performance.
Control variables: this paper selects six variables as control variables, such as size of operating income (SALE), total assets (ASSEET), gearing ratio (DAR), equity concentration (OC), current ratio (LIQ) and age of enterprise (AGE).
Digital transformation category
| Serial number | Lexical theme | Serial number | Lexical theme |
|---|---|---|---|
| Cluster 1 | Internet technology | Cluster 7 | Mobile payment |
| Cluster 2 | Communication technology | Cluster 8 | Intelligent decision-making and management |
| Cluster 3 | Data analysis and processing | Cluster 9 | Information control system |
| Cluster 4 | Cloud computing | Cluster 10 | Digital network |
| Cluster 5 | Artificial intelligence | Cluster 11 | Block chain technology |
| Cluster 6 | Virtual reality | Cluster 12 | Bio-identification |
For Hypothesis 1, in order to examine whether digital transformation can have a significant impact on firm performance, this paper constructs model (4):
For hypothesis 2, in order to examine whether business management capabilities have an impact on digital transformation and corporate financial performance, this paper constructs model (5), model (6) and model (7):
Table 2 shows the descriptive statistics of the main variables. The mean value of enterprise performance (Tobin Q) is 2.084, and the standard deviation is 1.207, which indicates that there is a difference in performance among the sample enterprises. Digital transformation (DIG) has a mean value of 1.172 and a standard deviation of 1.326, indicating that the level of digital transformation varies among the sample firms. The mean value of Enterprise Operation Efficiency (ATO) and Enterprise Management Efficiency (Admin) is 1.328 and 1.533 respectively, and the standard deviation is 1.694 and 1.574 respectively, indicating that there is a large gap in operation and management capabilities among different enterprises.
Descriptive statistics of major variables
| Count | Mean | SD | Min | P50 | Max | |
|---|---|---|---|---|---|---|
| Tobin Q | 3640 | 2.084 | 1.207 | 0.849 | 1.592 | 7.468 |
| DIG | 3640 | 1.172 | 1.326 | 0.000 | 0.674 | 4.852 |
| ATO | 3640 | 1.328 | 1.697 | 0.605 | 1.103 | 8.172 |
| Admin | 3640 | 1.533 | 1.574 | 1.009 | 1.326 | 4.938 |
| SALE | 3640 | 20.839 | 1.472 | 18.236 | 20.453 | 25.694 |
| ASSEET | 3640 | 22.319 | 1.306 | 18.942 | 22.047 | 26.589 |
| DAR | 3640 | 0.173 | 0.048 | 0.039 | 0.103 | 0.584 |
| OC | 3640 | 35.271 | 14.973 | 8.692 | 32.148 | 73.956 |
| LIQ | 3640 | 0.248 | 0.409 | 0.013 | 0.007 | 1.264 |
| AGE | 3640 | 8.429 | 0.537 | 5.342 | 8.376 | 13.974 |
Table 3 shows the regression results of the impact of the degree of digital transformation on business performance, column (1) is the result without adding control variables, column (2) is the result with the addition of control variables, and *, **, *** in the table indicate that they are significant at the 10%, 5% and 1% levels, respectively, and the following is the same as this. The results show that the SID coefficients without adding control variables are significantly positive at the 5% level, and the SID coefficients with adding control variables are significantly positive at the 1% level, indicating that digital transformation significantly improves the level of business performance, and the hypothesis H1 proposed in this paper is initially verified. In terms of economic significance, for every unit increase in the degree of digital transformation, enterprise performance will rise by 0.293. In addition, the coefficient of enterprise age in the control variable is significantly negative at the 1% level, indicating that with the growth of enterprise age and size, management problems gradually emerge, resulting in a downward trend in the expected growth potential of the enterprise.
Benchmark regression results
| Variable | (1) | (2) |
|---|---|---|
| Tobin Q | Tobin Q | |
| DIG | 0.172** (1.936) | 0.293*** (4.138) |
| SALE | 5.271*** (4.362) | |
| ASSEET | 0.674** (0.628) | |
| DAR | 0.932* (1.407) | |
| OC | 0.278*** (0.924) | |
| LIQ | 0.087* (0.448) | |
| AGE | -0.362*** (-1.376) | |
| Constant | 1.473*** (4.832) | 2.479*** (0.594) |
| Observed value | 3640 | 3640 |
| R2 | 0.047 | 0.398 |
| Enterprise fixed effect | Yes | Yes |
| Year fixed effect | Yes | Yes |
In this paper, ROE is used as a replacement variable for firm performance, and other variables are kept unchanged for re-regression, and the results are shown in column 1 of Table 4. The results show that the DIG coefficients are still significant at the 1% level and the sign of the coefficients has not changed, and the regression results are basically consistent with the benchmark regression results, indicating that the conclusions are robust.
Robustness test
| Variable | (1) | (2) | (3) | (4) | (5) |
|---|---|---|---|---|---|
| ROE | Tobin Q | Tobin Q | SIG | Tobin Q | |
| DIG | 0.039*** (2.68) | 0.273*** (4.12) | 0.548** (2.043) | ||
| DIG-t | 0.091*** (2.479) | ||||
| IV | 0.019*** (4.962) | ||||
| Constant | 0.956** (2.29) | 2.548 (0.573) | 1.582*** (10.095) | 8.309*** (42.876) | -2.038 (-0.742) |
| Observed value | 3640 | 3640 | 3640 | 3640 | 3640 |
| R2 | 0.973 | 0.284 | 0.048 | 0.217 | 0.262 |
| Control variable | Yes | Yes | Yes | Yes | Yes |
| Enterprise fixed effect | Yes | Yes | Yes | Yes | Yes |
| Year fixed effect | Yes | Yes | Yes | Yes | Yes |
Since the research conclusions may be affected by the economic environment as well as the special time period, this paper will present the data of the special time period to verify the stability of the conclusions. The time interval of this paper is 2011-2023, and 2020 is affected by special events with special characteristics, so the sample data of 2020 is excluded, and its regression results are shown in column 2 of Table 4. The SIG coefficients are significantly positive at 1% level, and its regression results are consistent with those of the benchmark regression results, which indicates that the conclusions are robust.
Considering the time-lagged nature of the impact of digital transformation on firm performance, digital transformation is lagged by two periods (DIG-t), and the results are shown in column 3 of Table 4. Digital transformation and firm performance are significantly positive at the 1% level, and the direction and significance of the regression coefficients are consistent with those of the benchmark regression results, indicating that the core findings of this paper are robust.
To further mitigate the endogeneity problem caused by reverse causality and first-floor variables, this paper adopts the instrumental variable method for re-regression. The interaction term (IV) between the number of fixed telephones and the number of Internet users in each prefecture-level city in 2003-2010 with one period lag is selected as the instrumental variable of DIG in 2011-2023. The number of landline telephones and the number of Internet users affect the technological level of the city, which in turn affects digitization, and the effects of landline telephones and the Internet on firm performance become very weak as the frequency of use declines. The results of the two-stage least squares regression for the selected instrumental variables are shown in columns (4) and (5) of Table 4, where the value of the Cragg-Donald Wald F-statistic is 25.748, which is higher than the 10% critical value of the Stock-Yogo weak instrumental variable test of 16.38, and there is no weak instrumental variable situation. Column (4) shows the results of the first-stage regression where the instrumental variable IV is significantly positive, and column (5) shows the results of the second-stage regression where the coefficients of the explanatory variables are significantly positive at the 5% level, indicating that the conclusions remain valid and the results of the benchmark regression are robust by ruling out endogeneity problems.
Using the median of the economic policy uncertainty index as a criterion, the economic policy is divided into high and low uncertainty to explore the asymmetric structural characteristics of the effect of digital transformation on enterprise performance, and the results are shown in Table 5. Compared with the period of low uncertainty of economic policy, the effect of digital transformation on enterprise performance improvement is weaker when economic policy is high uncertainty. The reasons for this may stem from the following three aspects, one, when economic policy is high uncertainty, it will cause difficulties in business operation and decision-making, and reduce the efficiency of enterprise production and operation. Second, the uncertain impact of the direction of economic policy adjustments and the effect of implementation leads to uncertainty in the expected returns of enterprises. At a given level of inputs, it is impossible to reliably predict the level of operating returns, while a high degree of economic policy uncertainty raises the transaction costs of enterprises, which face greater business risks under the given cost constraints. Third, under highly uncertain economic policies, financial institutions tend to propose higher rates of return on capital in order to avoid risks, which intensifies the financing constraints of enterprises and poses a great limitation on their development. Therefore, under the environment of high uncertainty of economic policies, enterprises are unable to make timely and efficient decisions based on economic policies, production and operation face more obstacles, and digital transformation has a limited effect on the improvement of enterprise performance.
Heterogeneity analysis of economic policy uncertainty
| Variable | The uncertainty of economic policy | |
|---|---|---|
| (1) High uncertainty | (2) Low uncertainty | |
| Tobin Q | Tobin Q | |
| DIG | 0.039* (1.942) | 0.271*** (5.684) |
| Constant | 8.132*** (17.423) | 9.628*** (25.371) |
| Observed value | 1739 | 1901 |
| R2 | 0.217 | 0.384 |
| Control variable | Yes | Yes |
| Enterprise fixed effect | Yes | Yes |
| Year fixed effect | Yes | Yes |
According to the industry attributes, the enterprises are categorized into high-tech industries or nonhigh-tech industries, and group regression is carried out, and the results are shown in Table 6. As can be seen from the table, compared with non-high-tech enterprises, the digital transformation of high-tech enterprises has a more obvious effect on enterprise performance improvement. The reasons for the asymmetric results are: first, there are differences in the degree of attention to digital transformation among enterprises with different attributes. High-tech industries naturally pay attention to technological hotspots, while non-high-tech enterprises will not have a higher degree of attention. Second, the technological innovation foundation of enterprises is different. The characteristics of the high-tech industry itself make it better able to adapt to the innovative technological conditions required for change, and can deeply embed digital transformation in its own organizational structure, business processes, decision-making system and production process. Non-high-tech enterprises lack the basic technological conditions to support themselves, and will encounter obstacles and difficulties in the process of digital transformation. According to the regression of the three regions of East, Central and West, the conclusion shows that only the eastern region has a significant impact of digital transformation on enterprise performance. Compared with the central and western regions, the eastern region has implemented the coastal opening policy earlier, carried out market reform earlier, and has a high level of market development, which is an economically developed region. In addition to the commodity market in the western region is more developed, the market for means of production and technology market are relatively backward. Therefore, enterprises in the eastern region have more complete financial and technological conditions to promote the process of digital transformation and thus improve enterprise performance.
Industry and regional heterogeneity analysis
| Variable | Industry properties | Regional difference | |||
|---|---|---|---|---|---|
| (1)High-tech industry | (2)Non-high-tech industry | (1)Eastern region | (2)Central region | (3)Western region | |
| Tobin Q | Tobin Q | Tobin Q | Tobin Q | Tobin Q | |
| DIG | 0.072*** (3.568) | 0.013 (0.971) | 0.079*** (4.726) | -0.003 (-0.006) | 0.037 (1.623) |
| Constant | 8.893*** (21.079) | 7.632*** (13.247) | 8.703*** (26.621) | 8.413*** (10.962) | 6.049*** (4.915) |
| Observed value | 1673 | 1967 | 1458 | 1782 | 400 |
| R2 | 0.326 | 0.283 | 0.317 | 0.309 | 0.281 |
| Control variable | Yes | Yes | Yes | Yes | Yes |
| Enterprise fixed effect | Yes | Yes | Yes | Yes | Yes |
| Year fixed effect | Yes | Yes | Yes | Yes | Yes |
In order to reflect the study of the impact of digital transformation on enterprise performance from the perspective of business administration, this paper adds enterprise management capability as a mediating variable, and measures it through enterprise operation efficiency ATO and enterprise management efficiency Admin to analyze the role played by enterprise management capability in the process of the impact of digital transformation on enterprise performance. Table 7 shows the regression results of the mediation effect test in this paper. Column (1) shows the baseline regression results, where the explanatory variable DIG and the explanatory variable Tobin Q are positive at the 1% want to capture level, indicating that the enhancement of enterprise digital transformation can promote the enhancement of enterprise performance. The explanatory variable Digital Transformation DIG and the mediating variable ATO in column (2) are positive at the 1% significant level, indicating that the improvement of enterprise digital transformation can promote the improvement of enterprise operational efficiency. In column (3), the explanatory variable digital transformation DIG and the mediator variable Admin are positive at the 1% significant level, indicating that enterprises' improved degree of digital transformation can enhance the management efficiency of enterprises. Combining the results obtained in columns (2) and (3), it can be seen that the increase in the degree of digital transformation of enterprises can significantly improve the management capabilities of enterprises. Column (4) is the focus model of the research concern, and the mediating variables ATO and Admin are added to the test model of the mediating effect to observe the coefficients and significance level changes of the explanatory variable DIG. It can be seen that compared to the coefficient of DIG shown in the baseline regression results in column (1), the coefficient of DIG in column (4) increases from 0.293 to 0.374 and is also significantly positive at 1% level, indicating that DIG can enhance corporate performance by improving the efficiency of corporate operations and management, i.e., the operational and managerial capabilities of the enterprise. Overall, the regression results in Table 7 support the mediating effect of operation and management capabilities, i.e., digital transformation can enhance enterprise performance by promoting operation and management capabilities, and Hypothesis 2 of this paper is proved.
Intermediate effect test
| Variable | (1) | (2) | (3) | (4) |
|---|---|---|---|---|
| Tobin Q | ATO | Admin | Tobin Q | |
| DIG | 0.293*** (4.138) | 0.072*** (5.164) | 0.174*** (15.489) | 0.374*** (2.973) |
| ATO | 0.268*** (21.432) | |||
| Admin | 0.279*** (20.228) | |||
| SALE | 5.271*** (4.362) | 4.832*** (3.087) | 4.326*** (3.124) | 8.793*** (5.842) |
| ASSEET | 0.674** (0.628) | 0.725*** (0.612) | 1.836*** (1.438) | 3.257*** (2.871) |
| DAR | 0.932* (1.407) | 0.998*** (1.402) | 1.356*** (2.749) | 3.443*** (3.092) |
| OC | 0.278*** (0.924) | 0.352*** (2.389) | 0.694*** (3.256) | 1.834*** (3.598) |
| LIQ | 0.087* (0.448) | 0.573*** (1.347) | 0.718*** (1.352) | 1.826*** (2.079) |
| AGE | -0.362*** (-1.376) | -2.371*** (-1.974) | -1.794*** (-2.082) | -1.326*** (-2.274) |
| Constant | 2.479*** (0.594) | -0.348*** (-33.271) | 12.874*** (43.263) | 0.482*** (32.935) |
| R2 | 0.398 | 0.352 | 0.397 | 0.413 |
| Observed value | 3640 | 3640 | 3640 | 3640 |
| Enterprise fixed effect | Yes | Yes | Yes | Yes |
| Year fixed effect | Yes | Yes | Yes | Yes |
This paper analyzes the impact of digital transformation on enterprise performance by constructing a regression model and introduces a mediating variable to explore the mediating role of business management capabilities in the industrial and commercial perspective. The coefficient of digital transformation in the baseline regression analysis after adding control variables is 0.293, which is significantly positive at the 1% level. It indicates that digital transformation can significantly improve business performance and that for every unit increase in the degree of digital transformation, there will be a 0.293 increase in business performance. In the robustness analysis, the results of replacing the explanatory variables, adjusting the sample time horizon, lagging the explanatory variables by two periods and endogeneity test all conclude that the coefficient of digital transformation is significantly positive at the 1% level, which is consistent with the results of the benchmark regression. It indicates that the findings of this paper on the impact of digital transformation on firm performance are well robust. The results of heterogeneity analysis indicate that different economic policy uncertainties, different industry attributes and different regional differences all lead to different degrees of impact of digital transformation on enterprise performance. In the mediation effect test, the regression coefficients of digital transformation and the variables ATO and Admin are significantly positive at the 1% level, and the regression coefficient of digital transformation on enterprise performance rises from 0.293 to 0.374 after adding the variables ATO and Admin, verifying the research hypothesis of this paper that the management ability plays a mediating role in the process of the impact of digital transformation on enterprise business performance..